I figured I would share this demo as a way to make folks aware that my gokrazy/rsync module can be used as a library — both its client and its server accept the io.ReadWriter interface type :)
I mostly like gokrazy for stateless systems though I know it supports persistent storage. I assume the rsync module is for file transfer to make a gokrazy system more stateful?
Another use-case I recently had is to copy static website files onto /perm/srv for serving with the Caddy web server. After installing gokrazy/rsync on my router (https://router7.org/), I can now use normal rsync commands to send data from my PC to /perm/srv :)
I know which bank the author worked for because I became aware of the ways these systems were named in the different banks. Orange site reference
The IDE is slow and clunky. Sourcecode stored as code meant I wrote my own grep tool to search a path in the database for source matching things, and even replace with a new pattern. It worked nicely for larger refactors.
Lack of git threw me initially. I grew to appreciate it. The systems running on here are not documented and not appreciated; it does not take much to attain wizard status with the system itself.
I wish more of it were available in the open source world. We’ve solved things differently in real life but your code being objects with a text attribute… unlocks all kinds of possibilities for autogeneration of repetitive code. None of this is actually done but it could be incredible if replicated elsewhere.
Worth noting the data store is not written in python, to my recollection it was written in C++ to meet the performance requirements. It just stored a whole lot of python. Essentially python is the DSL for working within the system.
A MASSIVE SIN in this system is the ease where a python function can call a function only defined in a subclass. Money comes down to simple addition many times, and it is very demoralizing to find this operation and start trying to work out how the data got there… only to discover that the concrete types of the objects involved is a mystery and thus so is the means of backtracking.
Thanks a lot for that! What I find fascinating about these stories is that they open a window on systems that have evolved and thrived outside of the mainstream. I’m particularly interested in the more radical choices made in these systems, and how they felt as a user/developer. In my past life, I ran a dataviz studio and built custom tooling, including languages (replacements for JS, HTML/XML and CSS), which offered a unique and extremely streamlined process. New hires would find the learning curve steep, but overall we got great consistency and productivity. Some later told me that they were missing some of the tooling, which was nice to hear.
I think the most radical idea that can be carried over elsewhere is the idea of infinite working environments, all visible to everybody, all usable by everybody, by any program at any time with almost no additional setup. You do the same setup you would for a regular program and you do a slightly different string and now you’ve got an entirely different base.
I have been chasing that with my CICD journey in the last couple years. I believe that developers should have infinite environments, and if you have a specific blessed stage environment, you’ve done it wrong.
And now that I think about it, that belief may have come from working on Bank Python and seeing how incredible it was to just pass a namespace to somebody else and they could just run the script I had without any additional review or checks. For business reasons, I would have my manager sign off before the operations team would mutate the production view of the database. But that was essentially it.
Actually, the database was the same way, wasn’t it? I could make changes to the database all day long to an overlay of the database. And I would. It was essential for the processing code to be able to process the data and write it somewhere. And then I would just run a diff between the production view and the view I created.
I don’t think most of our databases are set up to have that kind of infinite layering. The closest we get in the free and open source world is probably overlayfs for filesystems.
That’s amazing thanks a lot for the details. I find it fascinating when environments take the code and data as a whole, a bit like Smalltalk images, it is a very different model from the rather decomposed model we have now. In my previous job I actually started to do Postgres-based overlays using foreign data wrappers, so that you could mount the production database read-only and have all your writes local. Changes would then be the residual data on your local instance. In a way, I can imagine integrated Cloud-based environments (like Remplit or CloudFlare) go in the direction, where code and data are all accessible, shareable and “formable”. One last question: do you remember how permissions/security was managed? Banks are well known to be strict in that area, so I’m curious about how there was so much freedom of access/sharing.
Access was tightly gated on writes to the p database. This is what it meant to have code committed to the tree. I believe permissions could be added elsewhere in the system, I never saw it.
By having access to this system at all I was “private side,” and my stock market activity was seriously curtailed beyond the normal curtailing done for all bank employees.
Writes were done by SSO authorization, your username and password were attached to every software deployment as well as to writes.
Dev teams were not allowed to write. Ops teams did that after signoffs. Ops teams were not “allowed” to do dev. So I’d email my manager for a signoff, and then that’d go to the ops team to actually run the script.
There’s an implication that you might avoid certain data-oriented solutions if they would run into awkward parts of your type system. What I interpret then is that it’s data-driven because that’s the only thing you need to worry about up front.
As a self professed generalist this is one of the things I advertise about myself. I’ve given up trying to signal mastery in anything but my ability to figure things out with the help of things like vague awareness, trained over years of doing and exposing myself to as much as possible.
I’m not a very good salesman until I get my foot in the door. My resume is admittedly pretty honest about all of the cool stuff I’ve done, which happen to be not confined to one niche. So for better or worse I haven’t tailored my resume to particular roles nor found it easy to. I agree it’s not easy. If I were in a position to hire someone I’d like a team of generalists advised by one or two specialists until more expertise is needed. I get why companies would want a perfect vertical fit in theory but I think that leads to resume embellishing at best or an opportunity cost in hiring someone that’s self sufficient and capable to do the work now.
I think the Temporal programming model is pretty close to a distributed systems approach that respects existing languages. You write your pure code, and your non-pure code, you run it somewhere, and Temporal makes sure what your pure code describes happens.
It’s a different programming model than people are used to, and doesn’t feel Native to any language it’s in, but it does feel like it belongs in every language it’s in.
Meanwhile the work itself isn’t distributed. It’s highly local. Just… local to anywhere your code is running.
Yes, as the article describes, a two phase commit in Temporal is harder to pull off… but Temporal makes that pain apparent, and doesn’t make it any logically harder than it logically has to be. And if it works, you know it’ll continue to work, even with node and network unreliability.
Maybe we’ve only stalled because fewer of us are on the durable computing bandwagon, and the world will catch up soon?
Having owned a Framework since April of 2022, I cannot recommend them to people who need even basic durability in their devices. Since then, I have done two mainboard replacements, two top cover replacements, a hinge replacement, a battery replacement, several glue jobs after the captive screw hubs sheared from the plastic backing…
It’s just such an absurdly fragile device with incredibly poor thermals. They sacrificed a ton of desirable features to make the laptop repairable, but ultimately have released a set of devices that, when used in real-world settings, end with you repairing the device more often than not. And these repairs are often non-trivial.
I will personally be migrating to another machine. The Framework 12’s focus on durability may be trending in the right direction, but to regain trust, I’d need to see things like drop and wear tests. A laptop that can be repaired, but needs constant upkeep/incredibly delicate handling, is ultimately not an actual consumer device, but a hobbyist device.
Maybe they’ll get better in a few years. Maybe the Framework 12 will be better. Their new focus on AI, the soldered RAM in the desktop offering, and the failure to address the flimsy plastic chassis innards, among other things, mean that they have a long way to go.
It’s definitely a “be part of the community that helps solve our product problems” sort of feeling.
I have an AMD FW13, and was trying to figure out why it loses 50+% of its battery charge overnight when I close the lid, because I don’t use this computer every single day and don’t want to have remember to charge yet another device.
So I check the basics-I’m running their officially supported Linux distro, BIOS is current, etc. And half an hour into reading forum threads about diagnosing sleep power draw, I realize that this is not how I want to spend my time on this planet. I love that they’re trying to build repairable/upgradeable devices, but that goal doesn’t matter so much if people end up ditching your products for another option because they’re just tired of trying to fix it.
I’ll chime in with the opposite experience - I’ve owned an AMD Framework 13 since it came out, and had no durability issues with it whatsoever, and it’s been one of my 2 favorite computers I’ve ever owned. I’ve done one main board replacement that saved my butt after a bottle of gin fell over on top of it in transport.
Development and light gaming (on Linux, I very much appreciate their Linux support) have been great, and the reparability both gives me peace of mind, an upgrade path, and has already saved me quite a bit of money.
I’ve owned a framework since Batch 1. Durability has not been a problem for me. My original screen has a small chip in it from when I put it in a bag with something that had sharp edges and pressured the screen for a whole flight. Slowly growing. Otherwise, it’s been solid.
Same. I have a batch 1. There are quirks, which I expected and knew I am supporting a startup with little experience. I since have upgraded and put my old board into a cooler master case. This is so amazing, and what I cared about. I am still super happy with having bought the Framework and particular for tinkerers and people who will have a use for their old mainboards it’s amazing.
I get harbouring resentment for a company you felt sold then a bad product. But at the same time, you bought a laptop from a very inexperienced company which was brand new at making laptops, a pretty difficult product category to get right when you’re not just re-branding someone else’s white-label hardware.
3 years have passed since then, if I were in the market for a category which Framework competes in these days I would be inclined to look at more recent reviews and customer testimonials. I don’t think flaws in that 3 year old hardware is that relevant anymore. Not because 3 years is a particularly long time in the computer hardware business, but because it’s a really long time relative to the short life of this particular company.
I would agree that 3 years is enough time for a company to use their production lessons to improve their product. But nothing has changed in the Framework 13.
I don’t resent Framework. I think that’s putting words in my mouth. I just cannot, in good faith, recommend their products to people who need even a semi-durable machine. That’s just fact.
a very inexperienced company which was brand new at making laptops
Founded by people who had experience designing laptops already, and manufactured by a company that manufactures many laptops. Poor explanations for the problems, IMO.
I’ve had a 12th gen Intel since Sept 2022 (running NixOS btw) and I have not had any issues, I will admit it sits in one place 99% of the time. I might order the replacement hinge since mine is a bit floppy but not too big a deal.
As for the event, I was hoping for a minipc using the 395 and I got my wish. Bit pricey and not small enough for where I want to put it and I have no plans for AI work so it’s probably not the right machine for me.
I was originally interested in the HP machine coming with the same CPU (which should be small enough to fit) but I’ve been pricing an AMD 9950 and it comes out cheaper. I was also disappointed there wasn’t a sku with 385 Max w/64GB of RAM , which I might have have ordered to keep the cost down.
For reference a new machine is intended to replace a 10 year old Devils Canyon system.
I’ve also had my Framework 13 since beginning of 2022. I’ve had to do a hinge replacement, input cover replacement, and mainboard replacement. But I sort of expected that since it’s a young company and hardware is hard. And through all of it support was very responsive and helpful.
I would expect that nowadays the laptops are probably more solidly built than those early batches!
Support was definitely helpful. I just don’t have time or money to replace parts on my machine anymore.
From what I understand, the laptops aren’t any stronger. Even the Framework 16 just got some aftermarket/post-launch foam pads to put below the keyboard to alleviate the strain on the keyboard. The entire keyboard deck would flex.
The fact that these products have these flaws makes me wonder how Framework organizes its engineering priorities.
When compared to other similar laptops from brands like HP or Lenovo, how does the deck flex compare? I definitely feel sympathetic to not being better or on par with Apple - given the heaps of money Apple has for economies of scale + lots of mechanical engineers, but it would be a bit rough if mid-tier laptops in that category were far superior.
The deck flex is on par with or worse than an HP EliteBook circa 2019. The problem is that it’s incredibly easy to bend the entire frame of the machine, to the point where it interferes with the touchpad’s ability to click.
It’s really bad, bordering on unexcusable. The fact that there’s no concrete reinforcment says that they sacrificed build quality for repairability, which is equivalent to making a leaky boat with a very fast bilge pump.
I’m not sure what you’re doing to your laptop; how are you bending the entire frame of the machine?
It’s a new company that is largely doing right by open source, and especially open hardware. The quality isn’t incredible but it is worth its value, and I find these claims you’re making dubious.
It’s a fairly common flex point for the chassis, and a common support problem. The base of the mousepad, towards the front of the laptop where there’s a depression in the case, is where the majority of the flex is.
My laptop has seen nothing but daily, regular use. You can find the claims dubious, but others are having them too.
I’ll chime in too: I’ve had the Framework 13 AMD since it came out (mid 2023) and it has been great.
I upgraded the display after the new 2.8K panel came out, it took 2 minutes. Couple months later it developed some dead pixels, so they sent me a replacement. In the process of swapping it out, I accidentally tore the display cable. It took me a while to notice/debug it, but in the end it was just a $15 cable replacement that I’m fairly sure would have otherwise resulted in a full mainboard replacement for any other laptop. (When I had Macbooks, I lost count how many times Apple replaced the mainboard for the smallest thing.)
I haven’t been too precious with it, I toss it around like I did my Thinkpad before this. There’s some scuffs but it has been fine, perhaps the newer models are more sturdy? It’s comforting to know that if anything breaks, I’ll be able to fix it.
I also run NixOS on it, it does everything I need it to do, the battery life is great (8-10 hours of moderate use) and I’ll happily swap out the battery in a few more years once it starts losing capacity.
I spend so much of my life at the computer that feeling a sense of ownership over the components makes a lot of sense to me. I don’t want to feel like I’m living in a hotel.
It is, in fact, how I want to spend my time on this planet.
To add to the chorus, I bought a 12th gen intel framework 13 on release and it’s been flawless so far. Nixos worked out of the box. I love the 3:2 screen. I can totally believe that a small/young manufacturing company has quality control issues and some people are getting lemons, but the design itself seems solid to me.
On my old dell laptop I snapped all the usb ports on one side (by lifting up the other side while keyboard/mouse were still connected). Since they’re connected directly to the motherboard they weren’t repairable without buying a new cpu. If I did the same on the framework it would only break the $12 expansion cards and I wouldn’t even have to turn it off to replace them.
Later I dropped that same dell about 20cm on to a couch with the screen open. The impact swung the screen open all the way and snapped the hinges. They wanted me to send it back for repairs but I couldn’t handle the downtime, so for a year I just had the hinges duck-taped together. I’ve dropped my framework the same way, but because the screen opens the full 180 degrees it doesn’t leverage the hinges at all. And if it did break I’d be able to ship the part and replace it myself.
Not that I support the desktop offering as anything but waste, but the soldered RAM is apparently all about throughput:
We spent months working with AMD to explore ways around this but ultimately determined that it wasn’t technically feasible to land modular memory at high throughput with the 256-bit memory bus. (source)
I wish I could name something in good faith that was comparable to a hyper-repairable x86-64 laptop. Lenovo is pivoting towards repairability with the T14 Gen 5, but I can’t recommend that either yet.
Star Labs, System76, some old Thinkpad models.. there are “competitive” things, but few things that pitch the things Framework does.
While I agree on some of that, I must stress that I’ve had hardware that was fine until just one thing suddenly broke and everything was unusable. I’ll try an analogy: with repairability, if all your components are 99% reliable and working, the whole machine is at 99% but without it, even if all of them are at 99.9% instead, when you have 10 components, you’re not in a better situation overall.
And I say that while I need to finish going through support for a mainboard replacement due to fried USB ports on a first-gen machine (although not an initial batch). BTW, funnily I’m wondering if there’s an interaction with my yubikey. I also wish the chassis was a bit sturdier but that’s more of a wish.
As for thermals, while I think they could probably be better, the 11th gen Intel CPU that you have (just like I do) isn’t great at all: 13th gen ones are much better AFAIK.
I’ve experienced a full main board failure which led to me upgrading to a 12th gen on my own dime.
The thermal problems are still there, and their fans have some surprising QA problems that are exacerbated by thermal issues.
I wish I could excuse the fact that my machine feels like it’s going to explode even with power management. The fans grind after three replacements now, and I lack the energy and motivation to do a fourth.
I think 12th gen is pretty similar to 11th gen. I contemplated the upgrade for similar reasons but held off because I didn’t need to know and the gains seemed low. IIRC it’s really with 13th gen that Intel improved the CPUs. But I agree the thermals/power seems sub-par; I feel like it could definitely be better.
BTW, I just “remembered” that I use mine mostly on my desk and it’s not directly sitting on it which greatly improves its cooling (I can’t give hard numbers but I see the temps under load are better and max CPU frequency can be maintained).
Sorry to hear about the trouble with your Framework 13. To offer another data point: I have a 12th gen Framework 13 and haven’t needed to repair a thing, I’m still super happy with it. The frame-bending I’ve also not seen, it’s a super sturdy device for me.
It is, of course, complicated. This is a simplified view. I am leaving some very important assumptions unanswered (such as: how does iMessage do key exchange). Disclaimer: I am an idiot, probably compromised by the government, and you should independently verify the claims below, most of which are probably completely wrong.
most iCloud services, such as iMessage, are end to end encrypted.
for example, if you use “iMessage in iCloud” (which most people probably do) or iPhoto cloud storage, your data is stored in iCloud - as encrypted blobs that are encrypted on device. They also probably do some sort of server-side encryption, but that’s not relevant to this model.
during normal use, their systems cannot decrypt your data (they store device-encrypted blobs). (this is why that weird proposal about having your phone scan your photos for csam [0] was made - unlike google, their architecture is hostile to doing that in the cloud)
key updates are device-to-device, communicated via an imessage subset (also e2ee)
Apple maintains a Key Escrow, which they use to recover your account if you lose it. This also means they can, through specific channels, also turn over your iCloud backups (such as a subpoena)
If you use Advanced Data Protection [1], then Apple does not maintain key escrow [2]. They give you your keys and they promise you that they aren’t keeping a copy of those keys
Apple’s servers are not trusted in their model, which is why “access icloud on web” is an optional checkbox for ADP. Your phone communicates its icloud keys to apple’s servers for a one-off and you’re required to trust that they dont scrape your data and that they discard your keys.
There is a lot of trust, and a lot of external review around this. In some markets, such as China, you cannot disable Key Escrow, but iMessage is (supposedly) E2EE - a rare exception.
So like… make your own conclusions about this, but I’m relatively confident that Apple currently cannot access your data with ADP on. Of course, I can be wrong and even more of course, things may change. They could, for example, silently disable ADP with an iOS update - such a move would hopefully be catastrophic to their public image. Fun fact: this is (possibly) why Apple requires a PIN to update iOS on demand.
TLDR As it stands and as it’s publicly communicated (and verified by researchers): Apple can’t access your data on your phone but they can access anything stored in iCloud unless you use ADP in which case they can’t, unless you use iCloud on web. Your mileage may vary. Promotion not valid in China.
[2] Apple claims that Apple does not maintain key escrow. I’m not saying to the contrary, I’m just highlighting one of many very important assumptions people might make
Surely, there should be other so-called researchers to call the so-called researchers on their bluff for only so-called checking on so-called Apple’s so-called claims.
I don’t know that it’s right to ascribe to conspiracy what simply could be people doing their job well.
The important thing is, if Apple were lying about when or how much they retain keys, how would any so-called researcher know?
Maybe such evidence could become part of the public record in a court case, if it weren’t redacted for national security or whatever, or if the government sloppily didn’t keep it out of the record with their famous ‘parallel construction’ prosecutorial methods. Or maybe an Apple employee could leak, but they sure seem incentivized not to. I don’t know how much to infer from Apple’s selectively vague statements. I get that they’ve marketed themselves as privacy-forward and there’s reputational risk at play. But I also see they just paid $95M to make a specific claim that they might be up to some funny business just go away with no discovery or legal conclusions.
Bottom line for me is, unless I fully control the encryption process and key storage, E2EE isn’t all it’s cracked up to be, and I don’t care how slick the marketing is.
3rd parties continually review although that is at the 3rd party’s whim and not perfect.
And yes, there’s a lot of trust, but there’s also a lot of internal and external researchers.
I, personally, trust Apple is doing what they’re saying. I am also an idiot.
If I had (active) government level entities attacking me, or was worried that there were, I would probably completely change how I communicate and interact with communication technologies.
I’m being a little pithy here btw :p but trust and trusting is an important component of this system, and it’s up to you to inform yourself and individually assess your posture.
Ideally? Maybe this would be regulated by strict pro-privacy laws that deeply incentivized companies like Apple to be pro-privacy.
Idk
Like I said in my root comment, I’m probably a govt shill who shouldn’t be trusted and you should verify for yourself
Not just the idevice (though os control is enough to get all messages in clear text of course) but with iMessage also control of the app itself.
No matter how it is encrypted on their server to begin with, the app or os could easily eg upload the entire db of unencrypted messages to apple at any time.
At one point I said that a good day is a day where you know more at the end than you did at the beginning.
Then I started keeping a file, and I’ve been keeping this file for two years now, where every day has a list under it of the things I have learned that day. Some days don’t get an entry at all, in which case I don’t record the date. But the list items are basically as long as my terminal is wide, or shorter. Tweet-sized.
I optimized my system for handling grep so I can grep through my knowledge base and discover when I tried to do something before what it was I did. It’s my answer for “this thing I was confident about is on the tip of my tongue, but really, I don’t remember what it was.”
It’s even less structured than the TIL system. And it works great for me.
I like the idea of challenging norms, but its very smart to assess how critical something is before working on it.
Yes, it consumes time, but engineering resources are typically limited (even more so after hours ;)).
Triage that can be done by less qualified automations or even people, definitely aid in the mobilisation of an appropriate effort.
Your “SEV” levels just need simple definitions:
Blackout vs Brownout; User-facing/External vs Company-facing/Internal. Your issue severities should be a combination of those. Though this assumes your internal systems can survive a night or 5 of being unavailable- that database backup thats failing for 4 days could suddenly become incredibly important.
What about severities that are a lookup of t shirt levels of bad.
First digit for how broad the care group is. IT, engineering, company, some customers, most customers, all customers.
Second digit, data integrity risk. None, low, high, already lost some. Maybe this should actually be risk of altered performance after this incident.
Third digit, how long can it stay stable before the first digits get worse. A month, day, hour, It can’t get worse.
This would be a lot for an engineering team to manage, but not for an incident manager who’s primarily communicating to others. And also lets others see… We have a SEV-1, but will I care about it tomorrow.
I think that’s a pretty solid approach actually, if your company is Disney sized or something I can see a policy like this working extremely well.
Though you’d need punitive measures in place to prevent the stealing of resources, as soon as subjectivity enters something like this then managers game it to get more attention than they should to pad their numbers.
A smalltalk where every single message is defined as a file in the file system. To be one of the easiest glue languages.
No subclasses
Classes are defined as folders that have messages within them.
Message signatures are the file name.
If you want to share implementations you symlink from one message to the other message. This works perfectly well so long as the parameters you require being passed are at least the same in number between both messages.
If a message is actually a dynamic object, then that dynamic object is used for the invocation instead, allowing full FFI transparently.
If a message is actually an executable, then the program is run. The combination of this feature and the above feature allow for seamless integration between this language and any other.
When you create a class, you can of course decide to symlink most of the implementations present in a base class.
If you have a dependency, such as on another library, that dependency is probably bringing you a class. Well, that class can just be copied into your source tree. Tooling can then rise up around making sure that dependency is in date while innately you own your own dependency and can edit it like any other code.
When the daemon starts, the initial message to pass is defined by the user on the command line. It could have some same default such as calling a Main main
When I first dreamed of this, I figured I would write it in C for the ability to load objects. These days I would write it in Zig because the ability to compile and load and work with the compiler at runtime is incredible.
This would change the nature of the image in smalltalk to be fully compatible with Git and modern deployment mechanisms of “tar it and copy.”
I hadn’t considered it from that perspective; I guess so!
The way I’d internalised it was as mapping HTTP’s verbs and semantics to Smalltalk method dispatch. In section four they show how the principle can be extended to filesystems.
These things end up increasing reuse, since you can “just” trigger an event or a PUT/GET or open a file or dispatch a message and your obligation has ended. If that invokes a whole world of IT compliance or business logic mazes or triggers a supercomputer batch job… you don’t care. Suddenly previously built components can, with minimal knowledge, be glued together. Assemble a table of ways to get the result you need, and put a little “cost” next to each one, and even business people can make some decisions about what gets glued to where to achieve your goals.
We have a lot of languages to do work. We have far fewer good options for glue languages. TCL tried a bit. Python tried a lot. Lua tried a bunch. None of them managed to really explode into “Glue all the things together” territory.
Hence my dream of doing this as local objects, functions, commands, and interpreted piles. Escape from “GusteauTalk” (as my .git repo is named for this project) into a Real Language as soon as you like.
I can’t do all my BigInt processing as HTTP calls, the overhead is too high. But local process execution, or Dylib invocations? I think we can do that.
If you want to share implementations you symlink from one message to the other message. This works perfectly well so long as the parameters you require being passed are at least the same in number between both messages.
Holy hell, this solves the diamond inheritance problem so effectively. If I recall (it’s been a long time), Smalltalk typically wants you to use mixins instead of subclasses. This approach is such a simple and effective way to do highly granular mixins. Very nice!
If I recall (it’s been a long time), Smalltalk typically wants you to use mixins instead of subclasses.
Traditional Smalltalk doesn’t have mix-ins and does absolutely encourage subclassing; I think you’re thinking of Ruby. Many modern Smalltalks do have mix-ins in one form or another (e.g., Pharo has Traits), but that’s a fairly new additon, and IMHO doesn’t honestly mesh terribly well with the rest of the system
Why message per file? What about smalltalk makes it hostile to writing code for it like a regular programming language, where you put all the code for a class in a text file, and you do inheritance via syntax in a text file rather than symlinking a bunch of stuff?
Smalltalk has no syntax nor does it need syntax for defining the message interface. The per-file approach allows you to save the message interface as the filename, keeping it out of the syntax of the message itself. It also allows sharing implementations with different messages if suitable.
Smalltalk being image based makes it hostile to modern deployment mechanisms. Arguably these are not necessary with smalltalk, where you ship the image instead of shipping a separate built artifact, but it’s difficult to impossible to “dip your toe in” on doing so with such a radically different deployment story.
Smalltalk is designed generally around the Image. You write your code in the image and you ship the image. Anything to convert to git or whatever is a translation layer.
Is the image file (as a concept, not the ST implementation) really a barrier in this day and age of Docker? What if they were layered something like Docker, where you’d have the language runtime, on top of that you’d have two “branches” of images, one containing your running / frozen project, and another containing the common development tools alongside your project-specific tooling and any source code you consider important not exist in the distributable artefact? The two “VMs” would communicate using code defined at the lower language-runtime level that is common to both.
Is the image file (as a concept, not the ST implementation) really a barrier in this day and age of Docker?
No. The image is great for deploying server apps and that!s the one place Smalltalk is still used. The problem with the image abstraction is that it splits the world into Smalltalk and alien parts. If you want to use a library that is not Smalltalk, either it’s stateless or it breaks the image abstraction. This is why Étoilé moved away from image-based persistence to versioned object graphs for model objects, so that you had a clean language-level abstraction for separating persistent and ephemeral state and could either explicitly serialise foreign state or treat it as ephemeral.
So long as it can be packaged in an OCI image, I think it’d not be a blocker to deployment. Things like devpods show that OCI images make a compelling base.
Smalltalks generally don’t know what a filesystem even is, though.
I agree with the article, but I find the term “near-miss” to be grammatically confusing. The linked NPR article even switches to “near-collision” to be accurate.
There is nothing confusing about the phrase. Ask yourself a question like “Is the near future the future?” Now ask yourself “Is a near miss a miss?” You can’t answer “yes” to the first and “no” to the second without being inconsistent.
What I think trips people is that “nearly” is a word that negates what comes after, and for some reason in this single specific phrase they try to parse “near” as actually being “nearly”. But that’s not the word. You can even hear Carlin do it with his “they nearly missed”. But if you actually just look at the actual words, there’s nothing confusing about them.
I’ve used it. Most apps work great, as the environment really isn’t all that different from what they’re expecting, especially if you use a gapps image. High-profile games are less happy though. You can’t play Genshin Impact :)
It works great until it doesn’t. In particular, I run it in the “experimental” windowed mode, which is indeed experimental and has some really weird problems, some of which I still don’t understand, some of which I do as just “no one thought of this.” For example: If you move a window to the bottom of the screen, you can’t click on anything on the pixels near the bottom… Because the Android 11 navbar is down there and blocks that area from being interacted with.
I’d report the bugs, but the project seems very underfunded/underdeveloped for something that distros are shipping to end-users. As far as I know it’s still broken out of the box on Fedora.
I’m guessing most people are just running it in the normal fullscreen mode and therefore are having fewer issues.
I use it to listen to Apple Music on Linux instead of using their mediocre web UI. I have to use one version older of the app to get it to detect the internet correctly but it works pretty well overall.
Waydroid requires the “binder” kernel modules. Apparently that’s shipped with the upstream kernel now, but I used to have to use the linux-zen kernel to have it work in the past
Maybe names, logos and websites for severe bugs were actually the better solution ;)
Jerk. I read that right as I was swallowing some coffee, and I burst out laughing. Coffee in the nose is not a nice experience, and I blame you.
(And I think the overall problem is our urge to turn what should be a qualitative analysis into an easier-to-measure set of tickboxes. The tickboxes work to assess bulk trends, but not to assess individual situations, and we care about not having individual sites exploited.)
Unfortunately, I have to give this a CVE a 10 because you could be exploited for a free coffee with a simple social engineering attack, likely including the word “please”
Huh. I’ve made filesystems with a billion empty files, to compare generators and filesystems, and xfs performed poorly, with its default settings. ext4, ext2, and btrfs finished within a day or two; xfs was on track to take weeks.
not sure if the data-point is interesting here, but on my 13T external that’s formatted with modern NTFS, 1_296_232_240 ) inodes are allocated (not used). I could see it working and not as outlandish test it seems to be at first glance.
I checked my notes: it definitely bogged down within the first four million files, which doesn’t feel massive, but maybe that’s relative. I think it bogged down much sooner than that. After hitting ctrl-c, it took 20+ min to get a responsive shell. Unmounting took a few minutes.
Sorry if it came off as a glib “your favorite filesystem sucks”.
Nah, it felt more just like “it’s unreasonable that this thing fails under unreasonable circumstances”. I hope I wasn’t too glib in response; thanks for checking your notes. 4 million files is a much more reasonable circumstance, and bogging down earlier than that is definitely not a good thing. It does surprise me ‘cause XFS was supposedly designed for big chonky systems and in my experience has been quite bulletproof. If only I had the energy to reproduce it and do a a deep dive to find out what’s going on…
You can install Linux on a mobile phone, Steam Deck is Linux and almost all games nowadays just work on Linux.
We have a practical modular and repairable laptop like Framework.
Building your own PC has never been simpler and easier - it’s almost like legos at this point.
PCs are most silent, energy efficient, nicest form factors. You can get PC on a usb stick, or a tiny box.
We have reproducible Linux distros like NixOS and Guix.
We even have an Open Source instruction set: RISCV.
Everything is awesome if you want to do personal computing. It’s just turns out most people don’t want a PC, and they want a stripped down, dumbed down, appliance instead.
It looks most haven’t read the article up until the point where it addresses exactly what you’re talking about. And it says:
In the short term, we can do things like support open projects like Linux, support non-predatory and open source software, and run apps and store data locally as much as possible. But some bigger structural changes are necessary if we really want to launch the era of Personal Computer 2.0.
See, there’s a problem with all your examples (Linux, Framework…) except perhaps the Steam Deck: none of those are mainstream yet. And the effort to make them so is significant, possibly herculean. But most of all, political: companies do bad stuff only because we let them. They’re allowed to use DRM, and have been given the power to sue us if we break them — at lest in the US. They’re allowed to collect and sell our data. They’re allowed to “sell” us stuff, even though it’s not actually a sale, and when it’s inconvenient to keep us locked in they’re allowed to lock us out. We should question those allowances. That is, push for better laws.
Personally, I would love to see the 30 million lines problem solved, and have someone deliver a simple computer that’s good enough for most uses — including gaming. Probably has to be something like RISC-V, considering that one is not allowed to just make and sell a CPU that’s compatible with the mainstream ones (ARM, x86…) — which is utterly ridiculous by the way, instruction sets are interfaces and as such anyone should be allowed to copy it, no authorisation nor fee required.
But I’m not sure we could ever have that without first cleanly separating hardware companies from software companies. The rule is simple: companies that distribute hardware should be forbidden to distribute software — not even Open Source. They’re only allowed to provide the manual required to operate the hardware. It can be done in a kinda extra-territorial way too: if a company distributes both, we just don’t buy their hardware. (And there are ways to determine if 2 companies really are separate.) Do that, and suddenly hardware companies will be motivated to make simple hardware with standard interface, and software companies (and open source alternatives) will be able to actually compete, instead of having to do everything on top of the 3 top kernels out there (NT, Linux, Darwin). Though it might be too radical to seriously consider. I like the idea, but I’m pretty sure it would have a host of adverse effects.
People don’t want personal computers, just like they don’t want to do their own plumbing, do their own banking, etc. Before personal computers they just didn’t use computers at all, now they are just using appliances.
Masses as perfectly happy buying a locked down iPhone, Android, Macbook, Spotify subscription etc.The system works for them perfectly fine, also better than ever. They don’t want freedom and empowerment (but also responsibility) that comes with actually being in control.
But. Personal computing is not dead, just because masses don’t want it. There’s probably more total personal computer users now than there was 20 years ago. And what they can do is better than before too. It’s just that the total number of “computing as a service/appliance” market is now almost everybody else.
I call out just-world fallacy. Many things are decided at scale that people do not want. To give you 2 examples of blatant examples that happened in my country (France):
2005, there was a referendum asking if we wanted the new EU constitution. About 60% said no if I recall correctly. A few months later they just voted the same thing in congress, and it passed.
2022 or so, the government was dead set in enacting a retirement reform that would push the retirement age 2 years later (from 62 to 64). Every survey said over 75% of the population was against it, including ~90% of workers. They still passed it.
Also I recall a US study that say that whether a reform passes or not is mostly determined by what the rich want, not by what the people want: if the rich want something and the people don’t, it often passes, anyway, and if the people want something that the rich don’t, it never passes.
It would be more accurate to say that if rich people wanted it they would have legislated it. And I have a feeling that true personal computing actually goes contrary to the interests of the tech giants. Just do not forget that what tech giants want, may be very different from what people want — and that’s before we touch on whether “want” is conditioned on being informed or not.
People don’t want personal computers, just like they don’t want to do their own plumbing, do their own banking, etc.
That I can get behind. I’m sad about it, but yeah. People don’t (want to?) understand what computers actually are, and I can believe that most are happy with appliances.
Masses as perfectly happy buying a locked down iPhone, Android, Macbook, Spotify subscription etc.The system works for them perfectly fine, also better than ever. They don’t want freedom and empowerment (but also responsibility) that comes with actually being in control.
I’d agree that many people are like that. But are most? I’m not sure to be honest. If we just take the locked down part for instance, are people actually happy they can’t install third party apps on their iPhone without oversight from Apple? Are people actually happy they don’t even have the option?
But. Personal computing is not dead, just because masses don’t want it. There’s probably more total personal computer users now than there was 20 years ago. And what they can do is better than before too. It’s just that the total number of “computing as a service/appliance” market is now almost everybody else.
There’s a network effect though. I gotta use the messenger protocols (and therefore, app) everyone else is using, else I’m being locked out of my friends and family. If I want to pay for stuff online I need to install the 2FA banking app on my phone, and depending on the bank that phone better not be rooted. Overall, there’s a social pressure not to do anything that causes friction with the way other people work or communicate.
For instance, people tell me that recruiters are all on LinkedIn now. But I may have to set up an account and do some of their bullshit if I actually want to be seen by recruiters and raise my salary. Knowing that doing so will ever so slightly contribute to the strength of this network, and ever so slightly increase the incentive for everyone else to stay or follow suite. Same deal with Facebook, I have known some groups who only communicated through it. The result was I wasn’t informed, and soon my involvement dwindled. I don’t want this crap, but as a result I’m locked out of some things.
To be honest, I fear the day when personal computing starts to be banned altogether. I know it sounds tinfoil hat, but that conspiracy already started with the DMCA, where besides a few exceptions people are forbidden to even talk about flaws in their own computers that they bought. What if a sufficiently right-wing government with crazy-rich control freaks in it decide that people need to be “protected”, and all new computers sold must run Trusted Boot, no opt out, and only a few select vendors are allowed to sign operating systems for it? Sure they’d have special provisions for programmers, but in this world people who want actual personal computers are likely to find themselves outside of the law. It’s no question some tech giants actually want this world. Apple led the way with their locked down iPhone, and got away with it because unlike Microsoft back in the day they never dominated the market so completely, but ever since everyone else want a slice of the pie, with Microsoft itself attempting the locked down app store model on desktop computers — fortunately so far they have failed.
Granted, that dystopia is far from inevitable. But it’s probably more likely than most are willing to admit, especially taking into account the rise of fascism.
To achieve what you’re saying, you would need to have any hardware company pair with a software company to have this software available for their hardware when they launch and vice versa. Consider that IBM wrote software simply so that they could sell their computers. If you sell a very nice brick, nobody is going to buy it until they can do something with it. Unless, of course, you go very cheap.
Which reminds me, a company has tried this. They have a hardware team, and they make hardware, and they expect the community, or some other company, to make the software. They are called pine.
f you sell a very nice brick, nobody is going to buy it until they can do something with it.
In a world where companies are allowed to ship software with their hardware, a pure hardware company is indeed at a fatal disadvantage.
But what if no one can do both? One possible outcome is that no one tries radically different hardware: everyone would stick to existing designs and just make them faster, or add instructions here and there at a pace they can expect the software folks to catch up quickly. My hope is that we would see radically different designs, but most of all they’d be radically simpler (at the ISA level, the internals may optimise like crazy), so the software folks can pick it up quickly enough for the new thing to be competitive at all.
I agree. I would like a RISC-V desktop and a modular open-hardward ARM laptop, and they should run the particular set of packages that make up my desktop, assembled and tuned specifically for me.
That’s all doable, I just can’t afford it. But it’s still cheaper than the Apple gear people talked about recently on here.
A laptop for just €950 is bound to be crappy, have some issues, and not last very long. Or so you’d think.
Sorry what? Is it just me or is this kind of out of touch? (Or am I just too frugal?) I get its an Apple product but this is still expensive in my books. This is about how much I’d expect to pay for a new decently good laptop. Converted to AUD this about how much I paid for my brand new at the time, Lenovo X1 Carbon (gen5 - it’s on its way out now).
I’d not spend anywhere near this for a laptop I’d expect to be “crappy”.
Agree. I have never, and would never, spend this much on a laptop unless it were something incredibly special. Thinking of this as “cheap” feels like the author is on another planet
But as the author stated it’s for couch/personal use.
Who needs a “high-end” machine for that?
My personal laptop is a T460p from 2016. It can do everything I’ve ever wanted to do (except 3d games). I don’t see any real benefit that my work M3 MBP has. Nothing. Maybe compiling stuff faster, I’ve not compared anything.
I have a gaming laptop here that cost 1500 and played every game in HQ that was tried so far.
I like using a laptop that has battery life I don’t have to worry about. That, and high end machines tend to last longer - your T460p was absolutely high end when it was new. I doubt i.e. a Chromebook of the time would have lasted you as long.
It was about price - you can have something a little older for a good price. Also battery can be a point for some people, for me it’s not. I’ve lived through the ages for “lasts for 2h” and this one is not that that.
Also Chromebooks were to be had for like 300 - this was about if 900 is about a laptop that might be crappy.
But as the author stated it’s for couch/personal use.
For couch use I definitely wouldn’t want something other than a MacBook Air, tbh. Everything else seems to run hot and have a shorter battery life. Reasonable enough as semi-portable workstations, not at all good for casual use possibly away from an outlet.
*nod* “My laptop” is a hand-me-down Lenovo Ideapad from the mid 2010s running Kubuntu like my desktop (a 2023 Ryzen for faster rustc) and, to be honest, the laptop I use most is the hand-me-down Thinkpad T410 where I took advantage of the availability of Windows XP drivers and Inexperience Patchers to make it a fairly convincing fake Windows 98SE.
(And then, only a week ago, after pouring all that effort in, I remembered that my “stuff to get to” pile contains a hand-me-down Thinkpad T42 that actually has Windows 98SE drivers available for download and won’t make me dance around Intel HD video drivers which don’t want to remember “preserve aspect ratio” across changes to widescreen resolutions. Oh well… I had fun.)
Seriously, though. As long as you’re not using a web browser, playing modern games, or decoding H.265, it’s ridiculous how fully functional Windows XP is with WinAMP 2.95 and ffdshow for the media side of things. Hell, my “gaming rig” is a 2012 HP prebuilt with a 2009 Radeon, a SATA SSD, and Windows 7 (KVMed in to fix a bunch of Gaming+multi-monitor papercuts on the Ryzen) and it still does 99% of gaming related things I care about. As long as you’re not into AAA games, the slow-down in tech is real.
I’ve been self-employed for a very long time, so the incentives are different, I guess. I don’t replace my laptop very often, but when I do, I don’t think about spending double that or more since it’s the way I provide for myself and any optimisation there is a no-brainer.
As for a personal laptop, I would tend to agree with you, my personal one is an 11-year-old 15” MacBook Pro, still running smoothly.
Yes and no. It’s about where the line for a laptop I’d expect to work well is, but there are a lot of cheaper models on the market. Dell’s web site has ones starting at £300 (including VAT, 6-core i3 processor, 15” screen, 8 GiB RAM. 256 GB SSD).
My experience with those is that they’re a false economy. The last Dell we bought was about £450 had the battery die after six months. The replacements were over £100. Dell support claimed that batteries were a ‘consumable’ not a replacement part (in spite of being listed under replacement parts on their web site) and so not covered by warranty. I was advised I could take them to the small claims court but I didn’t have the time nor energy, I just decided to never buy a Dell again and tell this story at every opportunity: it will probably cost them more in the long run.
In comparison, I’ve had several batteries die in Macs after a couple of years. Back when they were easily replaceable, Apple would courier replacements out to me and they arrived at 9am the day after I called their service line (as an individual, no corporate support). Now, I have to take machines into the Apple Store or have them collect the machine and then repair it and send it back, which would be more annoying if I still lived somewhere where the nearest Apple Store as a couple of hours train ride away.
Sometimes you get lucky with the cheap machines, sometimes you don’t. Some of the mid-range Lenovo models seem to last very well. The old IBM ThinkPads were tanks. I tripped over the charging cable of my R31 in the middle of a big compile job (hitting the disk hard) and kicked it across the room. Compiler paused for two seconds and it then continued as if nothing happened and the machine suffered no ill effects (MagSafe has saved me from doing the same with Macs a few times).
Some of the more expensive machines are also not worth it. Microsoft gave me a top-of-the-line Surface Book 2 (2018 model) when I was there. The touchscreen was nice. Being able to detach the screen was sometimes nice but the battery life detached was awful (it wasn’t great anyway) and they didn’t do the obvious thing of having the base work as a Bluetooth keyboard when detached, which would have made the tablet mode much more useful. Windows updates broke it periodically. Thermal dissipation was bad and video calls made it slow to a crawl (typing in Teams took seconds for characters to appear when a video call was running, and because async JavaScript is hard, they would often appear in reverse order, one second apart). A lot of this is because every sensible kernel has a rule that the only thing you do in ISRs is prod a lock or trigger a work queue, whereas Windows actually does a load of stuff in ISRs, which makes latency absolutely terrible, but some of it was hardware. It definitely wasn’t worth the £2500 price tag.
My last MacBook Pro was a late 2013 model. It cost around £2500 (almost all of the options turned up to 11), but lasted ten years of daily use (battery finally died after six or seven years, long out of warranty, and I replaced it myself with a fairly cheap replacement). It was much better value than the £450 Dell that lasted six months.
I’d expect a €950 laptop to be something decent that should last many years. Not the most powerful, but at least well built and fast enough for 90% of what I do.
I’ve also had a lot of luck with refurbished machines. You can often get a decent, modern machine with 30% discount just because it’s got a small scratch, or is a corporate return for some reason (you can tell by the higher windows tier and 3 year on-site support included). There’s some risk of course (returned due to some defect that the service didn’t notice/fix), but I’ve been happy doing this for years.
For personal laptops, I don’t see myself buying another one brand-new. I’m just an average power user, not an AI developer or gamer. Web browsing, coding, videos, and light retrogaming are the extent of my activities. I try to buy used or refurbished from a reputable dealer.
I was once within seconds of buying a Framework 16” laptop for something like $1,800. That would have been a hefty price, more than I’ve ever paid for a laptop, but I liked the product, the company, and what they stood for. Before I noticed that it wouldn’t even ship for 5 months at the earliest. I decided to look around and found a great deal on a used Dell Precision workstation laptop with around the same specs give or take a few GHz for $900. I sat on it for a while and went to check the price on the morning of Black Friday and found that they dropped the price by 50%, so I got it for well under $500. It’s not my favorite laptop I’ve ever owned but it does the job, especially for what I paid.
Yup, my partner got a second-hand late 2013 MacBook Pro about five years after I got mine new. Hers cost around £500 and was great. She replaced it with a new MacBook Air about a year after I replaced mine.
I had a laptop I used for years. The WiFi driver was broken under windows. So it was half off when I bought it refurbished. Machine worked great under Linux.
I just decided to never buy a Dell again and tell this story at every opportunity: it will probably cost them more in the long run.
I’ve only ever had excellent experiences with Dell. I don’t have a policy of saying so at every opportunity, but I feel like this is as good an opportunity as any to share my experience.
I’ve had four Dell laptops over the past 15ish years. Three I bought, one was inherited. I believe I bought one replacement battery during that time, although I didn’t actually use them on battery power much. I have found the quality of the machines and the customer service absolutely top notch, and my experience has been so good that I’ll never bother looking at another manufacturer unless it’s a “speciality” manufacturer like Framework, unless I have some particular needs for some reason, or unless I have a bad experience with Dell (which may yet happen).
I totally agree with all of that. What I want doesn’t currently exist but I hope it will soon. It’s something with the build quality of a MacBook but which not only runs other operating systems but is guaranteed (within reason!) to always do so. The heroic nature of the work that the Asahi Linux people have to do is impressive and commendable sure but also worrying.
The heroic nature of the work that the Asahi Linux people have to do is impressive and commendable sure but also worrying
They’ve had to reverse engineer the GPU and some other things, but Apple put in a lot of effort to make it possible to boot other operating systems without impacting secure boot of macOS.
I haven’t used on personally but everything I’ve heard about the Framework laptops has been glowing.
Their cheapest model is more expensive than the MBA in the article. Their cheapest model upgrade potential might make it cheaper in the long run, but given that I replaced my last Mac because it was starting to physically wear out, I’m not 100% sure.
I used to build my own desktops and upgrade them, but then I wanted a new CPU. The new CPU needed a new motherboard. The new motherboard needed new RAM and a new graphics card. At the end, I kept the case and hard disk, and at that point I had almost enough spare parts left over for a complete new computer, so the savings from upgrading were fairly minimal.
They’ve had to reverse engineer the GPU and some other things, but Apple put in a lot of effort to make it possible to boot other operating systems without impacting secure boot of macOS.
Yes! In case I wasn’t clear, I didn’t mean that Apple failed to be helpful. Still, the end result is that we can be much more sure that a generic laptop will run arbitrary other operating systems than that a modern-architecture MacBook will. So I’m waiting for the best of both worlds :-)
It’s about where the line for a laptop I’d expect to work well is, but there are a lot of cheaper models on the market.
I’d expect a €950 laptop to be something decent that should last many years. Not the most powerful, but at least well built and fast enough for 90% of what I do.
I agree, totally!
I also agree that Macbooks are worth the premium for such a nice polished device over competitors of similar and more expensive prices, and like you say, they can last quite a while.
But my point is still like.. €950? Its just not what I expect for a crappy piece of hardware like the author says in the intro. I wouldn’t expect a crappy macbook or a crappy non-macbook for this price. It’s even only slightly more than how much I’d expect to pay for a second hand macbook of these specs too.
Yup, after reading that line I was expecting it to be about a netbook successor or something. Cheap crappy laptops start at around €300 these days. Find one of those that’s good and you have made a great discovery. Find a thousand Euro laptop that’s good and, uh, yes, that’s expected. Spending a thousand Euros on a laptop and discovering that it’s crappy would be a story.
A few years ago I got a Thinkpad T14 AMD for 1389 Euro and the build quality was pretty underwhelming. I opened it to add RAM, but it didn’t seem like something I should do a dozen of times. Also, the electronics looked like a patchwork. Luckily this was during COVID scarcity and I could sell it without much loss. I think there are better-built ThinkPads, but they were more expensive than that. So I wonder what these high-end 950 Euro laptops are.
I replaced it with a MacBook Air M1, which was the same ballpark with some upgrades and it was much better built, far better display, faster, and quiet (the T14 could get noisy under load).
I usually recoup a substantial part of the cost of a MacBook and use it to buy the next one. I used to upgrade every 1.5 years or so, but since getting my 14” M1 Pro I haven’t felt the urge in years.
When the point comes back to switch to a non-Apple laptop with Linux, I’d probably get a refurbished ThinkPad or something like that.
It’s reasonably priced for what you get, in fact there’s nothing to compare it to in the PC space.
The last non-Apple laptop I used was a ThinkPad X1 Carbon (Gen. 4 I think) where I ran Arch Linux on. The ThinkPad cost three times as much as a M1B Air, but suffered from CPU throttling, loud fan noise, sub 3.5h battery life, and various hardware issues that led to a complete freeze of the machine.
I bought an M1 MacBook Air in late 2020 and have no real reason to upgrade since then, battery still lasts for 8h, it’s dead silent because its fanless, and I can charge it with any random usb-c phone charger if I need to! I’ve been a die hard Linux user before switching to Apple, but those M-chip MacBooks are so good that they play in a completely different league and I can’t imagine going back to a Lenovo or Dell laptop.
Mine cost $3500. It’s a System76, not Apple. It’s amazing. I don’t know how I would ever be satisfied with the specs of something that’s less than a grand.
Yeah, I’ve never spent more than $500 USD on a laptop. My first laptop was one of the 15” IBM Thinkpad models, $150 (used) and it lasted 3 years before starting to freeze up when its motherboard flexed a bit from picking it up. After that got a $200 new one, Lenovo branded Atom netbook called an IdeaPad, about 11”, and got 3 more years out of that one - and kinda technically i still have it and it still works ok for some things, but a new job made me do video calls and it was just too slow; the calls would lag out.
Then I bought a $360 laptop, again Lenovo, its model number sticker came off, but it is not one of the think/pad things, just some number… and it worked for 9 years of daily on-battery use (the battery would last about 3-4 hours at a time). I did buy it more RAM and an SSD totaling about $140 more on it over the years, which made a big difference, so total of about 500. My kid stepped on the screen and snapped a plastic support so the left side of the laptop is floppy now… and the battery started getting only an hour use out of it anyway, so I bought another $400 one. (I still have that old one on my desk because it has so many files and programs installed that I still want periodically, but it hasn’t been a daily use laptop for almost two years now.)
The new one is a HP Probook, I got it slightly used from a going-out-of-business sale. Been very happy with it so far - it feels like HP’s clone of the Apple Macbook, 13” screen, fancy lightup keyboard, similar size, shape, color, even the keyboard seems similar to a macbook air. Get about 6 hours out of a battery charge and it goes to sleep / wakes up very quickly too. I don’t know what this model cost new, but I can’t complain at all. Had it for two years now, so we’ll see on longevity, but zero issues; it still feels new.
My policy was I’d get cheap laptops because odds are something would happen and they’d break anyway, so might as well make the pain of replacement less….. but the reality has exceeded my expectations. I like this HP a lot, the smaller size and weight compared to those 15” ones is lovely. Maybe in a decade (hopefully!) i’ll even fork out for one of these new.
This is a lot closer to my experience. Get something that was real expensive when a business bought it a few years ago when they start replacing that model in their fleet.
These are European prices, they really aren’t comparable to aus or usa for purchasing power. I would also be very very skeptical for anything under 1000 euro. Sure you can get something, but nothing in the “nice” category.
European here. €950 is a big sum of money; most people wouldn’t dream of spending that much on a laptop. You can (and should) expect something decent for that price.
I mean doesn’t this depend? I imagine people in Germany or Ireland have less of a problem paying 950€ for a laptop, but that number is unimaginable in Portugal or Greece. Purchasing power in the eurozone is heavily varied.
Prices for identical laptops tend to be more or less the same in the EU (or you’d see a lot more grey import). And €950 is a serious chunk of money even in the richest EU member states.
Really? I know a lot of non-tech people who buy the entry-level MacBook Air (or upgraded to 512GB or 1TB storage) in The Netherlands (given your username, I guess you might also be from NL?). It’s fairly cheap relative to what people spend on cars, etc. Also almost 40% of the population owns an iPhone, which is close to a MacBook Air price-wise.
iPhones, MacBooks, etc. are very common in the richest EU states. If you go to a library, you’ll see that even a lot of students use MacBooks.
Yeah, I’m very happy I get a better product for less when I spend my own money.
My point is that this thread is full of statements like “there are better laptops out there”, but there are very few links actually supporting that statement. Please note that a used Thinkpad running Linux is not comparable to a NIB Macbook Air in this case. It has to be something a consumer can buy online or in a big box store.
Yes, and I’ve always had better equipment and a better working environment at home, paid for with my own money, than any employer has provided me with.
agreed, my daily driver cost refurbished 160 EUR (Thinkpad x250) and I consider it the best machine I ever had (including a m1), on par with a 13” macbook unibody late 2008 (snow)leopard.
I don’t care too much which end the machine is - high or low - as long it’s a means to my end.
He quotes “Today, 47% of U.S. women ages 25 to 34 have a bachelor’s degree, compared with 37% of men”, but fails to account that 80% of engineering degrees go to men.
But I think he misses something much bigger. The “DEI backlash” as he calls it, stems from a desire to defend and promote meritocracy, which is under heavy attack from DEI proponents.
I think it’s a shame to assume that people in tech “hate diversity”, when there’s a simple and reasonable explanation to people’s resistance DEI, which is also more generous.
The “DEI backlash” as he calls it, stems from a desire to defend and promote meritocracy, which is under heavy attack from DEI proponents.
Entirely incorrect. The DEI backlash was created by racists and sexists because they hate that tech is actually becoming more meritocratic. People are racist and sexist, so they will prefer white men even when they are not as good. DEI practices counter this tendency. But people don’t like the idea that they might’ve only been hired due to the having a racist or sexist bias, and thus lash out against those practices.
My view is definitely falsifiable. For example, if you demonstrate tech has become more meritocratic in the past decade. Or if someone pointed me to DEI manifestos that put meritocracy at the utmost importance (I can find plenty who write the opposite). Or if someone pointed me to a significant organization in tech that explicitly tries to only promote “white men”, or fight against the fact that most tech CEOs aren’t white. I’m not aware of any.
If you’re not interested this discussion, that’s fine. But when I see people write things that are incorrect, and just also happen to be pretty horrible, I feel the need to state what I consider to be the truth.
stems from a desire to defend and promote meritocracy, which is under heavy attack from DEI proponents.
About that…
It quickly became clear that digital computers could be transformative not just for war, but for other purposes. Universities and industry research labs, which had been the origin of many of the innovations necessary to create digital computers, began to invest in CS education. The low-wage pathways for women to jobs as human computers — majoring in mathematics and working in science and industry — disappeared, replaced by new pathways that focused on White men in higher education. For example, one of the first CS departments was at the University of Cambridge; it began its first degree program in 1953, founded by a White man, with an all White male faculty, teaching primarily White male students. The same happened with the first CS department in the United States at Purdue University, which was founded 1962 by a White man, who hired an all White male faculty and taught mostly White men. These first programs, which graduated the doctoral students who became faculty at many of the other universities founding CS departments, established the early conditions for the lack of gender and racial diversity in CS education, shifting CS education from something informal amongst women mathematicians, to something only accessible to students at elite institutions, who tended to teach only White men.
I don’t understand, what in this paragraph contradicts what I said?
What I think is great about software, is that anyone can write a program, even anonymously, and its quality can be measured with relative objectivity. Do you think that if women wrote Linux, or C++, or Python, Rust, Google, Reddit, or LLMs, etc., that the tech world would ignore them because of sexism? I find it hard to imagine.
I figured I would share this demo as a way to make folks aware that my gokrazy/rsync module can be used as a library — both its client and its server accept the io.ReadWriter interface type :)
I mostly like gokrazy for stateless systems though I know it supports persistent storage. I assume the rsync module is for file transfer to make a gokrazy system more stateful?
Yes, gokrazy is stateless by default, with a permanent partition for data you want to keep.
One great use-case for rsync in this scenario is to offer downloading the permanent partition contents for backups as described at https://gokrazy.org/userguide/rsync-backups/
Another use-case I recently had is to copy static website files onto /perm/srv for serving with the Caddy web server. After installing gokrazy/rsync on my router (https://router7.org/), I can now use normal
rsynccommands to send data from my PC to /perm/srv :)I worked with bank Python too!
Please share your experience, this article is one of my favorite of the last 5 years.
I know which bank the author worked for because I became aware of the ways these systems were named in the different banks. Orange site reference
The IDE is slow and clunky. Sourcecode stored as code meant I wrote my own grep tool to search a path in the database for source matching things, and even replace with a new pattern. It worked nicely for larger refactors.
Lack of git threw me initially. I grew to appreciate it. The systems running on here are not documented and not appreciated; it does not take much to attain wizard status with the system itself.
I wish more of it were available in the open source world. We’ve solved things differently in real life but your code being objects with a text attribute… unlocks all kinds of possibilities for autogeneration of repetitive code. None of this is actually done but it could be incredible if replicated elsewhere.
Worth noting the data store is not written in python, to my recollection it was written in C++ to meet the performance requirements. It just stored a whole lot of python. Essentially python is the DSL for working within the system.
A MASSIVE SIN in this system is the ease where a python function can call a function only defined in a subclass. Money comes down to simple addition many times, and it is very demoralizing to find this operation and start trying to work out how the data got there… only to discover that the concrete types of the objects involved is a mystery and thus so is the means of backtracking.
This is a previous discussion about the Tables from this article: https://lobste.rs/s/s3ewtr/tools_you_d_miss_if_you_left_company#c_obuusf
Do you have questions?
Thanks a lot for that! What I find fascinating about these stories is that they open a window on systems that have evolved and thrived outside of the mainstream. I’m particularly interested in the more radical choices made in these systems, and how they felt as a user/developer. In my past life, I ran a dataviz studio and built custom tooling, including languages (replacements for JS, HTML/XML and CSS), which offered a unique and extremely streamlined process. New hires would find the learning curve steep, but overall we got great consistency and productivity. Some later told me that they were missing some of the tooling, which was nice to hear.
I think the most radical idea that can be carried over elsewhere is the idea of infinite working environments, all visible to everybody, all usable by everybody, by any program at any time with almost no additional setup. You do the same setup you would for a regular program and you do a slightly different string and now you’ve got an entirely different base.
I have been chasing that with my CICD journey in the last couple years. I believe that developers should have infinite environments, and if you have a specific blessed stage environment, you’ve done it wrong.
And now that I think about it, that belief may have come from working on Bank Python and seeing how incredible it was to just pass a namespace to somebody else and they could just run the script I had without any additional review or checks. For business reasons, I would have my manager sign off before the operations team would mutate the production view of the database. But that was essentially it.
Actually, the database was the same way, wasn’t it? I could make changes to the database all day long to an overlay of the database. And I would. It was essential for the processing code to be able to process the data and write it somewhere. And then I would just run a diff between the production view and the view I created.
I don’t think most of our databases are set up to have that kind of infinite layering. The closest we get in the free and open source world is probably overlayfs for filesystems.
That’s amazing thanks a lot for the details. I find it fascinating when environments take the code and data as a whole, a bit like Smalltalk images, it is a very different model from the rather decomposed model we have now. In my previous job I actually started to do Postgres-based overlays using foreign data wrappers, so that you could mount the production database read-only and have all your writes local. Changes would then be the residual data on your local instance. In a way, I can imagine integrated Cloud-based environments (like Remplit or CloudFlare) go in the direction, where code and data are all accessible, shareable and “formable”. One last question: do you remember how permissions/security was managed? Banks are well known to be strict in that area, so I’m curious about how there was so much freedom of access/sharing.
Access was tightly gated on writes to the
pdatabase. This is what it meant to have code committed to the tree. I believe permissions could be added elsewhere in the system, I never saw it.By having access to this system at all I was “private side,” and my stock market activity was seriously curtailed beyond the normal curtailing done for all bank employees.
Writes were done by SSO authorization, your username and password were attached to every software deployment as well as to writes.
Dev teams were not allowed to write. Ops teams did that after signoffs. Ops teams were not “allowed” to do dev. So I’d email my manager for a signoff, and then that’d go to the ops team to actually run the script.
Maybe I’m missing something, but I don’t really see what makes the Clojure example “data driven”. It’s just using a different board representation.
From the couple of articles I’ve read about this, it seems “data driven” simply means “represent everything with generic untyped dicts and lists”?
If you have structs and records use them to.
Focus on the data, not on attaching functions to your structs.
There’s an implication that you might avoid certain data-oriented solutions if they would run into awkward parts of your type system. What I interpret then is that it’s data-driven because that’s the only thing you need to worry about up front.
This is a really interesting comparison. I hope one of the experienced Clojure developers here can confirm (or deny!) your suspicions :)
Incidentally, “Peg Thing” is more usually called solitaire.
Which suspicions? I found it an interesting foray into the emphasis OCaml puts on types. The clojure stuff seemed right.
As a self professed generalist this is one of the things I advertise about myself. I’ve given up trying to signal mastery in anything but my ability to figure things out with the help of things like vague awareness, trained over years of doing and exposing myself to as much as possible.
How do you sell yourself on a resume? Job hunting as a generalist is not easy.
I’m not a very good salesman until I get my foot in the door. My resume is admittedly pretty honest about all of the cool stuff I’ve done, which happen to be not confined to one niche. So for better or worse I haven’t tailored my resume to particular roles nor found it easy to. I agree it’s not easy. If I were in a position to hire someone I’d like a team of generalists advised by one or two specialists until more expertise is needed. I get why companies would want a perfect vertical fit in theory but I think that leads to resume embellishing at best or an opportunity cost in hiring someone that’s self sufficient and capable to do the work now.
I think the Temporal programming model is pretty close to a distributed systems approach that respects existing languages. You write your pure code, and your non-pure code, you run it somewhere, and Temporal makes sure what your pure code describes happens.
It’s a different programming model than people are used to, and doesn’t feel Native to any language it’s in, but it does feel like it belongs in every language it’s in.
Meanwhile the work itself isn’t distributed. It’s highly local. Just… local to anywhere your code is running.
Yes, as the article describes, a two phase commit in Temporal is harder to pull off… but Temporal makes that pain apparent, and doesn’t make it any logically harder than it logically has to be. And if it works, you know it’ll continue to work, even with node and network unreliability.
Maybe we’ve only stalled because fewer of us are on the durable computing bandwagon, and the world will catch up soon?
Having owned a Framework since April of 2022, I cannot recommend them to people who need even basic durability in their devices. Since then, I have done two mainboard replacements, two top cover replacements, a hinge replacement, a battery replacement, several glue jobs after the captive screw hubs sheared from the plastic backing…
It’s just such an absurdly fragile device with incredibly poor thermals. They sacrificed a ton of desirable features to make the laptop repairable, but ultimately have released a set of devices that, when used in real-world settings, end with you repairing the device more often than not. And these repairs are often non-trivial.
I will personally be migrating to another machine. The Framework 12’s focus on durability may be trending in the right direction, but to regain trust, I’d need to see things like drop and wear tests. A laptop that can be repaired, but needs constant upkeep/incredibly delicate handling, is ultimately not an actual consumer device, but a hobbyist device.
Maybe they’ll get better in a few years. Maybe the Framework 12 will be better. Their new focus on AI, the soldered RAM in the desktop offering, and the failure to address the flimsy plastic chassis innards, among other things, mean that they have a long way to go.
It’s definitely a “be part of the community that helps solve our product problems” sort of feeling.
I have an AMD FW13, and was trying to figure out why it loses 50+% of its battery charge overnight when I close the lid, because I don’t use this computer every single day and don’t want to have remember to charge yet another device.
So I check the basics-I’m running their officially supported Linux distro, BIOS is current, etc. And half an hour into reading forum threads about diagnosing sleep power draw, I realize that this is not how I want to spend my time on this planet. I love that they’re trying to build repairable/upgradeable devices, but that goal doesn’t matter so much if people end up ditching your products for another option because they’re just tired of trying to fix it.
I’ll chime in with the opposite experience - I’ve owned an AMD Framework 13 since it came out, and had no durability issues with it whatsoever, and it’s been one of my 2 favorite computers I’ve ever owned. I’ve done one main board replacement that saved my butt after a bottle of gin fell over on top of it in transport.
Development and light gaming (on Linux, I very much appreciate their Linux support) have been great, and the reparability both gives me peace of mind, an upgrade path, and has already saved me quite a bit of money.
I’ve owned a framework since Batch 1. Durability has not been a problem for me. My original screen has a small chip in it from when I put it in a bag with something that had sharp edges and pressured the screen for a whole flight. Slowly growing. Otherwise, it’s been solid.
Same. I have a batch 1. There are quirks, which I expected and knew I am supporting a startup with little experience. I since have upgraded and put my old board into a cooler master case. This is so amazing, and what I cared about. I am still super happy with having bought the Framework and particular for tinkerers and people who will have a use for their old mainboards it’s amazing.
I get harbouring resentment for a company you felt sold then a bad product. But at the same time, you bought a laptop from a very inexperienced company which was brand new at making laptops, a pretty difficult product category to get right when you’re not just re-branding someone else’s white-label hardware.
3 years have passed since then, if I were in the market for a category which Framework competes in these days I would be inclined to look at more recent reviews and customer testimonials. I don’t think flaws in that 3 year old hardware is that relevant anymore. Not because 3 years is a particularly long time in the computer hardware business, but because it’s a really long time relative to the short life of this particular company.
I would agree that 3 years is enough time for a company to use their production lessons to improve their product. But nothing has changed in the Framework 13.
I don’t resent Framework. I think that’s putting words in my mouth. I just cannot, in good faith, recommend their products to people who need even a semi-durable machine. That’s just fact.
Founded by people who had experience designing laptops already, and manufactured by a company that manufactures many laptops. Poor explanations for the problems, IMO.
I’ve had a 12th gen Intel since Sept 2022 (running NixOS btw) and I have not had any issues, I will admit it sits in one place 99% of the time. I might order the replacement hinge since mine is a bit floppy but not too big a deal.
As for the event, I was hoping for a minipc using the 395 and I got my wish. Bit pricey and not small enough for where I want to put it and I have no plans for AI work so it’s probably not the right machine for me.
I was originally interested in the HP machine coming with the same CPU (which should be small enough to fit) but I’ve been pricing an AMD 9950 and it comes out cheaper. I was also disappointed there wasn’t a sku with 385 Max w/64GB of RAM , which I might have have ordered to keep the cost down.
For reference a new machine is intended to replace a 10 year old Devils Canyon system.
I’ve also had my Framework 13 since beginning of 2022. I’ve had to do a hinge replacement, input cover replacement, and mainboard replacement. But I sort of expected that since it’s a young company and hardware is hard. And through all of it support was very responsive and helpful.
I would expect that nowadays the laptops are probably more solidly built than those early batches!
Support was definitely helpful. I just don’t have time or money to replace parts on my machine anymore.
From what I understand, the laptops aren’t any stronger. Even the Framework 16 just got some aftermarket/post-launch foam pads to put below the keyboard to alleviate the strain on the keyboard. The entire keyboard deck would flex.
The fact that these products have these flaws makes me wonder how Framework organizes its engineering priorities.
When compared to other similar laptops from brands like HP or Lenovo, how does the deck flex compare? I definitely feel sympathetic to not being better or on par with Apple - given the heaps of money Apple has for economies of scale + lots of mechanical engineers, but it would be a bit rough if mid-tier laptops in that category were far superior.
The deck flex is on par with or worse than an HP EliteBook circa 2019. The problem is that it’s incredibly easy to bend the entire frame of the machine, to the point where it interferes with the touchpad’s ability to click.
It’s really bad, bordering on unexcusable. The fact that there’s no concrete reinforcment says that they sacrificed build quality for repairability, which is equivalent to making a leaky boat with a very fast bilge pump.
I’m not sure what you’re doing to your laptop; how are you bending the entire frame of the machine?
It’s a new company that is largely doing right by open source, and especially open hardware. The quality isn’t incredible but it is worth its value, and I find these claims you’re making dubious.
It’s a fairly common flex point for the chassis, and a common support problem. The base of the mousepad, towards the front of the laptop where there’s a depression in the case, is where the majority of the flex is.
My laptop has seen nothing but daily, regular use. You can find the claims dubious, but others are having them too.
This has been my experience with the Framework. It’s not Apple hardware, which is best in class all around, but it is on-par with my Dell XPS.
I’ll chime in too: I’ve had the Framework 13 AMD since it came out (mid 2023) and it has been great.
I upgraded the display after the new 2.8K panel came out, it took 2 minutes. Couple months later it developed some dead pixels, so they sent me a replacement. In the process of swapping it out, I accidentally tore the display cable. It took me a while to notice/debug it, but in the end it was just a $15 cable replacement that I’m fairly sure would have otherwise resulted in a full mainboard replacement for any other laptop. (When I had Macbooks, I lost count how many times Apple replaced the mainboard for the smallest thing.)
I haven’t been too precious with it, I toss it around like I did my Thinkpad before this. There’s some scuffs but it has been fine, perhaps the newer models are more sturdy? It’s comforting to know that if anything breaks, I’ll be able to fix it.
I also run NixOS on it, it does everything I need it to do, the battery life is great (8-10 hours of moderate use) and I’ll happily swap out the battery in a few more years once it starts losing capacity.
I spend so much of my life at the computer that feeling a sense of ownership over the components makes a lot of sense to me. I don’t want to feel like I’m living in a hotel.
It is, in fact, how I want to spend my time on this planet.
To add to the chorus, I bought a 12th gen intel framework 13 on release and it’s been flawless so far. Nixos worked out of the box. I love the 3:2 screen. I can totally believe that a small/young manufacturing company has quality control issues and some people are getting lemons, but the design itself seems solid to me.
On my old dell laptop I snapped all the usb ports on one side (by lifting up the other side while keyboard/mouse were still connected). Since they’re connected directly to the motherboard they weren’t repairable without buying a new cpu. If I did the same on the framework it would only break the $12 expansion cards and I wouldn’t even have to turn it off to replace them.
Later I dropped that same dell about 20cm on to a couch with the screen open. The impact swung the screen open all the way and snapped the hinges. They wanted me to send it back for repairs but I couldn’t handle the downtime, so for a year I just had the hinges duck-taped together. I’ve dropped my framework the same way, but because the screen opens the full 180 degrees it doesn’t leverage the hinges at all. And if it did break I’d be able to ship the part and replace it myself.
Not that I support the desktop offering as anything but waste, but the soldered RAM is apparently all about throughput:
With the focus of the desktop being “AI applications” that prioritize high throughpout, I’d say they could’ve gone with an entirely different chip.
I get the engineering constraint, but the reason for the constraint is something I disagree with.
Who else is making something competitive?
I wish I could name something in good faith that was comparable to a hyper-repairable x86-64 laptop. Lenovo is pivoting towards repairability with the T14 Gen 5, but I can’t recommend that either yet.
Star Labs, System76, some old Thinkpad models.. there are “competitive” things, but few things that pitch the things Framework does.
While I agree on some of that, I must stress that I’ve had hardware that was fine until just one thing suddenly broke and everything was unusable. I’ll try an analogy: with repairability, if all your components are 99% reliable and working, the whole machine is at 99% but without it, even if all of them are at 99.9% instead, when you have 10 components, you’re not in a better situation overall.
And I say that while I need to finish going through support for a mainboard replacement due to fried USB ports on a first-gen machine (although not an initial batch). BTW, funnily I’m wondering if there’s an interaction with my yubikey. I also wish the chassis was a bit sturdier but that’s more of a wish.
As for thermals, while I think they could probably be better, the 11th gen Intel CPU that you have (just like I do) isn’t great at all: 13th gen ones are much better AFAIK.
I’ve experienced a full main board failure which led to me upgrading to a 12th gen on my own dime.
The thermal problems are still there, and their fans have some surprising QA problems that are exacerbated by thermal issues.
I wish I could excuse the fact that my machine feels like it’s going to explode even with power management. The fans grind after three replacements now, and I lack the energy and motivation to do a fourth.
I think 12th gen is pretty similar to 11th gen. I contemplated the upgrade for similar reasons but held off because I didn’t need to know and the gains seemed low. IIRC it’s really with 13th gen that Intel improved the CPUs. But I agree the thermals/power seems sub-par; I feel like it could definitely be better.
BTW, I just “remembered” that I use mine mostly on my desk and it’s not directly sitting on it which greatly improves its cooling (I can’t give hard numbers but I see the temps under load are better and max CPU frequency can be maintained).
Sorry to hear about the trouble with your Framework 13. To offer another data point: I have a 12th gen Framework 13 and haven’t needed to repair a thing, I’m still super happy with it. The frame-bending I’ve also not seen, it’s a super sturdy device for me.
I can second that. I’ve had a 12th gen Intel system since late 2022 and no issues of the sort. Even dropping it once did nothing to it
I’d love to see the discussion in which they decided to rename dreamberd to gulfofmexico
I don’t understand the joke of the original name.
But there were blog posts about how Apple can’t access it even if they wanted to?
It is, of course, complicated. This is a simplified view. I am leaving some very important assumptions unanswered (such as: how does iMessage do key exchange). Disclaimer: I am an idiot, probably compromised by the government, and you should independently verify the claims below, most of which are probably completely wrong.
There is a lot of trust, and a lot of external review around this. In some markets, such as China, you cannot disable Key Escrow, but iMessage is (supposedly) E2EE - a rare exception.
So like… make your own conclusions about this, but I’m relatively confident that Apple currently cannot access your data with ADP on. Of course, I can be wrong and even more of course, things may change. They could, for example, silently disable ADP with an iOS update - such a move would hopefully be catastrophic to their public image. Fun fact: this is (possibly) why Apple requires a PIN to update iOS on demand.
TLDR As it stands and as it’s publicly communicated (and verified by researchers): Apple can’t access your data on your phone but they can access anything stored in iCloud unless you use ADP in which case they can’t, unless you use iCloud on web. Your mileage may vary. Promotion not valid in China.
[0] https://www.wired.com/story/apple-csam-scanning-heat-initiative-letter/
[1] https://support.apple.com/en-us/108756 <– this is the most important footnote
[2] Apple claims that Apple does not maintain key escrow. I’m not saying to the contrary, I’m just highlighting one of many very important assumptions people might make
Could you elaborate on the “verified by researchers” part?
There are lots of so-called researchers who regularly verify so-called Apple’s so-called claims.
Wat?
Surely, there should be other so-called researchers to call the so-called researchers on their bluff for only so-called checking on so-called Apple’s so-called claims.
I don’t know that it’s right to ascribe to conspiracy what simply could be people doing their job well.
The important thing is, if Apple were lying about when or how much they retain keys, how would any so-called researcher know?
Maybe such evidence could become part of the public record in a court case, if it weren’t redacted for national security or whatever, or if the government sloppily didn’t keep it out of the record with their famous ‘parallel construction’ prosecutorial methods. Or maybe an Apple employee could leak, but they sure seem incentivized not to. I don’t know how much to infer from Apple’s selectively vague statements. I get that they’ve marketed themselves as privacy-forward and there’s reputational risk at play. But I also see they just paid $95M to make a specific claim that they might be up to some funny business just go away with no discovery or legal conclusions.
Bottom line for me is, unless I fully control the encryption process and key storage, E2EE isn’t all it’s cracked up to be, and I don’t care how slick the marketing is.
how are people at a company supposed to do their jobs well without conspiring?
Oooh good point.
It’s a conspiracy! People are doing their jobs exactly as desired!
it would be really weird for engineers at a large company to be working in isolation from one another.
They easily can since they control the end point software and the end points have the unencrypted content
They “easily can” do “what” exactly?
Please expound.
Also when you say “endpoint” I assume you mean “iDevices”. Please confirm.
Their model is published regularly on their site, periodically: https://support.apple.com/guide/security/welcome/web
3rd parties continually review although that is at the 3rd party’s whim and not perfect.
And yes, there’s a lot of trust, but there’s also a lot of internal and external researchers.
I, personally, trust Apple is doing what they’re saying. I am also an idiot.
If I had (active) government level entities attacking me, or was worried that there were, I would probably completely change how I communicate and interact with communication technologies.
I’m being a little pithy here btw :p but trust and trusting is an important component of this system, and it’s up to you to inform yourself and individually assess your posture.
Ideally? Maybe this would be regulated by strict pro-privacy laws that deeply incentivized companies like Apple to be pro-privacy.
Idk
Like I said in my root comment, I’m probably a govt shill who shouldn’t be trusted and you should verify for yourself
Not just the idevice (though os control is enough to get all messages in clear text of course) but with iMessage also control of the app itself.
No matter how it is encrypted on their server to begin with, the app or os could easily eg upload the entire db of unencrypted messages to apple at any time.
At one point I said that a good day is a day where you know more at the end than you did at the beginning.
Then I started keeping a file, and I’ve been keeping this file for two years now, where every day has a list under it of the things I have learned that day. Some days don’t get an entry at all, in which case I don’t record the date. But the list items are basically as long as my terminal is wide, or shorter. Tweet-sized.
I optimized my system for handling grep so I can grep through my knowledge base and discover when I tried to do something before what it was I did. It’s my answer for “this thing I was confident about is on the tip of my tongue, but really, I don’t remember what it was.”
It’s even less structured than the TIL system. And it works great for me.
If this were ever an option, I think we’re too late for it now.
In any case I subscribe to the “computers must not speak unless spoken to” philosophy, and hope for that kind of future.
“Visitor at front door” is a computer speaking without being spoken to…
Yes, but playing a doorbell chime isn’t.
I like the idea of challenging norms, but its very smart to assess how critical something is before working on it.
Yes, it consumes time, but engineering resources are typically limited (even more so after hours ;)).
Triage that can be done by less qualified automations or even people, definitely aid in the mobilisation of an appropriate effort.
Your “SEV” levels just need simple definitions:
Blackout vs Brownout; User-facing/External vs Company-facing/Internal. Your issue severities should be a combination of those. Though this assumes your internal systems can survive a night or 5 of being unavailable- that database backup thats failing for 4 days could suddenly become incredibly important.
What about severities that are a lookup of t shirt levels of bad.
First digit for how broad the care group is. IT, engineering, company, some customers, most customers, all customers.
Second digit, data integrity risk. None, low, high, already lost some. Maybe this should actually be risk of altered performance after this incident.
Third digit, how long can it stay stable before the first digits get worse. A month, day, hour, It can’t get worse.
This would be a lot for an engineering team to manage, but not for an incident manager who’s primarily communicating to others. And also lets others see… We have a SEV-1, but will I care about it tomorrow.
Color code each column for maximum communication
I think that’s a pretty solid approach actually, if your company is Disney sized or something I can see a policy like this working extremely well.
Though you’d need punitive measures in place to prevent the stealing of resources, as soon as subjectivity enters something like this then managers game it to get more attention than they should to pad their numbers.
A smalltalk where every single message is defined as a file in the file system. To be one of the easiest glue languages.
Main mainWhen I first dreamed of this, I figured I would write it in C for the ability to load objects. These days I would write it in Zig because the ability to compile and load and work with the compiler at runtime is incredible.
This would change the nature of the image in smalltalk to be fully compatible with Git and modern deployment mechanisms of “tar it and copy.”
Have you seen Storage Combinators?
Not before. It looks like a tool for treating arbitrary endpoints in a RESTful world as actors?
I hadn’t considered it from that perspective; I guess so!
The way I’d internalised it was as mapping HTTP’s verbs and semantics to Smalltalk method dispatch. In section four they show how the principle can be extended to filesystems.
Smalltalk method dispatch gets a nice separation between invocation and implementation.
Similar effect with HTTP verbs.
So does event streaming.
So do Doors for the Solaris filesystem.
These things end up increasing reuse, since you can “just” trigger an event or a PUT/GET or open a file or dispatch a message and your obligation has ended. If that invokes a whole world of IT compliance or business logic mazes or triggers a supercomputer batch job… you don’t care. Suddenly previously built components can, with minimal knowledge, be glued together. Assemble a table of ways to get the result you need, and put a little “cost” next to each one, and even business people can make some decisions about what gets glued to where to achieve your goals.
We have a lot of languages to do work. We have far fewer good options for glue languages. TCL tried a bit. Python tried a lot. Lua tried a bunch. None of them managed to really explode into “Glue all the things together” territory.
Hence my dream of doing this as local objects, functions, commands, and interpreted piles. Escape from “GusteauTalk” (as my .git repo is named for this project) into a Real Language as soon as you like.
I can’t do all my BigInt processing as HTTP calls, the overhead is too high. But local process execution, or Dylib invocations? I think we can do that.
Where do you put Plan 9 / 9p in this?
It’s a useful implementation to study. I’m not really sure why it didn’t catch on when these other mechanisms did.
Holy hell, this solves the diamond inheritance problem so effectively. If I recall (it’s been a long time), Smalltalk typically wants you to use mixins instead of subclasses. This approach is such a simple and effective way to do highly granular mixins. Very nice!
Traditional Smalltalk doesn’t have mix-ins and does absolutely encourage subclassing; I think you’re thinking of Ruby. Many modern Smalltalks do have mix-ins in one form or another (e.g., Pharo has Traits), but that’s a fairly new additon, and IMHO doesn’t honestly mesh terribly well with the rest of the system
Why message per file? What about smalltalk makes it hostile to writing code for it like a regular programming language, where you put all the code for a class in a text file, and you do inheritance via syntax in a text file rather than symlinking a bunch of stuff?
Smalltalk has no syntax nor does it need syntax for defining the message interface. The per-file approach allows you to save the message interface as the filename, keeping it out of the syntax of the message itself. It also allows sharing implementations with different messages if suitable.
Smalltalk being image based makes it hostile to modern deployment mechanisms. Arguably these are not necessary with smalltalk, where you ship the image instead of shipping a separate built artifact, but it’s difficult to impossible to “dip your toe in” on doing so with such a radically different deployment story.
Smalltalk is designed generally around the Image. You write your code in the image and you ship the image. Anything to convert to git or whatever is a translation layer.
Is the image file (as a concept, not the ST implementation) really a barrier in this day and age of Docker? What if they were layered something like Docker, where you’d have the language runtime, on top of that you’d have two “branches” of images, one containing your running / frozen project, and another containing the common development tools alongside your project-specific tooling and any source code you consider important not exist in the distributable artefact? The two “VMs” would communicate using code defined at the lower language-runtime level that is common to both.
No. The image is great for deploying server apps and that!s the one place Smalltalk is still used. The problem with the image abstraction is that it splits the world into Smalltalk and alien parts. If you want to use a library that is not Smalltalk, either it’s stateless or it breaks the image abstraction. This is why Étoilé moved away from image-based persistence to versioned object graphs for model objects, so that you had a clean language-level abstraction for separating persistent and ephemeral state and could either explicitly serialise foreign state or treat it as ephemeral.
So long as it can be packaged in an OCI image, I think it’d not be a blocker to deployment. Things like devpods show that OCI images make a compelling base.
Smalltalks generally don’t know what a filesystem even is, though.
I agree with the article, but I find the term “near-miss” to be grammatically confusing. The linked NPR article even switches to “near-collision” to be accurate.
George Carlin would be proud
There is nothing confusing about the phrase. Ask yourself a question like “Is the near future the future?” Now ask yourself “Is a near miss a miss?” You can’t answer “yes” to the first and “no” to the second without being inconsistent.
What I think trips people is that “nearly” is a word that negates what comes after, and for some reason in this single specific phrase they try to parse “near” as actually being “nearly”. But that’s not the word. You can even hear Carlin do it with his “they nearly missed”. But if you actually just look at the actual words, there’s nothing confusing about them.
Does anybody here use this? What Android apps work well on PC?
I’ve used it. Most apps work great, as the environment really isn’t all that different from what they’re expecting, especially if you use a gapps image. High-profile games are less happy though. You can’t play Genshin Impact :)
It works great until it doesn’t. In particular, I run it in the “experimental” windowed mode, which is indeed experimental and has some really weird problems, some of which I still don’t understand, some of which I do as just “no one thought of this.” For example: If you move a window to the bottom of the screen, you can’t click on anything on the pixels near the bottom… Because the Android 11 navbar is down there and blocks that area from being interacted with.
I’d report the bugs, but the project seems very underfunded/underdeveloped for something that distros are shipping to end-users. As far as I know it’s still broken out of the box on Fedora.
I’m guessing most people are just running it in the normal fullscreen mode and therefore are having fewer issues.
I use it to listen to Apple Music on Linux instead of using their mediocre web UI. I have to use one version older of the app to get it to detect the internet correctly but it works pretty well overall.
I tried in the last month but the LXC container stubbornly refused to start. Ended up emulating Android instead in android studio
Waydroid requires the “binder” kernel modules. Apparently that’s shipped with the upstream kernel now, but I used to have to use the linux-zen kernel to have it work in the past
Oh, this was a humdrum inability of the network stack to find the bridge it actually wanted.
I haven’t tried, personally, but I read that this is the best way to run Minecraft Bedrock on Linux.
The problem with assigning meaning for a decision making process to numbers is that people will do math with it. You wouldn’t count CVEs, would you?
Maybe names, logos and websites for severe bugs were actually the better solution ;)
Jerk. I read that right as I was swallowing some coffee, and I burst out laughing. Coffee in the nose is not a nice experience, and I blame you.
(And I think the overall problem is our urge to turn what should be a qualitative analysis into an easier-to-measure set of tickboxes. The tickboxes work to assess bulk trends, but not to assess individual situations, and we care about not having individual sites exploited.)
My apologies. You may redeem this comment as a voucher for 1 (one) free coffee if you ever find yourself in Berlin, Germany. :-)
Unfortunately, I have to give this a CVE a 10 because you could be exploited for a free coffee with a simple social engineering attack, likely including the word “please”
I have been exposed! :-)
How about a trophy case section on your landing page with logos for all the brand name bugs closed in your product. :P
Huh. I’ve made filesystems with a billion empty files, to compare generators and filesystems, and xfs performed poorly, with its default settings. ext4, ext2, and btrfs finished within a day or two; xfs was on track to take weeks.
Not to be rude, but… so what? I’ve seen some very degenerate filesystems, but none with anywhere near a billion files.
Yet!
Surely this is a forum that rewards testing software to cursed limits!
not sure if the data-point is interesting here, but on my 13T external that’s formatted with modern NTFS, 1_296_232_240 ) inodes are allocated (not used). I could see it working and not as outlandish test it seems to be at first glance.
I checked my notes: it definitely bogged down within the first four million files, which doesn’t feel massive, but maybe that’s relative. I think it bogged down much sooner than that. After hitting ctrl-c, it took 20+ min to get a responsive shell. Unmounting took a few minutes.
Sorry if it came off as a glib “your favorite filesystem sucks”.
Nah, it felt more just like “it’s unreasonable that this thing fails under unreasonable circumstances”. I hope I wasn’t too glib in response; thanks for checking your notes. 4 million files is a much more reasonable circumstance, and bogging down earlier than that is definitely not a good thing. It does surprise me ‘cause XFS was supposedly designed for big chonky systems and in my experience has been quite bulletproof. If only I had the energy to reproduce it and do a a deep dive to find out what’s going on…
PC was never better and more alive?
You can install Linux on a mobile phone, Steam Deck is Linux and almost all games nowadays just work on Linux.
We have a practical modular and repairable laptop like Framework.
Building your own PC has never been simpler and easier - it’s almost like legos at this point.
PCs are most silent, energy efficient, nicest form factors. You can get PC on a usb stick, or a tiny box.
We have reproducible Linux distros like NixOS and Guix.
We even have an Open Source instruction set: RISCV.
Everything is awesome if you want to do personal computing. It’s just turns out most people don’t want a PC, and they want a stripped down, dumbed down, appliance instead.
It looks most haven’t read the article up until the point where it addresses exactly what you’re talking about. And it says:
See, there’s a problem with all your examples (Linux, Framework…) except perhaps the Steam Deck: none of those are mainstream yet. And the effort to make them so is significant, possibly herculean. But most of all, political: companies do bad stuff only because we let them. They’re allowed to use DRM, and have been given the power to sue us if we break them — at lest in the US. They’re allowed to collect and sell our data. They’re allowed to “sell” us stuff, even though it’s not actually a sale, and when it’s inconvenient to keep us locked in they’re allowed to lock us out. We should question those allowances. That is, push for better laws.
Personally, I would love to see the 30 million lines problem solved, and have someone deliver a simple computer that’s good enough for most uses — including gaming. Probably has to be something like RISC-V, considering that one is not allowed to just make and sell a CPU that’s compatible with the mainstream ones (ARM, x86…) — which is utterly ridiculous by the way, instruction sets are interfaces and as such anyone should be allowed to copy it, no authorisation nor fee required.
But I’m not sure we could ever have that without first cleanly separating hardware companies from software companies. The rule is simple: companies that distribute hardware should be forbidden to distribute software — not even Open Source. They’re only allowed to provide the manual required to operate the hardware. It can be done in a kinda extra-territorial way too: if a company distributes both, we just don’t buy their hardware. (And there are ways to determine if 2 companies really are separate.) Do that, and suddenly hardware companies will be motivated to make simple hardware with standard interface, and software companies (and open source alternatives) will be able to actually compete, instead of having to do everything on top of the 3 top kernels out there (NT, Linux, Darwin). Though it might be too radical to seriously consider. I like the idea, but I’m pretty sure it would have a host of adverse effects.
If people wanted it they would legislated it.
People don’t want personal computers, just like they don’t want to do their own plumbing, do their own banking, etc. Before personal computers they just didn’t use computers at all, now they are just using appliances.
Masses as perfectly happy buying a locked down iPhone, Android, Macbook, Spotify subscription etc.The system works for them perfectly fine, also better than ever. They don’t want freedom and empowerment (but also responsibility) that comes with actually being in control.
But. Personal computing is not dead, just because masses don’t want it. There’s probably more total personal computer users now than there was 20 years ago. And what they can do is better than before too. It’s just that the total number of “computing as a service/appliance” market is now almost everybody else.
I call out just-world fallacy. Many things are decided at scale that people do not want. To give you 2 examples of blatant examples that happened in my country (France):
Also I recall a US study that say that whether a reform passes or not is mostly determined by what the rich want, not by what the people want: if the rich want something and the people don’t, it often passes, anyway, and if the people want something that the rich don’t, it never passes.
It would be more accurate to say that if rich people wanted it they would have legislated it. And I have a feeling that true personal computing actually goes contrary to the interests of the tech giants. Just do not forget that what tech giants want, may be very different from what people want — and that’s before we touch on whether “want” is conditioned on being informed or not.
That I can get behind. I’m sad about it, but yeah. People don’t (want to?) understand what computers actually are, and I can believe that most are happy with appliances.
I’d agree that many people are like that. But are most? I’m not sure to be honest. If we just take the locked down part for instance, are people actually happy they can’t install third party apps on their iPhone without oversight from Apple? Are people actually happy they don’t even have the option?
There’s a network effect though. I gotta use the messenger protocols (and therefore, app) everyone else is using, else I’m being locked out of my friends and family. If I want to pay for stuff online I need to install the 2FA banking app on my phone, and depending on the bank that phone better not be rooted. Overall, there’s a social pressure not to do anything that causes friction with the way other people work or communicate.
For instance, people tell me that recruiters are all on LinkedIn now. But I may have to set up an account and do some of their bullshit if I actually want to be seen by recruiters and raise my salary. Knowing that doing so will ever so slightly contribute to the strength of this network, and ever so slightly increase the incentive for everyone else to stay or follow suite. Same deal with Facebook, I have known some groups who only communicated through it. The result was I wasn’t informed, and soon my involvement dwindled. I don’t want this crap, but as a result I’m locked out of some things.
To be honest, I fear the day when personal computing starts to be banned altogether. I know it sounds tinfoil hat, but that conspiracy already started with the DMCA, where besides a few exceptions people are forbidden to even talk about flaws in their own computers that they bought. What if a sufficiently right-wing government with crazy-rich control freaks in it decide that people need to be “protected”, and all new computers sold must run Trusted Boot, no opt out, and only a few select vendors are allowed to sign operating systems for it? Sure they’d have special provisions for programmers, but in this world people who want actual personal computers are likely to find themselves outside of the law. It’s no question some tech giants actually want this world. Apple led the way with their locked down iPhone, and got away with it because unlike Microsoft back in the day they never dominated the market so completely, but ever since everyone else want a slice of the pie, with Microsoft itself attempting the locked down app store model on desktop computers — fortunately so far they have failed.
Granted, that dystopia is far from inevitable. But it’s probably more likely than most are willing to admit, especially taking into account the rise of fascism.
To achieve what you’re saying, you would need to have any hardware company pair with a software company to have this software available for their hardware when they launch and vice versa. Consider that IBM wrote software simply so that they could sell their computers. If you sell a very nice brick, nobody is going to buy it until they can do something with it. Unless, of course, you go very cheap.
Which reminds me, a company has tried this. They have a hardware team, and they make hardware, and they expect the community, or some other company, to make the software. They are called pine.
In a world where companies are allowed to ship software with their hardware, a pure hardware company is indeed at a fatal disadvantage.
But what if no one can do both? One possible outcome is that no one tries radically different hardware: everyone would stick to existing designs and just make them faster, or add instructions here and there at a pace they can expect the software folks to catch up quickly. My hope is that we would see radically different designs, but most of all they’d be radically simpler (at the ISA level, the internals may optimise like crazy), so the software folks can pick it up quickly enough for the new thing to be competitive at all.
I agree. I would like a RISC-V desktop and a modular open-hardward ARM laptop, and they should run the particular set of packages that make up my desktop, assembled and tuned specifically for me. That’s all doable, I just can’t afford it. But it’s still cheaper than the Apple gear people talked about recently on here.
Sorry what? Is it just me or is this kind of out of touch? (Or am I just too frugal?) I get its an Apple product but this is still expensive in my books. This is about how much I’d expect to pay for a new decently good laptop. Converted to AUD this about how much I paid for my brand new at the time, Lenovo X1 Carbon (gen5 - it’s on its way out now).
I’d not spend anywhere near this for a laptop I’d expect to be “crappy”.
Agree. I have never, and would never, spend this much on a laptop unless it were something incredibly special. Thinking of this as “cheap” feels like the author is on another planet
Really? I feel like the average high-end laptop goes for $1,500+
But as the author stated it’s for couch/personal use.
Who needs a “high-end” machine for that?
My personal laptop is a T460p from 2016. It can do everything I’ve ever wanted to do (except 3d games). I don’t see any real benefit that my work M3 MBP has. Nothing. Maybe compiling stuff faster, I’ve not compared anything.
I have a gaming laptop here that cost 1500 and played every game in HQ that was tried so far.
“average” is way out of touch.
I like using a laptop that has battery life I don’t have to worry about. That, and high end machines tend to last longer - your T460p was absolutely high end when it was new. I doubt i.e. a Chromebook of the time would have lasted you as long.
It was about price - you can have something a little older for a good price. Also battery can be a point for some people, for me it’s not. I’ve lived through the ages for “lasts for 2h” and this one is not that that.
Also Chromebooks were to be had for like 300 - this was about if 900 is about a laptop that might be crappy.
For couch use I definitely wouldn’t want something other than a MacBook Air, tbh. Everything else seems to run hot and have a shorter battery life. Reasonable enough as semi-portable workstations, not at all good for casual use possibly away from an outlet.
That’s ok, I’m not telling you not to. But this discussion wasn’t based on the premise if anything that is not a mac is deemed adequate or not
*nod* “My laptop” is a hand-me-down Lenovo Ideapad from the mid 2010s running Kubuntu like my desktop (a 2023 Ryzen for faster
rustc) and, to be honest, the laptop I use most is the hand-me-down Thinkpad T410 where I took advantage of the availability of Windows XP drivers and Inexperience Patchers to make it a fairly convincing fake Windows 98SE.(And then, only a week ago, after pouring all that effort in, I remembered that my “stuff to get to” pile contains a hand-me-down Thinkpad T42 that actually has Windows 98SE drivers available for download and won’t make me dance around Intel HD video drivers which don’t want to remember “preserve aspect ratio” across changes to widescreen resolutions. Oh well… I had fun.)
Seriously, though. As long as you’re not using a web browser, playing modern games, or decoding H.265, it’s ridiculous how fully functional Windows XP is with WinAMP 2.95 and ffdshow for the media side of things. Hell, my “gaming rig” is a 2012 HP prebuilt with a 2009 Radeon, a SATA SSD, and Windows 7 (KVMed in to fix a bunch of Gaming+multi-monitor papercuts on the Ryzen) and it still does 99% of gaming related things I care about. As long as you’re not into AAA games, the slow-down in tech is real.
I don’t think a laptop can both be average and high end ;) But TIL that $2500 laptops exist, that seems nuts…
A fully loaded 16-inch MacBook Pro comes in at $7,500.
That shouldn’t be surprising, there are a lot of wealthy people in the US. Companies are going to do their best to get their hands on that money.
As far as Apple laptops go… Yeah.. $2500 isn’t even scratching the surface :P
I’ve been self-employed for a very long time, so the incentives are different, I guess. I don’t replace my laptop very often, but when I do, I don’t think about spending double that or more since it’s the way I provide for myself and any optimisation there is a no-brainer.
As for a personal laptop, I would tend to agree with you, my personal one is an 11-year-old 15” MacBook Pro, still running smoothly.
Yes and no. It’s about where the line for a laptop I’d expect to work well is, but there are a lot of cheaper models on the market. Dell’s web site has ones starting at £300 (including VAT, 6-core i3 processor, 15” screen, 8 GiB RAM. 256 GB SSD).
My experience with those is that they’re a false economy. The last Dell we bought was about £450 had the battery die after six months. The replacements were over £100. Dell support claimed that batteries were a ‘consumable’ not a replacement part (in spite of being listed under replacement parts on their web site) and so not covered by warranty. I was advised I could take them to the small claims court but I didn’t have the time nor energy, I just decided to never buy a Dell again and tell this story at every opportunity: it will probably cost them more in the long run.
In comparison, I’ve had several batteries die in Macs after a couple of years. Back when they were easily replaceable, Apple would courier replacements out to me and they arrived at 9am the day after I called their service line (as an individual, no corporate support). Now, I have to take machines into the Apple Store or have them collect the machine and then repair it and send it back, which would be more annoying if I still lived somewhere where the nearest Apple Store as a couple of hours train ride away.
Sometimes you get lucky with the cheap machines, sometimes you don’t. Some of the mid-range Lenovo models seem to last very well. The old IBM ThinkPads were tanks. I tripped over the charging cable of my R31 in the middle of a big compile job (hitting the disk hard) and kicked it across the room. Compiler paused for two seconds and it then continued as if nothing happened and the machine suffered no ill effects (MagSafe has saved me from doing the same with Macs a few times).
Some of the more expensive machines are also not worth it. Microsoft gave me a top-of-the-line Surface Book 2 (2018 model) when I was there. The touchscreen was nice. Being able to detach the screen was sometimes nice but the battery life detached was awful (it wasn’t great anyway) and they didn’t do the obvious thing of having the base work as a Bluetooth keyboard when detached, which would have made the tablet mode much more useful. Windows updates broke it periodically. Thermal dissipation was bad and video calls made it slow to a crawl (typing in Teams took seconds for characters to appear when a video call was running, and because async JavaScript is hard, they would often appear in reverse order, one second apart). A lot of this is because every sensible kernel has a rule that the only thing you do in ISRs is prod a lock or trigger a work queue, whereas Windows actually does a load of stuff in ISRs, which makes latency absolutely terrible, but some of it was hardware. It definitely wasn’t worth the £2500 price tag.
My last MacBook Pro was a late 2013 model. It cost around £2500 (almost all of the options turned up to 11), but lasted ten years of daily use (battery finally died after six or seven years, long out of warranty, and I replaced it myself with a fairly cheap replacement). It was much better value than the £450 Dell that lasted six months.
I’d expect a €950 laptop to be something decent that should last many years. Not the most powerful, but at least well built and fast enough for 90% of what I do.
I’ve also had a lot of luck with refurbished machines. You can often get a decent, modern machine with 30% discount just because it’s got a small scratch, or is a corporate return for some reason (you can tell by the higher windows tier and 3 year on-site support included). There’s some risk of course (returned due to some defect that the service didn’t notice/fix), but I’ve been happy doing this for years.
For personal laptops, I don’t see myself buying another one brand-new. I’m just an average power user, not an AI developer or gamer. Web browsing, coding, videos, and light retrogaming are the extent of my activities. I try to buy used or refurbished from a reputable dealer.
I was once within seconds of buying a Framework 16” laptop for something like $1,800. That would have been a hefty price, more than I’ve ever paid for a laptop, but I liked the product, the company, and what they stood for. Before I noticed that it wouldn’t even ship for 5 months at the earliest. I decided to look around and found a great deal on a used Dell Precision workstation laptop with around the same specs give or take a few GHz for $900. I sat on it for a while and went to check the price on the morning of Black Friday and found that they dropped the price by 50%, so I got it for well under $500. It’s not my favorite laptop I’ve ever owned but it does the job, especially for what I paid.
Yup, my partner got a second-hand late 2013 MacBook Pro about five years after I got mine new. Hers cost around £500 and was great. She replaced it with a new MacBook Air about a year after I replaced mine.
I had a laptop I used for years. The WiFi driver was broken under windows. So it was half off when I bought it refurbished. Machine worked great under Linux.
I’ve only ever had excellent experiences with Dell. I don’t have a policy of saying so at every opportunity, but I feel like this is as good an opportunity as any to share my experience.
I’ve had four Dell laptops over the past 15ish years. Three I bought, one was inherited. I believe I bought one replacement battery during that time, although I didn’t actually use them on battery power much. I have found the quality of the machines and the customer service absolutely top notch, and my experience has been so good that I’ll never bother looking at another manufacturer unless it’s a “speciality” manufacturer like Framework, unless I have some particular needs for some reason, or unless I have a bad experience with Dell (which may yet happen).
I totally agree with all of that. What I want doesn’t currently exist but I hope it will soon. It’s something with the build quality of a MacBook but which not only runs other operating systems but is guaranteed (within reason!) to always do so. The heroic nature of the work that the Asahi Linux people have to do is impressive and commendable sure but also worrying.
They’ve had to reverse engineer the GPU and some other things, but Apple put in a lot of effort to make it possible to boot other operating systems without impacting secure boot of macOS.
I haven’t used on personally but everything I’ve heard about the Framework laptops has been glowing.
Their cheapest model is more expensive than the MBA in the article. Their cheapest model upgrade potential might make it cheaper in the long run, but given that I replaced my last Mac because it was starting to physically wear out, I’m not 100% sure.
I used to build my own desktops and upgrade them, but then I wanted a new CPU. The new CPU needed a new motherboard. The new motherboard needed new RAM and a new graphics card. At the end, I kept the case and hard disk, and at that point I had almost enough spare parts left over for a complete new computer, so the savings from upgrading were fairly minimal.
Yes! In case I wasn’t clear, I didn’t mean that Apple failed to be helpful. Still, the end result is that we can be much more sure that a generic laptop will run arbitrary other operating systems than that a modern-architecture MacBook will. So I’m waiting for the best of both worlds :-)
There is a reason people I know call it dHell.
I agree, totally!
I also agree that Macbooks are worth the premium for such a nice polished device over competitors of similar and more expensive prices, and like you say, they can last quite a while.
But my point is still like.. €950? Its just not what I expect for a crappy piece of hardware like the author says in the intro. I wouldn’t expect a crappy macbook or a crappy non-macbook for this price. It’s even only slightly more than how much I’d expect to pay for a second hand macbook of these specs too.
Yup, after reading that line I was expecting it to be about a netbook successor or something. Cheap crappy laptops start at around €300 these days. Find one of those that’s good and you have made a great discovery. Find a thousand Euro laptop that’s good and, uh, yes, that’s expected. Spending a thousand Euros on a laptop and discovering that it’s crappy would be a story.
I’m French and I agree, a 950 € laptop would qualify as high end in my books. Not the highest end sure, but definitely not crappy.
A few years ago I got a Thinkpad T14 AMD for 1389 Euro and the build quality was pretty underwhelming. I opened it to add RAM, but it didn’t seem like something I should do a dozen of times. Also, the electronics looked like a patchwork. Luckily this was during COVID scarcity and I could sell it without much loss. I think there are better-built ThinkPads, but they were more expensive than that. So I wonder what these high-end 950 Euro laptops are.
I replaced it with a MacBook Air M1, which was the same ballpark with some upgrades and it was much better built, far better display, faster, and quiet (the T14 could get noisy under load).
I usually recoup a substantial part of the cost of a MacBook and use it to buy the next one. I used to upgrade every 1.5 years or so, but since getting my 14” M1 Pro I haven’t felt the urge in years.
When the point comes back to switch to a non-Apple laptop with Linux, I’d probably get a refurbished ThinkPad or something like that.
It’s reasonably priced for what you get, in fact there’s nothing to compare it to in the PC space.
The last non-Apple laptop I used was a ThinkPad X1 Carbon (Gen. 4 I think) where I ran Arch Linux on. The ThinkPad cost three times as much as a M1B Air, but suffered from CPU throttling, loud fan noise, sub 3.5h battery life, and various hardware issues that led to a complete freeze of the machine.
I bought an M1 MacBook Air in late 2020 and have no real reason to upgrade since then, battery still lasts for 8h, it’s dead silent because its fanless, and I can charge it with any random usb-c phone charger if I need to! I’ve been a die hard Linux user before switching to Apple, but those M-chip MacBooks are so good that they play in a completely different league and I can’t imagine going back to a Lenovo or Dell laptop.
I only buy refurbished laptops. +1000e laptop for 300e? Yes please.
Mine cost $3500. It’s a System76, not Apple. It’s amazing. I don’t know how I would ever be satisfied with the specs of something that’s less than a grand.
Yeah, I’ve never spent more than $500 USD on a laptop. My first laptop was one of the 15” IBM Thinkpad models, $150 (used) and it lasted 3 years before starting to freeze up when its motherboard flexed a bit from picking it up. After that got a $200 new one, Lenovo branded Atom netbook called an IdeaPad, about 11”, and got 3 more years out of that one - and kinda technically i still have it and it still works ok for some things, but a new job made me do video calls and it was just too slow; the calls would lag out.
Then I bought a $360 laptop, again Lenovo, its model number sticker came off, but it is not one of the think/pad things, just some number… and it worked for 9 years of daily on-battery use (the battery would last about 3-4 hours at a time). I did buy it more RAM and an SSD totaling about $140 more on it over the years, which made a big difference, so total of about 500. My kid stepped on the screen and snapped a plastic support so the left side of the laptop is floppy now… and the battery started getting only an hour use out of it anyway, so I bought another $400 one. (I still have that old one on my desk because it has so many files and programs installed that I still want periodically, but it hasn’t been a daily use laptop for almost two years now.)
The new one is a HP Probook, I got it slightly used from a going-out-of-business sale. Been very happy with it so far - it feels like HP’s clone of the Apple Macbook, 13” screen, fancy lightup keyboard, similar size, shape, color, even the keyboard seems similar to a macbook air. Get about 6 hours out of a battery charge and it goes to sleep / wakes up very quickly too. I don’t know what this model cost new, but I can’t complain at all. Had it for two years now, so we’ll see on longevity, but zero issues; it still feels new.
My policy was I’d get cheap laptops because odds are something would happen and they’d break anyway, so might as well make the pain of replacement less….. but the reality has exceeded my expectations. I like this HP a lot, the smaller size and weight compared to those 15” ones is lovely. Maybe in a decade (hopefully!) i’ll even fork out for one of these new.
This is a lot closer to my experience. Get something that was real expensive when a business bought it a few years ago when they start replacing that model in their fleet.
These are European prices, they really aren’t comparable to aus or usa for purchasing power. I would also be very very skeptical for anything under 1000 euro. Sure you can get something, but nothing in the “nice” category.
European here. €950 is a big sum of money; most people wouldn’t dream of spending that much on a laptop. You can (and should) expect something decent for that price.
I mean doesn’t this depend? I imagine people in Germany or Ireland have less of a problem paying 950€ for a laptop, but that number is unimaginable in Portugal or Greece. Purchasing power in the eurozone is heavily varied.
In Germany this would be 1mo of rent. Compared to say SF where it would be 0.2mo of rent. That is a big difference in purchasing power as well.
Prices for identical laptops tend to be more or less the same in the EU (or you’d see a lot more grey import). And €950 is a serious chunk of money even in the richest EU member states.
Really? I know a lot of non-tech people who buy the entry-level MacBook Air (or upgraded to 512GB or 1TB storage) in The Netherlands (given your username, I guess you might also be from NL?). It’s fairly cheap relative to what people spend on cars, etc. Also almost 40% of the population owns an iPhone, which is close to a MacBook Air price-wise.
iPhones, MacBooks, etc. are very common in the richest EU states. If you go to a library, you’ll see that even a lot of students use MacBooks.
My work laptop (HP Elitebook) costs 150% of that and is a less nice machine in many ways than my M1 Air. Battery ife especially is atrocious.
(I’m in Sweden)
Your work laptop was paid by an organization that gets to write off the expense, and is minimally invested in the user experience of the device :)
That’s not how it works. They didn’t have to pay VAT on it, but it still costs them money…
Yeah, I’m very happy I get a better product for less when I spend my own money.
My point is that this thread is full of statements like “there are better laptops out there”, but there are very few links actually supporting that statement. Please note that a used Thinkpad running Linux is not comparable to a NIB Macbook Air in this case. It has to be something a consumer can buy online or in a big box store.
Why? The author is apparently a ‘Senior DevOps engineer’, and the typical lobste.rs user is also not one who is primarily convenience-focused?
Also, the point of contention is not that there are better laptops out there, but that you should expect a €950 laptop to be crappy.
Yes, and I’ve always had better equipment and a better working environment at home, paid for with my own money, than any employer has provided me with.
Prices in Europe vary quite a lot, you can’t really speak on behalf of all of Europe.
And neither can the person I was responding to.
Two wrongs doesn’t make a right.
agreed, my daily driver cost refurbished 160 EUR (Thinkpad x250) and I consider it the best machine I ever had (including a m1), on par with a 13” macbook unibody late 2008 (snow)leopard.
I don’t care too much which end the machine is - high or low - as long it’s a means to my end.
The UK has 20% VAT, so you need to account for that.
He quotes “Today, 47% of U.S. women ages 25 to 34 have a bachelor’s degree, compared with 37% of men”, but fails to account that 80% of engineering degrees go to men.
But I think he misses something much bigger. The “DEI backlash” as he calls it, stems from a desire to defend and promote meritocracy, which is under heavy attack from DEI proponents.
I think it’s a shame to assume that people in tech “hate diversity”, when there’s a simple and reasonable explanation to people’s resistance DEI, which is also more generous.
Entirely incorrect. The DEI backlash was created by racists and sexists because they hate that tech is actually becoming more meritocratic. People are racist and sexist, so they will prefer white men even when they are not as good. DEI practices counter this tendency. But people don’t like the idea that they might’ve only been hired due to the having a racist or sexist bias, and thus lash out against those practices.
Two incompatible world views, neither falsifiable. Neither of which I want to read about on this site.
My view is definitely falsifiable. For example, if you demonstrate tech has become more meritocratic in the past decade. Or if someone pointed me to DEI manifestos that put meritocracy at the utmost importance (I can find plenty who write the opposite). Or if someone pointed me to a significant organization in tech that explicitly tries to only promote “white men”, or fight against the fact that most tech CEOs aren’t white. I’m not aware of any.
If you’re not interested this discussion, that’s fine. But when I see people write things that are incorrect, and just also happen to be pretty horrible, I feel the need to state what I consider to be the truth.
About that…
— https://criticallyconsciouscomputing.org/history#header-2
I don’t understand, what in this paragraph contradicts what I said?
What I think is great about software, is that anyone can write a program, even anonymously, and its quality can be measured with relative objectivity. Do you think that if women wrote Linux, or C++, or Python, Rust, Google, Reddit, or LLMs, etc., that the tech world would ignore them because of sexism? I find it hard to imagine.