AtNightWeCode

Go ecosystem is like a fraction of what .NET/Java is. Both in quality and size. Dart for instance is a better lang. It is also from Google. It also compiles to binaries. Why is that not the new...

The slim spot for Go is the devops tools space. Where Go may or may not survive. Considering the march of Rust.

No. It would only be a flag to allow obsolete Go code in practice I guess.

EDIT: C++ did not break C. The people managing Go have full understanding of dependencies even though they made breaking changes.

I suggest Go and Dart morph into one lang. Then you could use the power of Go with the more logical syntax of Dart.

Historically this service is known for being highly incorrect.

It can easily be done with WASM. You could even add the domain as a salt.

Are you aware of that secure pw hashing comes with a cost factor and that the cost is linear with the size of the input to the hashing algorithm?

EDIT: Down voters. What pw hashing algo are you using instead of BCrypt or similair?

What? I always use my daily backup of the Internet as the pw every time I sign up for something.

256 is fine in most user scenarios. No one will type in more chars than that anyway. And if a pw manager is used the length is not really a problem.

For services there are often better alternatives.

Passwords are sent to the server though in most cases. So there must be some limit.

I fail to see how these issues really matters. My issues with LLMs is that there is no score to tell the quality of an answer. The second thing is that it adds cognitive load and noise. In automations that might be fine but tools like Github Copiliot are more problematic. Once I got rid of it my dev speed clearly went up.

I think the author mixed up prototyping with proof of concept. A POC should be tossed out. I think most companies expect at least the senior devs to be able to write a feature on first try.

I mostly use 1.1.1.2 or 1.1.1.3 depending on the device. That is CF with different restrictions.

On Windows the big problem with DNS is that it can be changed randomly. Among all the garbage on Windows this is peak garbage. Any con can change the DNS that I explicitly told it use. DNS by it self is probably top 10 sec risk on desktop envs and it so completely broken by design. I hack it more or less daily to "debug" our sites.

Conservative people leave us with DNS, TLS/Public CAs and other sht and then they whine on HN day in and out why did a "root" cert of MS leak or whatever. Securing TLS or DNS with automatons at a large scale company is a complete nightmare.

I only use Chrome for web dev things. Been living for decades and I still don't understand how people can be so fking stupid.

Microsoft have created more bad API:s.

Open XML

Graph API

Large parts of the C# Azure SDKs

MFC - perhaps not bad just very difficult

Edge is just where something is hosted. It has nothing to do with data consistency. User signup is a unique problem that any large site has. Some sites like HN also actually needs usernames which makes things more complicated. Owasp always recommend to add usernames to any login. It is such a naive approach. User names are garbage in most cases.

Very good point.

A gotcha along the same path. Deserialization of things not needed as what you get with generated clients. An aspect of interfaces in Go I really like. Remotely type what I use. Skip the rest. Not fun to have incidents caused by changes to a contract that is not even used by a service. Also hard to find.

Agree. You get a discount if something breaks. But SLA really only works for larger services where the cost of fixing something is small when compared to the discounts.

I think the article is somewhat incorrect. I really dislike all these custom chatbots. But it is also fairly easy to measure how good they work. And for the sites I work with at least the chatbots perform better than other options and they do offload work from the support teams.

Don't use *. It is a common source for all kinds of problems. Many query langs does not support it at all for that reason.

If SO was worried about that drop I think they would have bought back some of that traffic. More likely something has changed how they count the visits or they blocked some bad traffic. Traffic data is often sampled as well.

The fall in the beginning of 2023 may be the introduction of ChatGPT. A more worrying idea is that the numbers reflect not just the decline of SO but a decline of the whole IT business.

ChatGPT. Both that people use it instead but also that some people try to game SO with answers from it. So the moderation of the site has probably become harder.

SEO is another factor. At least I don't get as many results on SO as I used to. That may be because of many reasons but if the popularity of a site starts to fall Google search might speed that up.

EDIT: I see now that the drop started long before ChatGPT became relevant. But it won't help SO I think. Github copilot and Google featured snippets may also be factors.

I have seen many failed or very expensive projects. An unordered list of issues.

1. Sunk cost. Quit before it hurts. 2. Non-technical leaders. Talk to experts when it comes to tech. I don't care if you watched a Youtube-vid. 3. Tech people that think they know it all. Don't build waste that solves problems that the business does not have. 4. Person cult. Anybody can make simple mistakes. 5. Department cult. Quit pretending that the guys at your department are smarter. 6. No scope. Define what to do. 7. No order. Prioritize what is important. 8. Premature or upfront decisions. Why create time estimates first and then figure out what to do. 9. Cowboys. Crazy time estimates that someone high up in the hierarchy just made up without talking to anybody. 10. Everything tech. Not sure why, but some dev teams can't add a simple feature without rebuilding half the tech stack. Features are built on current stack. Tech is built with tech decisions but may be necessary to add a feature. 11. Overlapping. Multiple people doing the same work. 12. Bad architecture. Architecture that does not simplify anything.

Rick is right I believe. But why not just use overloading instead? That is what is done in many OO-langs to avoid making breaking changes.

The typeface is not monospaced. It is a garbage sans serif clone with some stupid sans-letters like lower case i and l that looks just ridiculous. Sweden used to use Helvetica (or some typeface very close to that). They should revert to that.

Avro does not even come with deserialize to object in the libs. A terrible choice for interoperability.

I always try to avoid these special formats. Most people have not used them and even if there are claims that the tools work on different platforms there are often some quirks. Another approach is to add compression and tweak it for performance.

And doping. Now you think what? But. It is hard to compete with people doing drugs to be able crunch 16h seven days a week to get a result.

Governments need good design and a versatile typeface without any shenanigans.

Not what I said. Do you think they got the font for free? They used tax payers money to hire probably a brand corp that created this font as a part of the brand.

What you are not understanding is how corps are selling fonts. The Font is garbage.