Ariadne Conill

Here's a piece on writing portable ARM64 assembly: https://ariadne.space/2023/04/12/writing-portable-arm-assembly.html

The author thinks that musl, Alpine, and Linux are right and everything else is wrong. * https://ariadne.space/2022/03/27/the-tragedy-of-gethostbyname/ - musl's DNS resolver doesn't support TCP lookups, but it's your fault for expecting the standard DNS lookup functions to work. You should change your program to use a third-party library for DNS lookups. This post is a couple years old; musl now supports TCP lookups. * https://ariadne.space/2021/06/25/understanding-thread-stack-sizes-and-… - Alpine has a stack size that's substantially smaller (128k) than other widely used OSes (512k+). Your program may work everywhere else, but you're wrong to assume a reasonable stack size. Here's how to jump through a hoop and make it work on Alpine.

Cat is an interesting example because it's not as trivial as it seems, especially when it comes to performance and resilience: https://endler.dev/2018/fastcat/ https://ariadne.space/2022/07/17/how-efficient-can-cat1-be/ I think this is one of these "not built here" discussions. It's very easy to underestimate the number of edge-cases to support especially in the case of network transfer. You start implementing your own, then you've got to handle non standard certificates, 300 redirects, temporary failures, backoff strategies, optimizing chunk size, handling proxy calls, etc. Using a standard library makes sense.

Standardizing social interactions between humans and making them immutable by default is appealing to SV adjacent people because it makes it easier to collect and process data and sounds good in front VCs. It is negative value to actual users who have lots of open information exchange standards at their disposal already and can just... use them. Related: https://ariadne.space/2018/11/12/pleroma-litepub-activitypub-and-json-…

Based on my reading of [1] and [2] it sounds a bit more nuanced than "just incorporate s6 into alpine". [1]: https://ariadne.space/2021/03/25/lets-build-a-new-service-manager-for-… [2]: https://skarnet.com/projects/service-manager.html

Some more information in this introduction post: https://ariadne.space/2021/09/09/introducing-witchery-tools-for-buildi…

I think you mean glibc 2.34: https://sourceware.org/pipermail/libc-alpha/2021-August/129718.html You opt-in by building with -D_TIME_BITS=64: https://www.gnu.org/software/libc/manual/html_node/Feature-Test-Macros… An approach not without its problems: https://ariadne.space/2021/12/29/glibc-is-still-not-y2038-compliant-by… - but then again software that relies on time_t being 32 bits has been broken on 64-bit architectures for a very long time already, surely? I suppose I could have written that "GNU/Linux values ABI stability" in my comment, instead of plain "Linux"... ;)

Url seems to have an extra query parameter “s” that shows an error page. Correct url should be https://ariadne.space/2022/01/18/delegation-of-authority-from-the-syst…

And one big one just happened this year: https://ariadne.space/2021/06/14/the-end-of-freenode/

https://ariadne.space/2021/07/14/oracle-cloud-sucks/ this person had their account terminated for the crime of running an IRC server.

> The initial/main thread stack size is dynamic That seems doubtful, AFAIK all OS have a static allocation for the main thread, just one that is commonly (though not necessarily) much larger than what they give off-threads, as https://ariadne.space/2021/06/25/understanding-thread-stack-sizes-and-… quotes Windows gives 1MB to all threads, while macOS or OpenBSD 4.6+ give 8MB to the main thread and 512k to off-threads. The former would not merit consideration if it was automatically increased. > it grows until it runs into another allocation Which it won't because vmem. > and these days there are guard pages and such That just makes the access segfault if the stack allocation is not large enough to straight jump over the guard page.

I'm not an IRC person, so I have no idea myself about any of these things, but I notice that this blog post links to an earlier post where the author speculates about what's driving these changes: https://ariadne.space/2021/05/20/the-whole-freenode-kerfluffle/

There is some speculation about the motivations here: https://ariadne.space/2021/05/20/the-whole-freenode-kerfluffle/

> “Nginx is an incredible software and platform, but I was wondering if it wouldn't be time to just bite the bullet and create a more modern solution based on Rust.” I actually don't understand why I am seeing arguments like this all the time. Rewrite takes time and effort. It may be OK for new projects. But for a mature project that has a quarter million lines of code and load-tested for cross-platform compatibility and security, it is way more than "just biting the bullet". (Not to mention that memory safety in Rust is ... still improvable. [1]) Rust is not a panacea. [2] When a corporate tries to seize control of a open source project, some one (or some entity?) has to do something, and rewriting in Rust by no means reclaims the freedom and openness of the project. [1] https://news.ycombinator.com/item?id=39440808 [2] https://ariadne.space/2023/12/07/most-breaches-actually-begin-in-corp/

I am using that expression to separate them from the broader Open Source movement. I do think there is a major difference between general open-source and GNU/FSF "free as in freedom" ideology. I see the broader open source movement as "hey I made this thing to get stuff done, oh btw it's open source under a permissive license so feel free to take it/improve it". GNU/FSF is more like "here's a list of reasons why the way you do your computing is wrong, the right way is to go back to the dark ages of computing, oh and we have some half-baked software to help with that - and don't you dare use that other open-source thing over there, it's not free enough". > Name some examples The hundreds of different Linux distributions, none of which can match the stability, consistency and user-experience of proprietary OSes even from a decade ago? Imagine the collective effort that has been wasted reinventing the wheel and bikeshedding. The PinePhone, which despite there being a permissively-licensed, battle-tested mobile userspace (Android), would rather try to adapt a desktop-focused userspace to it (and the effort is obviously spread across multiple distros, because why wouldn't it?), and as a result you have a device for sale today that's less usable than a PocketPC from 2003. The whole "Respects Your Freedom" list from the FSF (https://ryf.fsf.org/index.php/), which looks like a low-budget computer parts catalog from a decade ago, or their absurd position against CPU microcode…

https://ariadne.space/about - she's pretty cool

Good for Debian! Related links: https://en.wikipedia.org/wiki/Year_2038_problem https://www.theregister.com/2020/01/30/linux_5_6_2038/ https://opensource.com/article/19/1/year2038-problem-linux-kernel https://www.gnu.org/software/gnulib/manual/html_node/Avoiding-the-year… https://ariadne.space/2021/12/29/glibc-is-still-not-y2038-compliant-by… Are manufacturers of IOT devices using Linux whose devices have a lifespan of 15 years thinking about this yet? I doubt it. Where is the startup for a consumer-visible certification body that ensures my device will fully function during Y2038?