To some extent I agree, however if there is too much code in the constructor it also is quite often an indicator that the code should be refactored. The convenience of what the RAII principle is providing allows also to avoid coding and resource management mistakes/bugs. In c++ exceptions are considered to be used for exceptional circumstances, and they are quite expensive for binary size, at least I made that experience when I used to work on some bigger applications. On the other hand I have been writing a lot of python code in the last few years and exceptions are the defacto way to handle errors. And it’s very much not a problem to use exceptions for non exceptional use cases.

In C++ I rather would argue that the approach of acquiring resources during initialization makes sense because it reduces your need for introducing checks that could lead to wrong code paths especially in the destruction. When applied correctly your program will be more robust if you use the raii principle. I recommend reading the exceptional C++ books by Herb Sutter those are very much still one of the best sources in writing C++ code that is handling exceptions properly.

Complexity in constructors and destructors is something to avoid which is something as I understand is your main criticism to the principle.

What’s the point though? If you don’t want to use it, why bother? Why do you care what “Linux users seem to be so obsessed” with?

The world would be a lot better without this kind of hate/phobic trash. Doesn’t matter if it’s desktops or dick, nobody should care about what another person likes if it’s not hurting anyone else.

Would you tell the community so that other devs don’t have to go through the same problem?

1. Switch to a bank that does invoice factoring: They pay roughly 85% of the invoice value when you invoice, so you need to raise your rates by 15%.
2. …then you can offer a 15% discount to any company who would prepay.
3. Consider subcontracting instead of referrals: It makes it easier for you to get the revenue you need for step #1, it makes it easier to engage a lawyer if it comes to it.

I am asking what would you do in my situation since in general controversy in business is always a bad idea even if you are the injured party. … The client is a startup that is about to do an IPO. … Some of their investors have a relationship with a very well known and established business magazine. I am not sure they aware of this.

I would find out which company is helping them do the IPO and talk to them I might find a commercial debt collection agency which is willing to take my case, then contact as many trade journalists and investors I can to explain that they’re really raising money to prepare for a lawsuit against creditors.

Then I might call the PR office that these articles are coming, and that I want to make sure that this wasn’t all some “honest mistake” on their part. Obviously they need to pay the full invoice immediately plus any expenses I incurred in this process, but it’ll also be good if they compensate me for any stress I had to endure.

If you need help with this, reach out to me privately.

Have you used a debt collection agency?

Yes. In general, they’re worth as much as you pay them, but except for the above I wouldn’t bother with them since it doesn’t sound like this company is unaware of their debt with you, but is being wilful.

It doesn’t sound useful in this case, but if I’m really worried someone is going to go bankrupt I get on DeBN. I might do that anyway since it’s easy. If they file for bankruptcy simply prepare a proof of claim and deliver it to the clerk. Only then do I engage a commercial debt collection agency. If they’re quiet and there’s no bankruptcy, I’ll keep adding interest until it becomes worth my time (to hire a lawyer).

Have you filled a legal complaint against a company before?

I’ve served notice to companies, and that’s generally all it takes. Companies keep the registered agent information up-to-date in a way they neglect for the accounts payable department.

Suing for unpaid debt in the USA is actually very easy:

1. Send a registered letter to the company’s registered agent informing them of their debt and that you will attempt to recover it if they do not respond in 30 days. They have a right to deny the debt, and they have a right to request an alternate way to resolve it. A lawyer should not charge you more than $1k to write such a letter. They may incur $2-5k in additional costs responding to replies, and of course: talking to you.
2. File the complaint with the court. If you file federal it will cost $400. If you are a US corporation, then your costs will vary based on the state you choose “at home”. It will be less controversial to file in the defendants state but it may be cheaper to file in another.
3. Serve them the complaint. A routine serve should cost between $50-$75.
4. You will receive a court date. You and your lawyer will need to show up. A lawyer may cost a few thousand dollars to show up and prepare (you will want them to prepare).
5. If they don’t show up (for example because you served the complaint by singing telegram) you can file request for default judgement. This will cost another $400 for the filing, but you won’t need as expensive a lawyer to show up.

Between travel, filing, and so on: $15k sounds extremely reasonable for your situation. I’ve never had to finish a court case I’ve started – I have however been sued and won, so I still have some idea what the entire process looks like.

Learning how to talk to lawyers and use them effectively is a very useful skill. It can easily mean the difference between spending $20k and $200k. A key thing is that they cannot tell you what to do. They can give you your options and point you to the most immediate one, but they won’t tell you to go nuclear early, and because you’re outside the US this might be your best move: You will have to tell them that you want to do it.

Giving them a sob is also a good way to spend a lot of money. If you say “here’s my invoice, I want to collect it” it’ll be much cheaper than what you tell them everything you told us.

I’ve never heard of this joke, but if you want to find a cycle in a linked list, the canonical way to do it is to use two pointers and have one walk one step at a time, while the other walks two. If they are ever equal after they take a step, there’s a cycle. Using signals is – I think it’s safe to say – way over the top. (Unless that’s the joke.)

Huh. If it’s fine to create your own “mediation pipe” to satisfy the API, and then splice(in, pipe); splice(pipe, out) … why doesn’t the kernel support that itself?

There is no way to measure the number of times a billboard on the side of a highway has been looked at.

Companies like Verizon Precision Marketing do exactly this: They monitor how many people walk or drive by (facing) a given billboard using radio data from mobile phones.

It is not possible to estimate the percentage of people who glanced at a magazine ad and subsequently bought the product.

Of course it is. You try a new publisher for three months and look at the sales uptick.

You can estimate, sure, but it is an inexact science.

Digital is in the same boat: An “impression” can be caused by your cat.

I think we all know this though.

The amount of personally identifiable information companies have about their customers is absolutely perverse.

It’s also of very low-precision. So much so that a well-targeted ad actually gets into the news. If you don’t use an ad blocker, you can see the topics Oracle/BlueKai are monitoring at http://bluekai.com/registry and while there are some aggressive persona-based approaches to data, they’re only good in the statistical sense.

Publishers will call [blocking ads] unethical…

I know one gaming review site that sees 50% of their users with ad blockers. Their strategy? Double the ads. More clickbait. Reduce the quality year on year. But what can we do?

It’s a dark path, but it’s difficult to get anything else out there.

It is the publisher’s responsibility to develop a business model that is sustainable and ethical.

Says the man who uses Google and Twitter trackers on his own content.

Small publishers lack the ability to do this, so the “end game” is an “Internet” with Google, Facebook/Instagram, and Twitter. That’s it. I don’t want that Internet. I think it’s a horrible place.

Something else to consider: Chrome, Firefox, node.js, and literally everything you think you like is funded by advertising either directly, or by “speakers” and “contributors” who work for companies that do advertising.

Advertising has a long tail – the current state of digital marketing is a long way from sponsorship, but unless we too make suggestions on how publishers can get paid, they’ll just continue to taxi in the toilet as long as they can.

Russ Cox wrote:

At Bell Labs, Rob switched acme and sam from black and white to color in the development version of Plan 9, called Brazil, in the late fall of 1997. I used Brazil on my laptop as my day-to-day work environment, but I was not a developer. I remember writing Rob an email saying how much I enjoyed having color but that it would be nice to have options to set the color scheme. He wrote a polite but firm response back explaining his position. He had worked with a graphic designer to choose a visually pleasing color palette. He said he believed strongly that it was important for the author of a system to get details like this right instead of defaulting on that responsibility by making every user make the choice instead. He said that if the users revolted he’d find a new set of colors, but that options wouldn’t happen.

It was really a marvelous email, polite yet firm and a crystal clear explanation of his philosophy. Over the years I have from time to time spent hours trying to find a copy of that email. It is lost.

Model M reporting in.

I’d like to find something new, but – and for reasons I haven’t investigated or discovered – even newer versions of the keyboard, where they claim to use the same switches, don’t feel the same. I wonder if it’s like leather shoes and I’ve simply developed a preference for the worn-in feeling.

The rest of the time, I use whatever keyboard is on my laptop (I have a MacBook Air 11” that is great for programming but aging, and a MacBook 12” with the shitty disgusting butterfly keyboard)

I’ve heard the “binary logs are evil!!!” mantra chanted against systemd so many times that it wasn’t funny anymore. It’s a terrible argument. With so many big players putting their logs into databases, the popularity of the ELK stack, it is pretty clear that storing logs in non-plaintext format works. Way back in 2015, I wrote two blog posts about the topic.

The gist of it is that binary logs can be awesome, if put to good use. That the journald is not the best tool is another matter, but journald being buggy doesn’t mean binary logs are bad. It just means that the journald is possibly not the most carefully engineered thing out there. There are many things to criticize about systemd and the journal, and they both have their fair share of issues, but binary storage of logs is not one of them.

a.out binaries are smaller than elf binaries, so let’s statically link everything into one big executable ala busybox.

Similarly, a modular kernel with all its modules takes up more total space than a single kernel with everything built in. So don’t even bother implementing modules. Linux 1.2 was the last great Linux before they ruined everything with modules.

64-bit code takes up more space than 32-bit, so let’s build for 32-bit instruction sets. Who has more than 4GB of addressable memory anyway?

Optimized code usually takes up more space, often a lot more when inlining is enabled. Let’s build everything with -Os so we can fit more binaries on our floppies.

Icons are really superfluous anyway, but maybe we’ll want X11R5 or some other GUI on a second floppy. (I’d say X11R6 but all those extensions take up too much space). Make sure to use an 8-bit storage format with a common palette – 24-bit or 32-bit formats are wasteful.

(I lament the bloated nature of the modern OS as much as the next Oregon Trail Generation hacker, but really – is “fits on a 1.7MB floppy really the right metric? Surely we can at least afford to buy 2.88MB drives now?)

I feel like APL took terseness too far. Every code snippet looks like somebody was playing “code golf”. It may be great for demoing the language, but won’t it be a nightmare for real code?

And I could technically achieve (nearly) the same thing in Lisp by giving my functions and variables names like “ῴ”, but it’s easier to understand when it’s spelled out like “solution-matrix”. Just because everything can be abbreviated to a single symbol doesn’t mean it should be.

Also, as neat as these purely algorithmic problems are, what does real life code look like in APL? What’s an HTTP request look like? How would I parse a JSON blob?

• Safari
• Firefox

Note especially that Safari doesn’t (and won’t) support pointer events and Chrome has flip-flopped on the idea a couple times.

A super common example for “clean code” is that instead of doing p * t to add sales tax, you write price * sales_tax. I pulled up an ecommerce suite and looked how they did it. They track an array of the tax markups, one for each applicable tax, and bundle that along with the cart.

btw, “clean code” (I hate that name, I prefer “intention revealing code”) is much more than nicely named variables, it’s well partitioned code at multiple levels. Short methods with meaningful names, divided into classes that are cohesive, divided into modules, etc., with cohesion at each level, and each one being at the same level of abstraction.

I took a look at the tax.php code and, while I’m no PHP developer, that’s not at all how I’d do it, because that code isn’t OO, it’s just code that happens to have method names. We can argue about OO vs. FP, but if I were to write this as OO code, I wouldn’t have code like this (in fact I’d hardly have if statements at all):

if ($calculate != 'P' && $calculate != 'F') {
	$amount += $tax_rate['amount'];
} elseif ($tax_rate['type'] == $calculate) {
	$amount += $tax_rate['amount'];
}

That’s a sign that the author doesn’t understand OOP, or didn’t want to do it.

The WSL is very good. The console is a bit shit but screen/tmux fixes that enough for me.

For IRC I use emacs+erc.

Someone who controls your network will simply drop the DNSKEY/DS records, so DNSSEC would not have provided any protection for “MyEtherwallet”. People who have already visited it were (hypothetically) protected by TLS, and people who hadn’t, would have received bogus records anyway.

So DNSSEC could, in an ideal setting, provide a benefit similar to HPKP, but why wait? HPKP is here now.

Furthermore, “DNSSEC wasn’t easy to implement” is a massive understatement.

“Protocol-level faculties that let you read, write, and hash specific chunks out of the middle of large files without making you transfer the whole large file? does SSH have that? Nope.” - sftp does have that actually. SFTP is basically just a protocol that fowards file descriptors.

two thing that always bothered me about openssh:

the naming id_rsa and id_rsa.pub means tab completion can cause you to accidentally send your secret key. I would have called it id_rsa.priv

It would be neat if it had more ways to support machine to machine workflows. I use ssh to forward unix sockets to do secure cluster networking, force commands are ok, but they are not the easiest to use.

Are you confident that every single user of your systems is going to out-of-band verify that that is the correct host key?

If your production infrastructure has not solved this problem already, you should fix your infrastructure. There are multiple ways.

1. Use OpenSSH with an internal CA
2. Automate collection of server public ssh fingerprints and deployment of known_hosts files to all systems and clients (we do it via LDAP + other glue)
3. Utilize a third party tool that can do this for you (e.g., krypt.co)

Your users should never see the message “the authenticity of (host) cannot be established”

kdb allows nested columns, so:

q)select email by team_id,name from users lj teams
team_id name | email                                                      
-------------| -----------------------------------------------------------
1       Citus| `craig@citusdata.com`farina@citusdata.com                  
2       ACME | `jennifer@acmecorp.com`tom@acmecorp.com`peyton@acmecorp.com

of course, being king of timeseries means it has lots of good date/time types (most with syntax) so regular operators are extended over them:

q)select email from users where created_at > 2018.06.22 - 7
email                
---------------------
craig@citusdata.com  
jennifer@acmecorp.com
tom@acmecorp.com     

q)select count i by created_at.week from users
week      | x
----------| -
2018.06.04| 2
2018.06.11| 3

JSON is often found in lieu of proper data structures and good design, but if you’ve got a data cleaning exercise, you can dip into q as needed:

q)select email,(.j.k each location_data)@'`state from users
email                 x   
--------------------------
craig@citusdata.com   "AL"
farina@citusdata.com  ()  
jennifer@acmecorp.com "CA"
tom@acmecorp.com      ()  
peyton@acmecorp.com   ()  

This has other great advantages, for example being able to build your application on top of the database cuts out a massive source of latency.

Most of the large resolver services such as Google. Quad9, OpenDNS and Cloudflare are all DNSSEC enabled.

OpenDNS does not support DNSSEC at this time.