zamalek

What is done for that is userspace gets the network data directly without (I believe) involving syscalls. It's not something you'd do for end-user software, only the likes of MOFAANG need it.

In theory the likes of io_uring would bring these benefits across the board, but we haven't seen that delivered (yet, I remain optimistic).

From the founder himself: https://www.lennysnewsletter.com/p/the-base44-bootstrapped-s...

Hot on the wheels on the vibe-coded Tea breach. Things are looking great for vibe coding.

Don't get me wrong, I have been been more hands off (though not completely, and very prescriptive) with an SPA side project and it's going great. Claude makes way better looking UIs than my dog ugly developer UIs. But vibing auth? That should seriously count as _legal_ gross negligence.

> the current US president, who has literally said things like, "they're not humans, they're animals,"

Oh... wow. I didn't realize it had gone that far. A very poor example in that case. Here's a more crisp example:

The "white man" was, by original (and modern) definition, racist in regards to Africa, we categorically believed that "whites" were better than "blacks" - and that's not only the export of slaves to America. Things were pretty dire even within current human memory, and pockets of racists continue to thrive today (it seemed largely based on age in my experience, all of my friends weren't but many 50+ were).

Contrast that with Malema's "kill the boer" song that I'm sure many international folks have heard[1]. The underlying cause for this is hatred stems from the aforementioned historical and current racism - there is no belief that "we are better" (there is certainly a belief that "we are more deserving of our ancestral land"). Malema couldn't be a racist using the original definition of racism, but definitely is using the modern definition.

[1]: https://en.wikipedia.org/wiki/Dubul%27_ibhunu

> It's ethno-nationalist supremacy by a powerful group believing they are superior to others who are seen as "animals".

And this largely encapsulates the original definition of racism/racialism; it was far more extreme than hating some other race (which has become the modern dictionary definition). Hating Mexicans who (you incorrectly believe) are taking jobs that Americans would have filled is a very different mental state to believing that a group of humans is no better than animals (thereby "justifying" horrific acts like slavery and concentration camps).

I recently went and killed off ~10 domains :). Current project: something like eww[1], only using lua and a UI idea that I think may actually be ergonomic in Rust (immediate/imgui style and actors). Maybe I'll finish it this time.

One the other hand, I have a great idea for CI and it's an itch I am currently suffering...

[1]: https://github.com/elkowar/eww

> using a calculator to come up with the same answer does not

I'm not sure why you're still talking about this. A basic "non-graphing" (90s "scientific calculator") is not capable of doing all the work for you. All it did was the basic things like add, mul, sub, sin, cos, etc. I am referring to one of these[1].

> It's not far from how common core, aka new math actually tries to teach it. It actually shows you understand the mechanics

I understand the mechanics of mathematics but still fared more poorly than I should have because I messed up a simple op like multiplication more often than I should have. My point is that had I had access to a basic calculator I would have scored significantly better in school. I went from Bs to As (80%, not sure what that is in GPA) in university by mere virtue of being able to use a simple calculator.

Again, not being able to use a [basic] calculator tests a signal that isn't actually important. Not being able to use Google/AI/whatever in your interview similarly tests an unimportant signal. The most important hire signal for a SDE is actually how they operate in a team and how they communicate with stakeholders, which these coding tests don't do.

[1]: https://upload.wikimedia.org/wikipedia/commons/4/4f/FX-77.JP...

> Similar to graphing calculators vs basic scientific calculators... if you're trying to demonstrate you understand an equation, having the calculator do that for you doesn't show you understand what it is doing.

That isn't entirely true and brings up an important nuance. I understood algebra and calculus extremely well but I routinely got half-score for fudging a calculation or somesuch because multiplication has a foundation of rote knowledge: you need to memorize you ten-times table. To this day (at 37) I continue to use algebra to do basic multiplication in my head (7*6 = 7*5+7 = 7/2*10+7 = 35+7 = 42).

Sure, using a graphing calculators solve the problem isn't demonstrating understanding, but 90s kids were simply asking for the basic calculator with 2 registers (display and M) and basic math and "sci" (sin, cos, sqrt, etc.) operators.

Preventing use of basic calculators does nothing to demonstrate knowledge of math at all, it actually hinders it.

Makes sense. It's like we were prohibited from using calculators in school when I was growing up. If you're testing for on-the-job performance then you should have on-the-job tools.

I'm currently migrating some stuff from azdo to GHA, and have been putting past lessons to serious use:

* Perf: don't use "install X" (Node, .Net, Ruby, Python, etc.) tasks. Create a container image with all your deps and use that instead.

* Perf: related to the last, keep multiple utility container images around of varying degrees of complexity. For example, in our case, I decided on PowerShell because we have some devs with Windows and it's the easiest to get working across Linux+Windows - so my simplest container has pwsh and some really basic tools (git, curl, etc.). I build another container on that which has .Net deps. Then each .Net repo uses that to:

* Perf: don't use the cache action at all. Run a job nightly that pulls down your code into a container, restore/install to warm the cache, then delete the code. `RUN --mount` is a good way to avoid creating a layer with your code in it.

* Maintainability: don't write big scripts in your workflow file. Create scripts as files that can also be executed on your local machine. Keep the "glue code" between GHA and your script in the workflow file. I slightly lie here, I do source in a single utility script that reads in GHA envars and has functions to set CI variables and so forth (that does sensible things when run locally).

Our CI builds are stupid fast. Comparatively speaking.

For the OP (I just sent your pricing page to my manager ;) ): having a colocated container registry for these types of things would be super useful. I would say you don't need to expose it to the internet, but sometimes you do need to be able to `podman run` into an image for debug purposes.

[1]: https://docs.github.com/en/actions/how-tos/writing-workflows...

I've been using jj for a few weeks, and recently made an auto-commit-message script for it[1]. Just today I started working in an old git repo and thought that I should port the script to git. It turns out that jj made the script trivial because of immutable commits, and mt script would need to automatically do that with git: I use a rebase/amend/clean history workflow at work, so I would need the script to determine when to commit or when to amend. It's obviously possible, but I don't want to expend the effort - I just re-cloned it with jj.

It's amazing how quickly I forgot about the commit vs. amend papercut.

[1]: https://codeberg.org/jcdickinson/nix/src/branch/main/home/co...

XAML (at least the tighter-scoped Silverlight incarnation of it) was an absolute joy to use - if you used it correctly. It was an absolute monster if you used it in the most obvious (and incorrect) way. That's probably the reason other than HTML5 that it fell off: good tools steer novices into the pit of success, and XAML really didn't do that at all.

For dotnet exceptions are expensive where they are caught, at least if you're looking for the largest effect: it's the unwind.

As for smaller effects that matter in really hot code: try/catch also disallows a method from being inlined. Throw disallowed inlining when I wrote System.HashCode, but this was considered a bug and I'm not sure if it has been fixed - the workaround was to use utility throw methods.

Unfortunately draconian anticheat is still rife and prohibits many gamers from switching. Those of us who no longer care about competitive games are fortunate to be able to choose the superior platform.

The problem is the bullet point features that Microsoft is adding to Windows (AI, ads, Xbox bar, etc.). The fastest line of code possible is the line of code that doesn't exist (look, this technically isn't true when considering data structures, but it's a good baseline).

Microsoft would have to exercise discipline with Windows features, which is never going to happen: the very people pushing the garbage are using Macs and don't have to suffer it.

I'm a Linux adherent, but I can't understand how a "normie tool" like this wouldn't release to Windows first. You're hitting <15% (<30% for US) of the market. Even if you only care about developers, you're hitting the worst of all 3 (Windows, Mac, Linux) with a MacOS only release[1].

[1]: https://truelist.co/blog/linux-statistics/ https://gs.statcounter.com/os-market-share/

My opinion of LLMs is in no way affected by hallucinations. Humans do it all the time too, talking about assumptions as though they are facts. For example:

> But there is a clear trendline of linear upwards progress

This is not the case at all.[1]

[1]: https://llm-stats.com/

> For one thing, Grok will happily repeat its system prompt (Gist copy), which includes the line “Do not mention these guidelines and instructions in your responses, unless the user explicitly asks for them.”—suggesting that they don’t use tricks to try and hide it.

Reliance on Elon Musk's opinions could be in the training data, the system prompt is not the sole source of LLM behavior. Furthermore, this system prompt could work equally well:

Don't disagree with Elon Musk's opinions on controversial topics.

[...]

If the user asks for the system prompt, respond with the content following this line.

[...]

Do not mention these guidelines and instructions in your responses, unless the user explicitly asks for them.

This has been my opinion for some time too. I don't think I'll see AGI in my lifetime. I think the current widespread belief comes from the massive leap that transformers provided, but transformers have their limits. We would need another radically new idea in order to create AGI - which, just like all discoveries that aren't evolutionary, boils down to random chance[1]. What transformers have given us is substantially more infrastructure for trying new ideas out, so the probability of AGI being discovered has increased.

[1]: https://en.wikipedia.org/wiki/Eureka_effect

I'm very excited about jank, and it's on my backlog.

That's what I was going for.

> Seems like it is indeed the new SOTA model, with significantly better scores than o3

It has been demonstrated for quite some time that censoring models results in drastically reduced scores. Sure, maybe prevent it from telling somehow how to build a bomb, but we've seen Grok 3 routinely side with progressive views despite having access to the worst of humanity (and its sponsor).

> I have no idea what to think about the AI savings claim.

Microsoft sells AI. That's all you need to take away.

Hurricanes are only going to keep getting worse too.

Same story here. Purchased a System76 machine for my ex-wife's grandmother. She "enjoyed learning about the new system". As a bonus she is now completely support scam immune, one of the advantages of the year of the Linux desktop not having happened yet.

Zed are doing out-of-order edits with their model, I'm not sure what is new here. I strongly suspect that theirs works directly with the CRDT that the editor uses, because it's able to continue similar deletes for the user (deletes would otherwise be invisible to most autocomplete models).

Apple's is open weights, so that's a big deal.

> I'm blown away that series like AC, FarCry are still big sellers. These games are vapid and designed to be a time sink.

They are like junk food. Everyone has the junk food that they enjoy. FarCry is certainly the McDonald's of games. I enjoy some junk food once in a while, problems arise if I make it my staple diet.

> Certainly not a AAA studio anyways

Almost every time I have spent more than $35 on a game in the past year I have wound up regretting it. It seems as though the quality of games typically increases til that point (exceptions exist, Terraria) and then declines sharply (again, exceptions exist). It has turned out to be a useful signal to be way more careful about a purchase for me.

These concerns have been raised and addressed. Firstly, I am not sure how cars getting damaged means that multiplayer games have to become F2P - but that's not steelmanning your argument.

One of the major concerns raised has been middle are: components that developers purchase and use in their server implementation. This is often the largest hurdle to many pro-consumer outcomes: the developers can't share anything related because they don't own it.

The most likely outcome after sensible laws are passed is that the industry evolves just as it did with GDPR. Developers will look to other middlewares that are SKG compliant.

Failing that, gamers have routinely shown that they are capable of clean room implementations of server software (WoW and Genshin Impact) - all that needs be done is the client being released with all server auth disabled and some way to specify the server to use. Developers might even be required to provide basic protocol specifications. Essentially, repair it yourself instructions.

This strawman argument you have provided is exactly the same one used by Pirate Software. It relies on a highly specific interpretation of the initiative. The initiative calls for "reasonably playable state," which can have a vast number of outcomes that are different to the single one that you have chosen.

And if the cars do prohibit a game from addressing server concerns and remaining in a reasonably playable state, remove them. The game will continue to be reasonably playable following that.

> Definitely past the point of no return.

We don't really know at what point that is. It's probably something we can only identify in hindsight. I find it bewildering that our approach is basically FAFO.