Yes. Multiple historical layers (...)

This is reaching a level of idiocy that's completely unheard of.

Just say you know nothing about what you're saying and you're completely oblivious, and sit out the rest of the discussion.

Well ok, but the concern is about the weaknesses, Mr. Stroustrup.

I don't think these discussions on "weaknesses" come from a place of intelectual honesty. None of these arguments even touches the fact that there are already a myriad of freely available static code analysis tools and memory profilers that do a very good job catching memory safety issues.

For some unexplainable reason, these criticisms of C++ always focus on a single strawman: these tools do not exist and no developer in the world cares about the topic.

What I find surprising is that there are a lot of steps between a free-for-all and state intervention through regulation that those experts seemed to have skipped altogether, such as voluntary auditing, state-sponsored industry initiatives to specify best practices, invest in the development of static analysis tools and memory profilers, or making vulnerable companies liable for the consequences of attacks.

But no, they jumped straight into state-imposed regulation. Because keeping people out is a solution?

…it’ll still be 35 users/month

I'm not sure you are aware how irrelevant this is. This could mean as little as a single user opening the community page daily, or 30 different users accidentally navigating into the community page from the main page just because an article showed up in their feed.

To frame the absurdity of this argument, I moderate !nodejs@programming.dev , which in the past month registered also 30 users/month, and that community is also dead.

I enjoyed the read so much that I've created a community dedicated to data structures and algorithms: !data_structures@programming.dev

No, you’re lying by using a different definition of “dead”.

Now you're being silly and acting defensively. I don't need to do anything for the !dotnetmaui@programming.dev group to be dead or remain dead, as it was expected to be. Anyone can take a look at it and see that if they filter out your personal inorganic traffic, which is already of dubious relevance, nothing remains.

You can stay up all night arguing otherwise, but it is what it is.

It's ok if you feel that it's your personal mission to generate traffic for a particular channel on a lemmy instance. Just don't try to pretend it's something that's relevant for anyone beyond yourself.

None of the communities you have mentioned are dead. In fact according to the January stats, all the communities you mentioned have above average usage.

I'm sorry, you're trying to blatantly lie with statistics.

"Above average" means nothing if the majority of communities is already dead. You're just arguing that some communities are more dead, which is pointless.

You're also lying regarding what traffic is being posted to !dotnetmaui@programming.dev. All posts ranging back to the last two weeks come from a single user account: https://programming.dev/u/SmartmanApps .

To make this even more pathetic, the bulk of the posts going into !dotnetmaui@programming.dev were posted by your account after I pointed out the community was dead and already dead on arrival.

You're not refuting the point: you're proving the point that the community is dead.

It's ok if you don't feel a need to change your persona workflow.

Nevertheless I'm not sure you understood the example, so I'm not sure you fully grasp the differences.

The whole point of my example was to point out the fact that, thanks to interactive rebase, you do not even need to switch branches to work on multiple unrelated PRs. You can just keep going by doing small commits to your local feature branch and keep doing what you're doing. In the end all you need to do is simply reorder, squash, and even drop commits to put together multiple PRs from commits that are built upon each other.

Simple, and straight to the point.

I think it's hard to provide examples of a binary format beyond a few benchmarks.

I think that this article is missing the most obvious and perhaps most used package manager for C and C++ projects: Linux distro's package repositories.

It should be stressed that package managers are a requirement for Windows and, to a far lesser degree, macOS. On any Linux distro, you'd just apt install <dependency>-dev and you'd be mostly set. It's mainly Windows development that draws this need to install packages to be able to develop anything.

Honestly, I would argue that git submodule should count as a package manager. I simply list out the repos I want to pull in and update them as needed.

I see your point, but I think that when developers refer to package managers, implicitly that means accessing standalone precompiled binaries that are ready to just be integrated into a project.

With git submodules, unless they are used to track standalone projects or even precompiled binaries, you still have to resolve their dependencies, which is the responsibility of a package manager and the main reason they are used.

Mercure's spec has a bit more meat on its bones in terms of providing an adequate description of the protocol.

From https://mercure.rocks/spec :

Mercure provides a common publish-subscribe mechanism for public and private web resources. Mercure enables the pushing of any web content to web browsers and other clients in a fast, reliable and battery-efficient way. It is especially useful for publishing real-time updates of resources served through sites and web APIs to web and mobile apps.

Subscription requests are relayed through hubs, which validate and verify the request. When new or updated content becomes available, hubs check if subscribers are authorized to receive it then distribute it.