1. 14

    The choice is between an Electron app or no macOS app.

    Really? Were there no other cross-platform GUI toolkits you could have used?

    1. 15

      I wrote and maintain an Electron application. If I weren’t using Electron, if the application would exist at all, it would be Linux-only. There’s a very simple reason why: we use interactive SVGs a lot throughout the application, because we need an image that resembles a keyboard, and SVGs make that easy. GTK can’t do that, QT can’t do that. They can display SVGs, sure, but they don’t support interactive SVGs, only via a webview. Since about 90% of the stuff the app does is tied to interacting with that SVG, there’s no point in using a native toolkit, when most of it will be spent in the same WebKit anyway.

      We could, of course, build our own widgets instead of using SVGs, but that’s way more time investment than we can afford. We - well, I - explored the native options. None of them would’ve allowed me to build the app as it is today. Electron did.

      Sometimes there are good reasons for using Electron.

      1. 1

        not svgs, but the tk vector canvas is surprisingly capable and nice to use.

      2. 12

        Honestly, I have a dim view of cross-platform toolkits having used them and developed on them. UX-wise, they’re about as bad mouthfeel-wise as Electron. I had to do quite a bit of work to get Qt to have not awful mouthfeel on Mac OS. I find most of the advocates of cross-platform UI toolkits tend to be on Linux, which was historically a pretty low bar UX-wise and clamouring for any application. You’d get better performance, but it’s not a strong argument from the UX side.

        1. 6

          Honestly, in my experience they’re worse than Electron apps. At least standard text input shortcuts work in Electron! The “cross-platform” toolkits tend to look non-native, just like an Electron app (although often more dated), and have very… weird shortcut support, e.g. macOS Emacs-style keyboard input.

        2. 7

          What would you recommend? I don’t mean this adversarially, it’s just that every time I’ve looked for a good cross-platform GUI toolkit, I’ve come back disappointed. I hate working with Qt because Qt bindings vary in quality across languages and I’d rather not use Qt’s stdlib over the C++ stdlib when writing C++ because I have much more experience with the C++ stdlib. Gtk similarly has some pretty poor bindings in certain languages. Tk is probably the toolkit I’ve enjoyed using the most, but it’s rapidly losing mindshare from what I can tell.

          1. 2

            I agree with you that the state of cross-platform GUI toolkits is bad. I love GTK on Linux, and as far as I can tell, its language bindings are consistently good, even when they’re third-party. But GTK support on Windows is second-class, and on MacOS has historically been terrible (but is maybe getting better?).

            When I was looking at a toolkit to use for a cross-platform graphical Common Lisp app, the best I could find was Tk, despite its limitations.

          2. 4

            I think so? There are options you can use, but there are no really good options. https://blog.royalsloth.eu/posts/sad-state-of-cross-platform-gui-frameworks/ is a nice survey.

            1. 2

              They missed some third-party XAML implementations like https://avaloniaui.net/ It’s going to be closer to javafx, but with a great RAD environment (visual studio)

              I hope MAUI will get a community Linux back-end. That would make it a good alternative too.

            2. 4

              i mean you have to buy into the react paradigm but react native can compile to windows and mac in addition to ios and android

              https://microsoft.github.io/react-native-windows/

              1. 2

                I guess people really haven’t tried how fast you can get stuff running with QT. Yeah it’s not completely native (and it can also be used the electron way since some version), but that’s not something you get with electron either. To be fair, you have to use c++ (with QT additions) or python for it..

                1. 2

                  I inherited a Qt project once. It was awful. I’ve never used Electron, but I know enough about it to pick it over Qt in most circumstances.

                  Not sure there are many other options if you’re targeting desktop. Proton looks promising. Microsoft’s React Native for Windows and Mac does as well. Both are similar concepts with ostensibly less overhead than Electron. Anyone here try those?

                1. 2

                  The earliest I could find is a bug report I sent to Debian, against CVS, back in late 2000:

                  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=76494

                  I’ve maintained websites prior to that, dating back to ~1997 or so, but none of those are available outside of archives. Most aren’t even in them, either.

                  1. 47

                    A good few years ago, I was working for a company that made a proxy with deep protocol analysis, as in, it looked inside the protocols, analysed them, and so on and so forth. Pretty exciting stuff. I was L3 customer support, we got the bugs that baffled L1 & L2, and required diving deep into the code. One such bug was from a customer who had trouble SSHing into some of their machines when using our software. It only happened when targeting certain machines, from certain clients, and wasn’t 100% reproducible. Below is my recollection of the bug hunt, years later, so some details might be slightly wrong, but I believe that it is mostly accurate nevertheless. Perhaps not in every detail, but close enough.

                    First, we narrowed down that the target system was always SunOS/sparc, and the client usually Linux. So we set up a test environment and tried to reproduce it locally, and that initially failed. After a bit of back and forth, we figured out that they were using a SunOS older than what we had in our test system, so we switched to the same, and lo and behold, bug reproduced! So whatever the bug was, it was likely fixed or worked around in later versions of the OS, which strongly suggested the bug wasn’t in our software. Nevertheless, the same SSH client could always connect directly, so there’s something we were doing differently, and that needed to be fixed, or worked around.

                    The symptom was that when connecting through our proxy, sometimes we got immediately disconnected. Looking at the proxy logs, it looked like the server side just dropped the connection. So I went looking server side, and it turns out, it dropped the connection, because the sshd crashed. Oopsie. Not the main sshd, mind you, but the process it forked off to handle the new connection.

                    Now, why does it crash, and how can we prevent it from happening? After a lot of hair pulling, and diffing the exchange between a good connection and a failed one, we narrowed it down to key exchange. In particular, to which key exchange is agreed upon during handshake. There was one which made the server crash. Nice, we can work with that! We patched our proxy to blacklist said KX, and as far as the customer was concerned, the problem was fixed.

                    I wanted to know more, though, so I tried to dig deeper: I compiled the same version of SSH, with debug symbols and whatnot - but I was unable to replicate the problem with that. Weird. No matter, I can do blackboxing! I enabled debug logging in the system sshd, and followed the source going by the logs, which led me to a block of hand-crafted assembly code. Okay, that’s scary territory, I don’t speak sparc assembly, but this is a good opportunity to learn some! There wasn’t anything particularly bad in that code, and newer SSH versions did not have a change there, either. So it wasn’t the asm. What else could it be?

                    The compiler, of course. There was a bug in the compiler that miscompiled that particular asm code. If using a different optimization level, or a different version of the compiler, the generated code was correct. But one particular version of the compiler did miscompile it, and if the agreed KX was this particular one, with a hand-coded asm part, we had a high, but not 100% chance of corrupting memory and crashing. Luckily, the miscompilation was pretty mild, as in, the generated binary differed in about 5 bytes, so we could even provide a binary patch for the customer. They opted not to use it, though, and just go with the workaround of blacklisting said KX.

                    Why didn’t it happen when connecting directly, though? Because the KX priority was different. You see, with our proxy, you can override the priority, and we shipped our own (more modern, at the time) defaults than the system ssh clients. When we asked the client to connect directly with the problematic KX, the problem became reproducible without the proxy too.

                    This took me quite some time to hunt down, find a workaround, and then dig deeper for the root cause. But it’s one of my favourite bug hunts of all time.

                    1. 4

                      I’m going to echo reddit and say I don’t entirely understand why anyone would prefer an ergodox over a kinesis. (Or a dactyl or maltron, though those are a bit more obscure and expensive, respectively.)

                      The kinesis is a bit less programmable out of the box, true, but that’s nothing you can’t replicate easily enough on the software side. And the hardware, which you can’t do anything about, is indisputably much better.


                      TFA makes reference to a keyboard with only four columns. I think that’s a great idea, which is why I was very excited about moonrim; it has four columns, with more rows to make up for it. Sadly, that project doesn’t seem to have panned out.

                      1. 5

                        There are quite a few things you can’t do with the Kinesis, like position the halves at a distance you find more comfortable, or tent it at a different angle. The ErgoDox can do both. Something like the Keyboardio Model01, even more so.

                        As for being programmable: there’s a whole lot of things that are just easier to do in firmware. While you can change the controller in the Kinesis, to make it possible to run QMK on it, that’s not something you do as a software guy.

                        1. 1

                          positioning

                          Sure. Personally, I haven’t found that particularly limiting, and the bowl shape more than makes up for it.

                          While you can change the controller in the Kinesis, to make it possible to run QMK on it, that’s not something you do as a software guy.

                          No need for that. You can do it all in x (or wayland, autohotkey, etc.).

                          1. 3

                            haven’t found that particularly limiting, and the bowl shape more than makes up for it.

                            The bowl shape might make it up for you, it certainly didn’t for me. Different people are built differently, and as such, what works for you, may not work for someone else. So, try to imagine that someone might not be happy with the Kinesis being a single piece board, and it becomes much easier to understand why someone would choose an ErgoDox/Dactyl/Model01/etc instead. :)

                            No need for that. You can do it all in x (or wayland, autohotkey, etc.).

                            There are things you really can’t, or only with much more effort than in firmware. Especially if you move your keyboard between OSes: doing the same thing 3+ times is much more effort than doing it once. Even more so when you’re otherwise unfamiliar with some of the OSes, and you just want to use your keyboard.

                            Layers, oneshot keys, tapdance - those are a few features I use all the time, which are trivial in firmware, but considerably harder to do on the host side if you want to do them well. One of the things I absolutely love about my Model01, is that when I use a OneShot modifier, I can light up the key until it is active. Not sure how you’d accomplish that without a programmable board. Or without LEDs, which the Kinesis doesn’t have either.

                        2. 3

                          The switch selection on the Kinesis is quite poor last I checked; none of the stock options were particularly pleasant. Plus you can take an Ergodox in a backpack; the Kinesis is too large to move. (Less of an issue during pandemic days.)

                          1. 1

                            This is just a useful reminder that everybody has a different decision procedure when it comes to keyboards. I use and endorse the use of the Kinesis Advantage, but then again – I’m happy with the switches, and I have two so I never need to carry one in a backpack.

                            I am very interested by the Dactyl, which is the only split keyboard I am aware of that uses the Kinesis negative camber.

                          2. 2

                            I’ve worn out four kinesis contoured, and I have two more in a box. The kinesis contoured firmware had some glitches where I can type faster than the keyboard can handle, and the bent circuit board wears out faster than the ergodox. I have four ergodox keyboards and only one is showing signs of wear. I had to repair my kinesis keyboards many times where the finger bowls connect to the center PCB. It’s much easier to modify the layout on ergodox keyboards as well. I could go into much more detail, such as my shoulders are too wide for the kinesis. I like the kinesis just fine, but ergodox are much better for me.

                            1. 2

                              Personal preference. Some people have quite broad shoulders so need the flexibility, custom layouts, hot swappable switches, some might even want open source firmware, etc. I know I own one for a few of those reasons. The layout might be replicable in software but that’s a lot more effort and requires doing it on every OS you use. Ergonomics are an incredibly individual thing and it’s never one size fits all. Hell, I have plans to build my own keyboard that’s as perfect for me as possible but the Ergodox does the job for now.

                              1. 2

                                And the hardware, which you can’t do anything about, is indisputably much better.

                                Can’t say I agree with that. I used a Kinesis Advantage for a while and I found the build quality to be lacking. It felt cheap and plasticky, and the function keys were all wobbly and rubbery.

                                I tried an Ergodox EZ for a while too, and found it nicer in terms of hardware.

                                (They were both too big for my hands, making me stretch in uncomfortable ways, and the Keyboardio Model 01 was the keyboard that finally made it possible for me to type without pain again.)

                              1. 2

                                I would find it a little unsettling to have such a powerful MCU in my keyboard. It’s part of what’s kept me out of the custom mechanical keyboard scene so far. I’d like my input devices to be as dumb as possible, at least to the point of being incapable of being used as a keylogger.

                                1. 6

                                  Most keyboards likely have absurdly overkill microcontrollers on them anyway; I’d rather control it rather than someone else.

                                  1. 2

                                    Bluetooth is another can of worms altogether, but there’s no economic incentive to put such a costly part on a cheap USB keyboard that’s not programmable in the first place. So, I doubt it.

                                    On the other hand, it’s much easier to develop flexible firmware for a variety of physical configurations with an overpowered general-purpose MCU and its toolchain. And the economics of open source make DIY experimentation and low-volume manufacture much easier. So I understand why QMK exists… but if I ever make my own keyboard, it won’t have fancy LEDs or USB-C, and it damn sure won’t have RAM. (shaking fist at cloud)

                                    1. 5

                                      if I ever make my own keyboard, it won’t have fancy LEDs or USB-C, and it damn sure won’t have RAM

                                      Out of curiosity, have you ever tried to code something in two and a half kilobytes of memory? Because that’s what the ATmega32u4 has, which is the most popular chip QMK is used on. QMK also works with the ATtiny85, which has five hundred and twelve bytes of RAM. (Fun fact: that’s less than the number of bytes in this comment.)

                                      I wrote a Forth implementation for the atmega32u4 chip, because I heard that Forth is famous for working well in low-resource situations: https://github.com/technomancy/orestes Once the Forth booted, I got it loading my program in, and it got about twenty lines of code loaded before it exhausted the RAM.

                                      I’m not saying it’s impossible to do anything nefarious, but at that point there are much easier ways to inject a keylogger. If I were you I’d draw the line at networking instead of RAM.

                                      1. 3

                                        Yeah, the no-RAM thing is kind of an unreasonable restriction for a programmable device. I regret my little outburst. As I say elsewhere, all I really want is for the programming interface to be physically separated from the keycode interface.

                                        It’s quite impressive that QMK would work with such a modest 8-bit MCU. The ATtiny looks like a pretty good candidate for my low-end aesthetic. Thanks for the tip!

                                        EDIT: Wait, can you give me a link to QMK on ATtiny85? It’s looking less plausible since the docs say STM32 and ARM exclusively, and then Reddit says No. Bit-banging USB doesn’t seem like a great idea anyway.

                                        Now I’m genuinely curious what is the lowest-spec MCU that can both speak USB HID (seems like the harder part) and scan a key matrix (which looks pretty easy, given enough pins). I don’t need any of the fancy QMK features, and I’d enjoy the size-coding challenge. No need for a Forth runtime, ASM would be fine for an application like this.

                                        1. 3

                                          Wait, can you give me a link to QMK on ATtiny85?

                                          https://github.com/qmk/qmk_firmware/pull/9371

                                      2. 1

                                        Bluetooth is another can of worms altogether

                                        Bluetooth implies latency, so I wouldn’t go anywhere near it. Even ignoring it’s wireless and I don’t trust it to be secure.

                                        it’s much easier to develop flexible firmware for a variety of physical configurations with an overpowered general-purpose MCU and its toolchain.

                                        Microcontrollers that are common today are, even at the low end, of course going to be overpowered for the purpose.

                                        1. 2

                                          Even the something like the MSP430 is overpowered… but that’s probably what I’d use, because it’s cheap, very low power, with a reasonable open source toolchain. I’m sure it’s possible to go lower than that, but USB HID isn’t something I would want to do from scratch. I suppose a PS/2 keyboard would be simpler.

                                          Really all my paranoia requires is that the device isn’t programmable over the same physical interface that it sends keycodes. Beyond that it’s just aesthetics… and of course latency.

                                          1. 1

                                            msp430 overpowered

                                            Indeed it’d be. msp430 is a nice 16bit µarch with very power efficient implementations, but avr8 is 8bit (!) and sufficient. I imagine older keyboards have z80, 8085 and the like in them.

                                            Really all my paranoia requires is that the device isn’t programmable over the same physical interface that it sends keycodes.

                                            There’s indeed a USB firmware update standard, which would be nice to support, as long as flashing needed some switch turned on, or some special handshake like holding some keys while plugging keyboard in, to force it into some flashing mode in the bootloader.

                                    2. 4

                                      Pretty much any keyboard that can connect via USB has enough capacity to have a keylogger in the firmware. For some values of keylogger, at least… a tiny thing that sends the pressed keys over on rawhid needs very little resources, and the host-side of it can appear pretty harmless too. If your keyboard supports being configured via a GUI on the fly, then whatever protocol is used for communication, can be used for keylogging as well.

                                      The best way to avoid a keylogger, is to run open source firmware, and lock your MCU down so that flashing new firmware will require an action from you (ie, you must not be able to jump to the bootloader from within the firmware).

                                    1. 5

                                      My daily driver is a Keyboardio Model01, the best keyboard I ever owned. The camera mount is amazing for tenting, and the palm keys are incredible. The custom keycaps - while they do limit your options - are a joy to type on.

                                      I also have a number of other keyboards for different purposes. For gaming, I use a Dygma Raise and an Azeron Classic. I usually use the Azeron (though I still need to port my firmware of choice to it, it’s default one is meh), but when I need more keys, I fall back to the Raise.

                                      (I also have a few ErgoDox EZs, an original Atreus, a Keyboardio Atreus, a Planck EZ, Splitography, Ginni, and a KBD4x, off the top of my head. The Planck EZ is my wife’s, Splitography is for Steno, the rest are currently unused.)

                                      1. 2

                                        Thank you, Algernon, for all the work you’ve put in on the software for the Model01 and Dygma Raise.

                                        As for OPs question, on my own keyboard journey, after years of using stock keyboards, then an entry level keyboard with cherry mx browns (to try mechanical switches), then a Mistel Barocco (to experience split layout — which I now can’t live without). I was given a Keyboardio Model01, but had a hard time with the ortholinear key layout and I didn’t have the time to relearn. I was going to get a Kinesis Edge Gaming, but preordered the Dygma Raise instead because of its full programmability.

                                        The Dygma Raise checks all the boxes for me:

                                        • mechanical switches
                                        • split
                                        • fully programmable (Sticky Keys!)
                                        • normal key layout (nothing to relearn)

                                        Plus it has:

                                        • thumb buttons
                                        • built in wrist rests
                                        • so many LEDs (useful to know which later you’re on)
                                        • can be used with only one side plugged in (for gaming mostly)
                                        • a WIP tenting solution

                                        I recommend checking out the Dygma Raise (even if you don’t play games, I bought it primarily for programming).

                                        And if not the Raise, in general, I very strongly recommend a split keyboard… your wrists and shoulders will thank you.

                                      1. 2

                                        Now I need to find someone who can build one of these for me… so I can port Kaleidoscope to it. For science!

                                        1. 2
                                          • A bunch of static sites
                                          • Two writefreely instances
                                          • One NodeBB forum
                                          • One single-user Wordpress
                                          • Nextcloud (for file sharing and bookmarks, primarily, but trying to make better use of it still)
                                          • Bitwarden
                                          • Gitea
                                          • Drone for my CI needs
                                          • My own email (postfix, rspamd, dovecot, roundcube to top it off)
                                          • DNS for some of my domains
                                          • XMPP for the family (Prosody)
                                          • Mastodon for myself & a few friends
                                          • ZNC
                                          1. 3

                                            A language is viable for production as soon as someone is crazy enough to put it there, and commit to keeping it there.

                                            A lot depends on where “production” is. Is it something you put on your home server? An important piece of infrastructure at home, but if anything breaks or goes down, nothing really serious happens. It’s still “production”, though. Putting something on thousands of on-premise devices that are never going to be updated is another end of the production spectrum, with very different requirements.

                                            Mind you, I’ve put immature languages onto both, with great success.

                                            1. 3

                                              I use split ergonomic keyboards both at work and at home: https://github.com/omkbd/ErgoDash The repo contains all the files one needs to have the PCB and case plates fabricated (not my work).

                                              I had the PCBs made at jlc pcb and soldered it all together at home, see: https://images.yourfate.org/#15521316686517 It runs QMK firmware: https://qmk.fm/

                                              While it has more keys than a 40% I still use a lot of layered keys. My favorite layer feature is having a “numpad” on the righ thalf of the board, with 456 being on jkl when I press a mod key on the other half. Makes entering many numbers very fast.

                                              AMA I guess…

                                              1. 3

                                                I feel like it would be difficult to type on this if the two parts weren’t always the same distance apart and at the same angle of rotation. Do you find that you’re constantly making micro-adjustments to get the two parts into your preferred position?

                                                1. 3

                                                  No, I actually sometimes move them apart to have stuff like documents, a notepad, or food between them. I can use them In lots of distances, as long as the angles of my wrists are right, which you can adjust on the fly.

                                                2. 2

                                                  What do you think of the QMK firmware? Did you use a ready made solution before this?

                                                  1. 3

                                                    I like that in QMK I can easily remap keys and create new functions on layers. They have an online configurator where you can edit the layout to your liking and get a new firmware binary: https://config.qmk.fm/#/ergodash/rev1/LAYOUT

                                                    You can also edit the layout locally and compile it yourself, some advanced functions are not available in the configurator.

                                                    I have used many ready made keyboards before, but liked the idea of the ergodash. I’m very happy with these keyboards so far.

                                                  2. 2

                                                    Can you report on how painful it is to have the {/} and [/] keys split between the left half and the right half?

                                                    I experimented with similar designs here, but I was always concerned about this issue.

                                                    1. 2

                                                      I have them somewhere else. I have them on a layer on P and the button to the right of it. I press the layer button on the left half, then P and the button to the right of it are [].

                                                      This works nicely for me. In general I have my layout set up so that it’s always the opposite hand pressing the layer button (i.e. left hand switching layers for the right hand and vice versa).

                                                      If you’re interested, my config.json is here: https://gitlab.com/youRFate/keymaps/blob/master/ergodash/layers.json

                                                      You can dl it and plug it into https://config.qmk.fm/ to see the layers etc.

                                                      1. 2

                                                        I have ([{/}]) on separate hands, it needs a little getting used to, but once that happens, it’s very, very nice to have them that way.

                                                        It’s only a problem when one of your hands is mousing, but for that, I have a trackball between the halves, so very little hand movement is required between keyboard and ball, and thus, no real issue with hitting either part of the pair.

                                                    1. 2

                                                      I’ve had very nice discussions on the Fediverse, about all kinds of things (unlike on twitter). I’m also on a tiny, personal instance, so the local timeline is nigh useless, because its me, my bots, and a friend of mine.

                                                      Nevertheless, I’m @algernon@trunk.mad-scientist.club, and I post about mechanical keyboards, their firmware, parenting, self-hosting things, and whatever else I feel like. There’s no real theme or purpose, it’s a loudspeaker I shout random things into, and sometimes people shout back and we have a great time.

                                                      1. 1

                                                        I think that’s how I discovered the Dygma Raise project, which you’re kinda involved with. Hopefully they’ll manage to ship keyboards this quarter… 😁

                                                        Mastodon’s nice for that. Of course, it could happen as well on Twitter, or Facebook, but I feel that I’ve discovered more things than on big advertisement silos, where the platform aims at maximizing buzz rather than serendipity.

                                                        I’m oz@mastodon.social BTW.

                                                      1. 1

                                                        Some thoughts:

                                                        Git is also incredibly opinionated about how you should work with e-mail: one mail per commit, nicely threaded, patches inline.

                                                        This is because it was made for the kernel project. You can easily modify this behavior with scripting.

                                                        Collaboration includes receiving feedback, incorporating it, change, iteration

                                                        With e-mail this is well, via e-mail, who anyone can set up as they want (and likely already have)

                                                        I wasn’t subscribed at the time, and the mailing list software silently dropped every mail

                                                        This is incredibly bad. I really think mailing list software and mail servers are to blame: if you aren’t subscribed, you should get a confirmation of success or failure. If you are classified as spam, the mail should not be delivered and thus get a notice back.

                                                        Having to subscribe to a list to meaningfully contribute is also a big barrier: not everyone’s versed in efficiently handling higher volumes of e-mail (nor do people need to be).

                                                        Again I think this is the mailing list software’s fault. Most are ancient and are not designed with this workflow in mind. Ideally you should have subscription options (only get updates to your threads, or threads you are active in, etc) with sane defaults.

                                                        In all of these cases, I had no control. I didn’t set the mailing lists up, I didn’t configure their SMTP servers. I did everything by the book, and yet…

                                                        Not that you are wrong at all, but I feel the need to add that it is also true about forges!

                                                        Patches never arrive out of order and with delays

                                                        If patches are prefixed with [PATCH v1 0/7] the mailing list should queue them up to guarantee in order and without delays between them delivery.

                                                        have easy access to all discussions, all the commits, all the trees, from the comfort of your IDE

                                                        I agree this is better than looking through your list archives or online ones. But there could be a tooling solution for that.

                                                        No need to care about SMTP, formatting patches, switching between applications and all that nonsense.

                                                        Forges make the boring, tedious things invisible.

                                                        Tooling can be distributed within the repository if the project chooses this workflow.

                                                        I think all in all the problems you saliently describe are the fault of (1) lack of appropriate software to assist the workflow (2) lack of effort to support the workflow from the maintainer’s side. Ideally I would like a forge-mailing list that supports both workflows with the features I suggested above because not everyone is a power user but most people eventually turn into one.

                                                        1. 4

                                                          This is because it was made for the kernel project. You can easily modify this behavior with scripting.

                                                          Tell that to a newbie, and see them run away never to be seen again.

                                                          If you are classified as spam, the mail should not be delivered and thus get a notice back.

                                                          Alas, that’s now how many systems work in practice. It would be nice if they did, but they don’t, and I have zero control over that.

                                                          Not that you are wrong at all, but I feel the need to add that it is also true about forges!

                                                          The difference is that I have not come across a forge yet that let me submit invalid PRs or issues, sat on it for hours, and (at best) mailed me a response much later. I have had e-mail issues countless times on the other hand.

                                                          If patches are prefixed with [PATCH v1 0/7] the mailing list should queue them up to guarantee in order and without delays between them delivery.

                                                          No list does that currently, as far as I’m aware. Probably because it would be so very easy to abuse it: make sure some of the series is spammy, so it gets held up, or even rejected, and never send a good replacement. The mailing list keeps the rest queued, wasting resources. You can, of course, combat this by expiring stuff after a while, or building various abuse prevention methods into the mailing list software, but then we’re building workarounds again, instead of using purpose-built tools which don’t even have this problem.

                                                          Sending patches attached would solve 90% of the problems listed in the blog post, yet, core git doesn’t support that.

                                                          But there could be a tooling solution for that.

                                                          Could be, but there isn’t, even though e-mail and the desire to use it for All Kinds Of Stuff has existed for far longer than the Forges. Yet, noone built anything like the forges over e-mail, until Sourcehut, very recently.

                                                          Tooling can be distributed within the repository if the project chooses this workflow.

                                                          Uhm, yeah, no. You’ll never be able to support IDE integration at the level magit/forge (and many other integrations do) that way. You may be able to support an IDE or two. With a Forge that has an API, IDEs can support the Forge, and have it work for all projects using that forge. Resources far better spent.

                                                          1. 1

                                                            Uhm, yeah, no. You’ll never be able to support IDE integration at the level magit/forge (and many other integrations do) that way.

                                                            Yes I know, I was talking about e-mail. :)

                                                            Your other points can be summed up as “no software that does this exist” and this was exactly my point: it’s basically lack of appropriate & user-friendly tooling that makes e-mail version management so difficult for unaccustomed users. This is a problem the community should solve. I’m not sure if the git maintainers would accept such patches, but it’s worth trying.

                                                          2. 3

                                                            What you are describing is not a bad thing, but you’re not really doing email at that point. You’re designing a Code Forge Peering Protocol, and you just happen to have built it on top of SMTP.

                                                            I’ve been horribly tempted to do just that, by taking git-am/sr.ht’s email behavior and designing a forge that speaks the same protocol but bridges it with a pull request based web UI. You could fork a repository from any git:// URL, submit a “pull request” by sending email to whichever email list the project uses, and we would provide an email address for you that’s only used for code review (and, thus, only accepts replies, not unsolicited messages, because we don’t want to do anti-spam heuristics if we can avoid it). All of this could be done entirely through your browser, though you could clone your fork and do local work exactly like GitHub does. The only part where I’d have to do even remotely novel work would be in providing a good UI for git rebase -i, so that you could appropriately convert your in-progress commits into a good patchset that the ever-picky project owner would accept, again without forcing you to touch the git CLI.

                                                            And since we directly know that all incoming unsolicited emails must be patchsets, we could both do the queueing work that you describe to ensure that patchset submission is atomic, and we could reject 99% of incoming spam instantly.

                                                            Alas, not nearly enough hours in the day to build everything that I’m horribly tempted to build.

                                                            1. 1

                                                              providing a good UI for git rebase -i, so that you could appropriately convert your in-progress commits into a good patchset that the ever-picky project owner would accept, again without forcing you to touch the git CLI.

                                                              Wait, do existing web-based git management things (git{lab,hub}, sr.ht, etc?) provide a way to interactively rebase? I think having everything else you proposed would be amazing.

                                                              1. 1

                                                                Hm I don’t think that’s possible on the web, since you have no way to resolve conflicts unless you startup a web editor. However you can do it with git “clients”, ie wrappers around the command line.

                                                                1. 1

                                                                  Yea I didn’t think so, which is why I was a bit confused to see GP basically say “I could implement everything but I cannot implement this one thing that no one else has so I won’t implement any of it” (my rough paraphrasing)

                                                              2. 1

                                                                You’re designing a Code Forge Peering Protocol

                                                                Yeah, just like mailing lists are a discussion forum on top of SMTP.

                                                                If you ever want to start this thing together send me a message :).

                                                                (By the way, I’ve a tool like your ammonia project. I use it as a filter for html view in a cli MUA i’m working on.)

                                                            1. 8

                                                              Find the message ID to reply to. [switch to a browser, navigate to the archives page, find the thread in question, expand the details section, select the ID, copy it, and paste it into your terminal]

                                                              It seems like the whole point of sourcehut is to allow for a collaborative flow with git which lets you stick with your comfortable tools instead of having everything center around a web application. This section (apart from being very tedious) seems to imply the opposite.

                                                              Meanwhile:

                                                              Some people think that they can get away with sending patches through some means other than git send-email, but you can’t.

                                                              You know what mail clients are really good at? Saving you the hassle of having to track down message IDs manually like some kind of cave-person! Instead of stating that it’s completely impossible to use a real mail client to send a patch (which is obviously not true) why not explain why most mail clients are bad at it?

                                                              1. 8

                                                                The web interface is my comfortable tool. I had a patch for Sourcehut and I literally just gave up on submitting it because of how big a PITA it was to submit. Email is terrible, I really don’t want a workflow based on email.

                                                                1. 8

                                                                  I literally just gave up on submitting it because of how big a PITA it was to submit

                                                                  That’s probably why SirCmpwn made a tutorial. ;P

                                                                  I personally agree with you about GUIs. I think terminals are garbage for any kind of low cardinality discovery-based workflow. If I know where I’m going and what I’m doing, terminal. If I’m looking through a few different things, or I’m not quite sure what I’m looking for, GUI. If I’m looking for / operating on a gajillion things, back to the terminal.

                                                                  By most metrics I’m a terminal power user, but “fear of mice” is not one of them. If you’re fast and accurate with a mouse, no reason not to use one.

                                                                  1. 5

                                                                    That’s probably why SirCmpwn made a tutorial. ;P

                                                                    A tutorial is nice, but what I really dislike is that I’ll do the configuration once then not touch it for months (and forget how it works) until one day it breaks as I’m trying to use it. It’s the same reason I don’t run my own web or email server: I really, really don’t want to increase my “sysadmin” burden.

                                                                    That being said, kudos to him for making it easier for people who were more “on the fence” than myself. But personally I’ll wait for the web interface :-)

                                                                    1. 4

                                                                      “By most metrics I’m a terminal power user, but “fear of mice” is not one of them. If you’re fast and accurate with a mouse, no reason not to use one.”

                                                                      Now that’s a great philosophy for UI’s. :)

                                                                  2. 1

                                                                    This section (apart from being very tedious) seems to imply the opposite.

                                                                    It doesn’t imply the opposite, it’s basic awareness by the author. Those instructions are universal and easy, all kinds of readers will follow along just fine. If you care deeply about avoiding a web interface, you probably know how to find the Message-ID header in your mail client on your own. If you don’t, figure it out on your own. It’s not important to this tutorial.

                                                                    You know what mail clients are really good at? Saving you the hassle of having to track down message IDs manually like some kind of cave-person!

                                                                    What’s your point? Does your mail client automatically populate replies to threads with your updated commits? No. So why complain about git send-email not automatically populating Message-ID? You’re copying something somewhere either way. Between the two, copying Message-ID is a lot less error prone.

                                                                    Instead of stating that it’s completely impossible to use a real mail client to send a patch (which is obviously not true) why not explain why most mail clients are bad at it?

                                                                    Okay again, this is a tutorial. And the implication here is obvious too. If you are reading a beginner tutorial about mailing patches, you clearly don’t know anything so just use the damn tool that stops you from mailing malformed garbage to a ton of people. It’s way out of scope and anyone that cares can easily go read more elsewhere.

                                                                    For anyone who actually does want to know, instead of just bitch: a mailed git patch isn’t just a diff, it’s an entire commit serialized into an email. That serialization is extremely fragile since leading whitespace and single newlines have meaning for patches, but for email they don’t.

                                                                    When you say “most” mail clients are bad at it, you’re probably thinking of Gmail, Apple Mail, Outlook, and so on. And yes, it is literally impossible to make the Gmail web interface handle patches correctly. But you probably didn’t guess that neither Alpine nor Mutt handle patch emails correctly without multiple configuration changes. All mail clients screw it up.

                                                                    All. Of. Them.

                                                                    Use git send-email.

                                                                    1. 1

                                                                      All mail clients screw it up.

                                                                      All. Of. Them.

                                                                      Had no problems generating a patch with git am, copying it with xclip and pasting it into Thunderbird. As long as you make sure word-wrap is off everything is fine.

                                                                      1. 5

                                                                        I don’t think this is actually true; I imagine that even tho it worked in some cases it doesn’t guarantee it will work in others, and that further configuration is required for it to work consistently. (For instance, for some patches it’s important that the MUA preserve trailing whitespace.)

                                                                        But that just illustrates my point: if you tell people something that’s obviously false like “you cannot send patches with a MUA” without giving any details, they’re going to ignore you.

                                                                        I’m a big kid. I can configure my MUA.

                                                                        1. 1

                                                                          Exactly, it’s about the edge cases.

                                                                          I’m a big kid. I can configure my MUA.

                                                                          The point is you shouldn’t until you have a meaningful understanding of the issues, and some actual experience with emailed patches. Because big kid or no, there’s still a nontrivial chance you’ll screw it up. There’s no edit button on an email blasted out to everyone on a mailing list, it’s a hassle for everyone else.

                                                                          1. 3

                                                                            I think that’s a reasonable stance; what I disagree with is the wording on the site and on sr.ht proper.

                                                                            It’s like the author has never interacted with children. If you tell a kid “don’t touch the stove” they’re going to touch it. You tell the kid “if you touch the stove, you will get burned because it’s hot” and then they’ll actually listen. Hackers are a lot like children in that way.

                                                                            1. 4

                                                                              I apologize for being so harsh in my earlier comments. My frustration stems from experience with people who don’t actually listen. People who insist on using their favorite workflow at the expense of everyone else, and have the entitled audacity to complain when other people won’t go out of their way to support them. When you replied in that way, it reminded me of that kind of person. But I’m happy to explain things to people who ask nicely.

                                                                              I’m extremely biased in favor of people who generously give up their time to write documentation and tutorials. I don’t think they deserve criticism for deciding something is out of scope, it’s much better than writing nothing at all. They already deal with so much crap from selfish needy people, they’ve earned some slack in my book.

                                                                            2. 0

                                                                              I use Thunderbird for all my email accounts and to keep track of my RSS feeds. I have filters set up to automatically categorize mailing lists. I can send and view mail in plain-text, or properly signed and/or encrypted with pgp. Assuming the world of mail clients consists solely of gmail/apple mail/outlook and telling people never to send mail outside of git send-email is bogus. I will continue to use my fully-featured, properly-configured MUA to send patches, tyvm.

                                                                              1. 2

                                                                                Cool. I recommend that you continue to use Thunderbird for all your email accounts, RSS feeds, filters, and categorization, exactly as you are now. And copy the Message-ID into git send-email. It breaks literally nothing in your workflow.

                                                                                When you copied git format-patch output into Thunderbird, did you copy the entire git format-patch output into the message body? If you did, you did it wrong. Or did you correctly delete the first line, part of the mbox format, and copy the subsequent header lines out of the message body and into Thunderbird? If you sent multiple commits, did you ensure each commit was sent as its own email, each email subject properly used [PATCH n/m] format, and patches n>0 were all In-Reply-To patch 0?

                                                                                Did you perform all of these manual steps perfectly without fail, and will you perform them perfectly without fail every time in the future? Did you find all of that easier than copying the Message-ID to git send-email?

                                                                                And when you verified your patch sent properly, did you test by sending the patch to a different email address via a forwarding address a la mailing lists, copy the entire raw message source received into a file (not the message body), git am that file to a different repo clone with different name and email configured, and verify the authorship, timestamps, and all other commit metadata was perfectly preserved?

                                                                                If not, your test was worthless. And if you did, can you prove it works for all variety of edge case patches, like patches with extremely long lines, non-ASCII characters, patches received from someone else and modified by you, and so on and so forth?

                                                                                Assuming the world of mail clients consists solely of gmail/apple mail/outlook

                                                                                Yes, that was clearly my assumption when I also specifically addressed Alpine and Mutt in the same paragraph. Good catch.

                                                                          2. 1

                                                                            Since git am doesn’t generate patches, I’m skeptical.

                                                                            1. 1

                                                                              Sorry, I meant git format-patch

                                                                          3. 0

                                                                            If you are reading a beginner tutorial about mailing patches, you clearly don’t know anything so just use the damn tool that stops you from mailing malformed garbage to a ton of people.

                                                                            Or you’re used to sending patches via PRs or whatever other method, and come across a project that wants patches via email. You might be a git wizard otherwise, but if you haven’t used git send-email, a beginner tutorial can be helpful.

                                                                            Does your mail client automatically populate replies to threads with your updated commits?

                                                                            Yes. It did need a bit of configuration, but then, git send-email requires a tutorial. Thus, if an email client requires configuration to do things correctly, I’d still consider that a viable thing to do.

                                                                            And yes, it is literally impossible to make the Gmail web interface handle patches correctly.

                                                                            You can attach patches, and it won’t mungle them. It won’t be inline, but any reasonable e-mail client will be able to show you the patches in a reasonable manner whether attached or inlined. It’s only impossible if you aim at the lowest common denominator: terrible email clients. (Not saying anyone should use Gmail’s web interface for sending patches, but it is possible to do that.)

                                                                            1. 0

                                                                              You can attach patches

                                                                              No you can’t, those emails aren’t valid patch emails.

                                                                              but any reasonable e-mail client

                                                                              Literally irrelevant. Does not matter at all. This is about tooling and conventions.

                                                                              The tutorial literally says “some people think that they can get away with sending patches through some means other than git send-email,” that’s you.

                                                                              But you can’t. You’re flat out wrong and that’s exactly why the tutorial tells you you’re flat out wrong.

                                                                              1. 1

                                                                                No you can’t, those emails aren’t valid patch emails.

                                                                                But they are. That you don’t like them, or can’t work with them, that’s your and your tooling’s problem. There’s tooling that works just fine with attached patches. They may not be acceptable for git am, but git am is not the only tool. It is if you stick to core git, but if you do that, you’re limiting yourself unnecessarily. Mind you, you can always filter the mails through a filter from your MUA first, and pass the attachment as inline patches to git aim, it’s not even hard.

                                                                                Besides, like I mentioned in my previous comment, you literally can send valid, inline patch emails, with correct threading without git send-email. Your e-mail client may need a bit of configuration, but so does git send-email.

                                                                                This is about tooling and conventions.

                                                                                You may be surprised, but there are communities where the convention is sending patches attached, because those are far harder to mungle, and you don’t need to send a gazillion mails for a longer branch, just one larger mail. This makes it easier to review and reply to multiple commits at the same time, which can be very useful when they’re closely related (but still separate enough to warrant being separate commits).

                                                                                Inline patches are not the only way to email patches around. It’s the only way stock git supports, and as we can see from this discussion, also the most fragile one.

                                                                                1. 1

                                                                                  They may not be acceptable for git am

                                                                                  This tutorial and discussion is about using the core git email workflow.

                                                                                  It’s the only way stock git supports

                                                                                  This tutorial and discussion is about using the core git email workflow.

                                                                                  you can always filter the mails through a filter from your MUA first, and pass the attachment as inline patches to git aim, it’s not even hard

                                                                                  Alright, send a patch to the LKML, and when they reject it be sure to tell them “it’s not even hard,” I’m sure that will clear everything up!

                                                                                  You may be surprised

                                                                                  Shocker, I’m not.

                                                                                  there are communities where the convention is sending patches attached

                                                                                  Communities irrelevant to this tutorial and discussion about the core git email workflow.

                                                                                  To be clear, this tutorial and discussion are about the core git email workflow. See the very first sentence of the post, a website called git-send-email.io:

                                                                                  Git ships with built-in tools for collaborating over email.

                                                                                  1. 1

                                                                                    This tutorial and discussion is about using the core git email workflow.

                                                                                    git does not include a MUA, or any other tool to fetch email, which you will still need for the workflow, and the tutorial hints at this too, with “Check your email”. It already goes beyond core git email workflow. I wasn’t criticising the tutorial either, I was criticising your comments related to it, namely to give up on one’s usual MUA in favour of git send-email, just because without configuration they screw things up.

                                                                                    Alright, send a patch to the LKML, and when they reject it be sure to tell them “it’s not even hard,” I’m sure that will clear everything up!

                                                                                    I did, way back when, it was accepted.I sent plenty of patches by email over the years, none of them were rejected on the grounds of bad formatting/threading/whatever. None of them were sent using git send-email.

                                                                                    Git ships with built-in tools for collaborating over email.

                                                                                    No, it does not. It ships with tools to send email, that’s only one part of collaboration. Besides, what I was replying to is this assertion:

                                                                                    All mail clients screw it up. All. Of. Them.

                                                                                    Which is incorrect.

                                                                                    1. 1

                                                                                      Fine. All mail clients in popular or semi-popular use, in default configuration, or configured by those who don’t know what they’re doing, will not get it right 100% of the time. Happy?

                                                                                      1. 1

                                                                                        Yes, thank you.

                                                                        1. 15

                                                                          After the recent announcement of the F5 purchase of NGINX we decided to move back to Lighttpd.

                                                                          Would be interesting to know why instead of just a blog post which is basically an annotated lighthttpd configuration.

                                                                          1. 6

                                                                            If history has taught us anything, the timeline will go a little something like this. New cool features will only be available in the commercial version, because $$. The license will change, because $$. Dead project.

                                                                            And it’s indeed an annotated lighttpd configuration as this roughly a replication of the nginx config we were using and… the documentation of lighttpd isn’t that great. :/

                                                                            1. 9

                                                                              The lighttpd documentation sucks. Or at least it did three years ago when https://raymii.org ran on it. Nginx is better, but still missing comprehensive examples. Apache is best, on the documentation font.

                                                                              I wouldn’t move my entire site to another webserver anytime soon (it runs nginx) but for new deployments I regularly just use Apache. With 2.4 being much much faster and just doing everything you want, it being open source and not bound to a corporation helps.

                                                                              1. 1

                                                                                Whatever works for you. We used to run our all websites on lighttpd, before the project stalled. So seemed a good idea to move back, before nginx frustration kicked in. :)

                                                                                1. 3

                                                                                  Im a bit confused. You’re worried about Nginx development stalling or going dead in the future. So, you switched to one that’s already stalled in the past? Seems like the same problem.

                                                                                  Also, I thought Nginx was open source. If it is, people wanting to improve it can contribute to and/or fork it. If not, the problem wouldn’t be the company.

                                                                                  1. 2

                                                                                    The project is no longer stalled and if it stalls again going to move, again. Which open source project did well after the parent company got acquired?

                                                                                    1. 3

                                                                                      I agree with you that there’s some risk after a big acquisition. I didnt know lighttpd was active again. That’s cool.

                                                                                      1. 2

                                                                                        If it was still as dead as it was a couple of years ago I would have continued my search. :)

                                                                                        1. 1

                                                                                          Well, thanks for the tip. I was collecting lightweight servers and services in C language to use for tests on analysis and testing tools later. Lwan was main one for web. Lighttpd seems like a decent one for higher-feature server. I read Nginx was a C++ app. That means I have less tooling to use on it unless I build a C++ to C compiler. That’s… not happening… ;)

                                                                                          1. 3

                                                                                            nginx is 97% C with no C++ so you’re good.

                                                                                            1. 1

                                                                                              Thanks for correction. What’s other 3%?

                                                                                              1. 2

                                                                                                Mostly vim script with a tiny bit of ‘other’ (according to github so who knows how accurate that is).

                                                                                                1. 1

                                                                                                  Alright. I’ll probably run tools on both then.

                                                                                                  1. 2

                                                                                                    Nginx was “heavily influenced” by apache 1.x; a lot of the same arch, like memory pools etc. fyil

                                                                                      2. 2

                                                                                        SuSE has been going strong, and has been acquired a few times.

                                                                                        1. 1

                                                                                          SuSE is not really an open-source project though, but a distributor.

                                                                                          1. 3

                                                                                            They do have plenty of open-source projects on their own, though. Like OBS, used by plenty outside of SuSE too.

                                                                                2. 5

                                                                                  It’s a web proxy with a few other features, in at least 99% of all cases.

                                                                                  What cool new features are people using?

                                                                                  Like, reading a few books on the topic suggested to me that despite the neat things Nginx can do we only use a couple workhorses in our daily lives as webshits:

                                                                                  • Virtual hosts
                                                                                  • Static asset hosting
                                                                                  • Caching
                                                                                  • SSL/Let’s Encrypt
                                                                                  • Load balancing for upstream servers
                                                                                  • Route rewriting and redirecting
                                                                                  • Throttling/blacklisting/whitelisting
                                                                                  • Websocket stuff

                                                                                  Like, sure you can do streaming media, weird auth integration, mail, direct database access, and other stuff, but the vast majority of devs are using a default install or some Docker image. But the bread and butter features? Those aren’t going away.

                                                                                  If the concern is that new goofy features like QUIC or HTTP3 or whatever will only be available under a commercial license…maaaaaybe we should stop encouraging churn in protocols that work well enough?

                                                                                  It just seems like much ado about nothing to me.

                                                                                  1. 6

                                                                                    maaaaaybe we should stop encouraging churn in protocols that work well enough?

                                                                                    They don’t work well enough on mobile networks. In particular, QUIC’s main advantage over TCP is it directly addresses the issues caused by TCP’s congestion-avoidance algorithm on links with rapidly fluctuating capacities. I share your concern that things seem like they’re changing faster than they were before, but it’s not because engineers are bored and have nothing better to do.

                                                                                  2. 4

                                                                                    New cool features will only be available in the commercial version, because $$.

                                                                                    Isn’t that already the case with nginx?

                                                                                1. 11

                                                                                  This may be incomplete or wrong: My understanding is that you send a search query to a searX instance, which then sends the query to various conventional search engines (google, bing, etc.), receives the results and passes them on to you.

                                                                                  It seems to me that hosting your own searX instance offers very little benefit in terms of privacy. If there are only a few searX instances, each with many users, then it acts a bit like a VPN. Google still gets all the search queries, but it can’t identify individual users of the instance. However, if you have your own searX instance which only you use, then google can still build up a picture of an individual (you), because the searX instance is just a simple proxy between you and google.

                                                                                  Is this correct, or does searX do something more complex under the hood (such as caching or federation with other instances)?

                                                                                  1. 2

                                                                                    I don’t know about SearX, but there’s YaCy, which does federate. It’s effectively a distributed search index. As with all approaches not relying on getting results finally from Google or Bing, the search results aren’t of high quality, though.

                                                                                    1. 2

                                                                                      It seems to me that hosting your own searX instance offers very little benefit in terms of privacy.

                                                                                      But it does. It strips, or at least normalizes, any identifying information in the request.

                                                                                      From the searx faq:

                                                                                      Searx protects the privacy of its users in multiple ways regardless of the type of the instance (private, public). Removal of private data from search requests comes in three forms:

                                                                                      • removal of private data from requests going to search services
                                                                                      • not forwarding anything from a third party services through search services (e.g. advertisement)
                                                                                      • removal of private data from requests going to the result pages

                                                                                      Removing private data means not sending cookies to external search engines and generating a random browser profile for every request. Thus, it does not matter if a public or private instance handles the request, because it is anonymized in both cases. IP addresses will be the IP of the instance. But searx can be configured to use proxy or Tor. Result proxy is supported, too.

                                                                                      1. 2

                                                                                        The “generating a random browser profile for every request” part of it is what makes SearX interesting to self-host, because a random browser profile for every request makes it that much harder for the asked engines to track me.

                                                                                        1. 2

                                                                                          Exactly, and the fact that it comes from a single IP means that they may attempt to build a ‘user profile’ for that IP address but any number of users could be hidden behind it, so it would definitely be an exercise in futility for google, etc.

                                                                                    1. 3

                                                                                      This just inspired me to create my own minimal, VTE-based terminal emulator, to get rid of some of the minor things that I found annoying with my previous terminal (gnome-terminal). The major benefit is that there are no menus now, neither app menus nor context menus, and no window decorations, which makes the thing get out of my way even better.

                                                                                      The last time I was unhappy with my terminal of choice (ROXTerm at the time), I went hunting for a good one, didn’t find any, and settled on gnome-terminal because that was the closest. Seeing how easy it is to build something on top of VTE, I no longer need to settle. Thank you!

                                                                                      1. 0

                                                                                        A sane versioning scheme […] That was easy.

                                                                                        No, semver is probably not the final solution.

                                                                                        From what I understood one problem is when you want to order versions. For example, you specify a dependency as “>=2.7”. Now is that condition fulfilled for “2.7-beta”? Is “2.7-beta” earlier or later than “2.7-rc3”?

                                                                                        Also, there is CalVer which uses dates. Example: Ubuntu 18.04 (meaning April 2018). Also ComVer a simplified SemVer.

                                                                                        1. 3

                                                                                          Semver uses 3 .-separated decimal numbers. -rc3 isn’t semver.

                                                                                          1. 2

                                                                                            Exactly. If you want to use a beta or rc release, you should vendor it or select a git commit, as in Cargo.

                                                                                            1. 2

                                                                                              -rc3 isn’t semver, because it should be -rc.3, see SemVer#9. SemVer has rules governing pre-releases.

                                                                                            2. 2

                                                                                              There’s also ConVer (content versioning) which uses git commit hashes, although the fact that I can no longer find a link to the site/article probably speaks to how popular that idea is.

                                                                                              1. 2

                                                                                                In practice, the answer to those questions is no, and earlier respectively.

                                                                                                And there is work on making those things official: https://words.steveklabnik.com/what-s-next-for-semver

                                                                                                1. 2

                                                                                                  SemVer#11 lists the ordering. >= 2.7 is not satisfied by either of your examples. As for their ordering, 2.7-rc.3 > 2.7-beta.1, as specified by SemVer.

                                                                                                1. 16

                                                                                                  Advertising, with no prior engagement. :(

                                                                                                  1. 17

                                                                                                    Hi, my apologies, I might have been not used to the rules/culture of this place… Let me apologize if this link submission made you see some unwanted ad spam. I was thinking that the story we posted might have some value as a technical reading, but ultimately you’re right, it serves as an advertisement as well. From now on I’ll be more careful when posting links here and try to separate technical content. Again, I’m sorry for disturbing your pleasant afternoon (although you might be in a different timezone…).

                                                                                                    1. 10

                                                                                                      No worries, and congrats on a product launch. The problem is just that this site would quickly drown in marketing and advertising if people’s first few interactions here were posting their products or services. :(

                                                                                                      1. 6

                                                                                                        I like the video you have demonstrating the board: https://www.youtube.com/watch?v=th7KZzGz17s -

                                                                                                        1. 8

                                                                                                          Thank you… I’m pretty sure that, if I manage to survive another decade, my friends will then share that video to embarrass me… But that was the best I could come up with with the limited budget (The video took $15 to buy a tripod to stabilize the filming). I’m glad that at least one person liked it… 😂

                                                                                                          1. 7

                                                                                                            All those friends should envy you since you actually did something instead of being an armchair critic 😉

                                                                                                      2. 11

                                                                                                        Agreed, but there is interesting technical content as well: https://moonrim.io/compare - although you could call that advertisement as well, but a 10 page advertisement then.

                                                                                                        Maybe I just like keyboards too much to be annoyed by advertisement that appeals to me. (So ads about cool ergo boards).

                                                                                                        This should have a buyer beware title, because it’s a crowdfunding and you might not get a product (or three years later like the UHK) and lost your money.

                                                                                                        1. 4

                                                                                                          I want to upvote but the thing seems super suspicious. Just at the crowdfunding stage.

                                                                                                          Are there other, similar keyboards on the market?

                                                                                                          1. 10

                                                                                                            Hi. Maker here. I hope the following helps remove some of the suspicion:

                                                                                                            • Some design ideas of this has already been tried and produced. If it helps, you might want to take a look at ErgoDox, Keyboardio, and DataHand.
                                                                                                            • We’ve built a prototype and, admittedly, the technological barrier is not high to make this thing if you have some knowledge in electronics and programming. The issue is more of how to gather enough people to make mass production feasible; To make plastic injection molding financially viable, we need at least about one thousand people willing to buy it, hence the fixed funding goal. Speaking of which… The goal is fixed (instead of flexible funding goal) so if it doesn’t reach $200k before the end of the campaign, the funding will be canceled automatically and your payment will be refunded in its entirety.
                                                                                                            • If you doubt whether I’m a fraud or not, you are totally rightfully so. It seems there are so many crowdfunding scams these days… So, if, by any chance, you are interested in my track record: I’ve worked as a software engineer in South Korea for a while. I’ve been an organizer of Haskell-KR for a while, so my name is not entirely unknown to the Korean programming community. (If you look at the backers list, you’ll see the early backers are mostly Koreans with typical Korean names). My previous workplace was LINE, where I held a programming language boot camp that was featured officially on LINE’s engineering blog. And… If there’s anything else I can do for you to trust us, please never hesitate to tell us.
                                                                                                            1. 4

                                                                                                              I love me some split keyboards. I just wrote this article on my five year adventure in split keyboards: https://raymii.org/s/articles/Split_keyboards_a_five_year_review_including_the_ErgoDox_EZ_Matias_Ergo_Pro_and_Kinesis_Freestyle_2.html

                                                                                                              The fixed goal is nice, that gives people a bit more trust to join in with the funding. I’ve sadly got bad experiences with Indigogo’s where money just vanished without a product. That is, you are not buying a product but supporting a project and maybe getting a perk.

                                                                                                              I’m absolutely not saying you’re a fraud and I love the project, will probably back it since I do love me a good keyboard.

                                                                                                              1. 2

                                                                                                                Ditto wrt indigogo and vanished money. I’d be more inclined to contribute to crowdsuppy, kickstarter, or, alternatively, massdrop, all of which I’ve had great experiences with, but indigogo is a red flag as I see it.

                                                                                                                1. 2

                                                                                                                  Indeed if you compare Kickstarter and Indiegogo, the former has a higher reputation. We did want to use Kickstarter, but opening a project on Kickstarter demanded permanent residency of one of 26 countries (1). We’re Koreans living in South Korea, so we sadly couldn’t. :(

                                                                                                                  (1)

                                                                                                                  Project creation is currently available to individuals in the US, UK, Canada, Australia, New Zealand, the Netherlands, Denmark, Ireland, Norway, Sweden, Germany, France, Spain, Italy, Austria, Belgium, Switzerland, Luxembourg, Hong Kong, Singapore, Mexico, and Japan who meet the requirements below.

                                                                                                                  https://help.kickstarter.com/hc/en-us/articles/115005128594-Who-can-use-Kickstarter-

                                                                                                                  1. 2

                                                                                                                    I was sold the second I saw this keyboard. I’m not thrilled about indigogo, but I see what you’re saying regarding kickstarter and you have my support. I hope to see moonrim II get fully backed. Good luck!

                                                                                                                  2. 1

                                                                                                                    FWIW, there have been some pretty successful projects on IGG, like the ErgoDox EZ. Of the 5 projects I backed there, three already delivered, a the fourth is on track to deliver too. The only project that vanished with my money on IGG was the Jolla Tablet.

                                                                                                                    1. 1

                                                                                                                      I completely forgot about the Ergodox Ez - as a long time user of the classic ergodox (massdrop kit) I backed the EZ on indigogo well and have never regretted it.

                                                                                                                2. 3

                                                                                                                  We’ve built a prototype and, admittedly, the technological barrier is not high to make this thing if you have some knowledge in electronics and programming. The issue is more of how to gather enough people to make mass production feasible.

                                                                                                                  I appreciate your candor on this point, and overall in the campaign description and demonstration video.

                                                                                                                  I’ve been considering alternative keyboards for a couple of years now, and thought about using my 3D printer to build a Dactyl keyboard, which is also a concave columnar layout, but did not find the time. One thing I missed was an integrated pointing device: I thought about adding one, then contemplated all the changes I would have to do for it, and decided to leave it for later.

                                                                                                                  Having you deal with all that is well worth the crowdfunding price for me, and I just backed it.

                                                                                                                  1. 2

                                                                                                                    Thank you very much 😂 We really want to deliver a quality product to you. As this is a fixed funding goal and it gets canceled if it doesn’t reach the goal, we desperately need to spread the word to gather about 1,000 people. If there’s any chance anyone around you might also be interested, please tell them. Again, thank you so much for backing Moonrim II.

                                                                                                                  2. 1

                                                                                                                    The only thing that sticks out is the price. It is oddly low. :)

                                                                                                                    In any case: good luck on this mission! I love the idea, and yes, I’ve heard of those keyboards. Yours doesn’t look anything like those, but like a kinesis advantage pro that is domed.

                                                                                                                  3. 1

                                                                                                                    You could buy an ergodox and glue / Velcro it to a bookshelf stand (bookend)? or two planks at an angle (reverse T)

                                                                                                                1. 10

                                                                                                                  I dunno. This looks like feature creep.

                                                                                                                  Want to start a conversation but no code has been written: Open an issue. You’re PR is going to fix a bug or add a feature. This starts as a issue that is discussed, voted upon and milestoned

                                                                                                                  Your PR is not ready for merging yet: Keep it in your branch/fork and keep working on it. It’s not ready for a PR. Have questions, ask it on the issue. Want a code review, request one on the issue. Or do a regular PR and request comments, because you can keep adding commits to the PR.

                                                                                                                  1. 10

                                                                                                                    Keep it in your branch/fork and keep working on it.

                                                                                                                    I suspect this doesn’t work for most people because you can’t (easily) see and comment on a diff between your branch and the base one without opening a PR. Which is exactly what they should have done instead of creating another redundant first-class entity in the already over-complicated UI.

                                                                                                                    1. 6

                                                                                                                      t’s not ready for a PR. Have questions, ask it on the issue. Want a code review, request one on the issue.

                                                                                                                      Send the patch to a mailing list and discuss it there inline within the comfort of your favorite mail client of choice.

                                                                                                                      1. 3

                                                                                                                        Hard to deny that these days most people find a Web UI more comfortable than any of the (stagnating) mail clients. Minus the hassle of subscribing to a mailing list.

                                                                                                                        I mean, I’m totally with you that all of that is just as easy, but most people seem to have a different perception.

                                                                                                                        1. 4

                                                                                                                          Minus the hassle of subscribing to a mailing list.

                                                                                                                          This is a common misconception; there is no need to subscribe to the mailing list if it’s set up correctly. If you send your patch there, people will include you in the reply list regardless.

                                                                                                                          I agree it’s distressing how much GMail’s near-monopoly has hurt the ecosystem and caused people to forget that better clients exist.

                                                                                                                          1. 1

                                                                                                                            In some cases, like work, you’re forced into GMail to avoid someone exfiltrating your IMAP synced mailbox via malware. So…

                                                                                                                            1. 2

                                                                                                                              Yeesh; hopefully work policies don’t prevent you from making OSS contributions with your personal mail account tho.

                                                                                                                              1. 2

                                                                                                                                In some cases, yes. For the types of stuff I’d do in my spare time, most likely not.

                                                                                                                                But my guess is that this restrictive email thing will become more popular, not less, meaning that important projects, like those you get paid to work on during work hours, won’t be able to successfully support contributions that way.

                                                                                                                                I dunno though. I could certainly be completely offbase here and we will see a resurgence of plain text email, and a new generation of mail tools. Fastmail is betting on JMAP, which uses JSON, so the barrier to entry might get much much lower in years to come…

                                                                                                                        2. 1

                                                                                                                          A mailing list you say? So I have to fight through all the spam, and everything else that comes with it? Should I set up a list for every single repo? Should I write my own automation for code review and assignment?

                                                                                                                          I mean, on GitHub, I can clearly see who I requested code review from (and whether they reviewed it, and what their review is), who’s assigned to the PR/issue, what milestone it belongs to and so on. Can I do that with a mailing list? Yes, I could search, use labels, bookmarks, whathaveyou in my client of choice, but there’s no built-in thing that provides me with this information. One can build it on top of mailing lists, but then I’d use the API of that, and the mailing list would be reduced to transport only.

                                                                                                                          As it happens, I can reply to GitHub notifications by email, from the comfort of my favourite mail client. And I get to enjoy the benefits of the API too. So no, a mailing list is not a substitute. For tiny things, or if you’ve established an email-based workflow already, it can work. For a lot of projects, it doesn’t.

                                                                                                                          1. 2

                                                                                                                            So I have to fight through all the spam, and everything else that comes with it?

                                                                                                                            The last time spam got thru on any of my mailing lists was over a decade ago. (Basically never happened since I moved off Google.)

                                                                                                                            Should I set up a list for every single repo?

                                                                                                                            Another advantage of this setup is that you aren’t locked in to a 1:1 mapping between issue tracking and repos. You can easily set up a single mailing list to discuss patches for a group of related repositories under a single project, or a separate mailing list for a subset of development on a single large repository.

                                                                                                                            1. 4

                                                                                                                              The last time spam got thru on any of my mailing lists was over a decade ago. (Basically never happened since I moved off Google.)

                                                                                                                              Lucky you. Plenty of my non-google addresses get quite a bit of spam, can’t imagine it being any different for mailing lists. The single list I run gets plenty of spam. Most are caught, but only because I spend non-negligible time making sure they are. I’d happily let go of this task.

                                                                                                                              Another advantage of this setup is that you aren’t locked in to a 1:1 mapping between issue tracking and repos. You can easily set up a single mailing list to discuss patches for a group of related repositories under a single project, or a separate mailing list for a subset of development on a single large repository.

                                                                                                                              I can do the same on GitHub with GitHub projects. I get e-mail notification, I can reply via e-mail, but I also have an API to access all of that information - and more - whenever I want, from wherever I may be, without having to carry my email archive around, or host my email on a server I can access on the go. I can also selectively subscribe to issues (via the API, so I don’t ever need to fire up a web browser), and get e-mail notifications. The API lets me query on labels, milestones, etc. Doing the same on a mailing list is very far from being trivial, or convenient, especially when I want to search backwards to times I weren’t subscribed yet.

                                                                                                                              If you don’t need the labels, don’t use milestones and various other features GitHub provides, a mailing list might be a suitable alternative, and a convenient workflow. The moment you start to use these, a mailing list will stop being adequate. No, Subject: [project:subsystem bug#N]: blah is not a solution, and doesn’t work. Client-side labeling/filtering doesn’t either, because the point of labels & milestones is that they’re centralised. N+1 mailing lists for subsystems is a pain to maintain, and even bigger a pain to work with (been there, tried it, no thanks).

                                                                                                                              In my experience, working with Magit & Forge is much more convenient than mailing patches and whatnot around.

                                                                                                                        3. 5

                                                                                                                          You say feature creep, I say the end of a long period of feature stagnation.

                                                                                                                          Enough companies use Github as part of their development infrastructure – not just process – that it makes sense to have a programmatic way of flagging a PR as “not ready yet”. The rest of us will just type “[WIP]” less often. Is it that bothersome?

                                                                                                                          1. 4

                                                                                                                            Do you all not use a WIP label?

                                                                                                                            1. 6

                                                                                                                              I do, and include that same title. No one looks at labels, even if they are blinking and bright red.

                                                                                                                              1. 2

                                                                                                                                Reminds me of how many PRs we’ve merged that have a “Do Not Merge” label on them.

                                                                                                                                1. 1

                                                                                                                                  We have a policy of only allowing PRs to be merged by a reviewer, never by the assignee. You may self assign as a reviewer but normally a PR will be created by one person and merged by another. If nobody has been assigned it sits until someone is.

                                                                                                                            2. 2

                                                                                                                              I’d prefer to open a PR to start a discussion, but I’ve worked with people who prefer to see PRs as ‘ready to go’ and get shirty when you open a PR that’s not merge-ready.

                                                                                                                              Making it explicit is handy for dealing with people who aren’t interested in learning how to use patches-by-mail but you want a place to discuss work in progress.