Red Rozenglass

a young man from a far away land.
corposlave ("software engineering consultant" lol).

GNU/Linux (Slackware). Programming (C, sh, Lisp, Lua, Web, JS, etc.). Likes writing documentation, understanding & maintaining systems, archiving, organization. Interested in paper machines, books, visual novels, writing, drawing, anime. EN/AR fluent, TR can order food, JP early learnings.

RƎD.

Public Key

npub1aj05n6h74ys35hfujyddczfcmr4drhs79xtrfdxw8sutdq2gnw9q994pgc

NIP-05 Address

rozenglass@fedi-dreamscape-link.mostr.pub

Profile Code

nprofile1qqswe86fatl2jgg6t57fzxkupyud36k3mc0zn935kn8rcw9ks9yfhzspz4mhxue69uhhyetvv9ujumt0wd68ytnsw43qlze2l8

Publishing to

relay.mostr.pub

Author Public Key

npub1aj05n6h74ys35hfujyddczfcmr4drhs79xtrfdxw8sutdq2gnw9q994pgc

Show more details

Last Notes

2025-04-02 20:13:29

- reply

@nprofile…jgyg That is not lisp. You lost homoiconicity, and you're introducing large amounts of syntax. That is not the lisp way. A lispy implementation, I would argue, would look something like this:
<list>
<atom>defun</atom>
<atom>frob</atom>
<list><atom>x</atom></list>
<list>
<atom>+</atom>
<atom>2</atom>
<atom>x</atom>
</list>
</list>
This way, if you write macros to manipulate code, you don't care that the lists are function calls or arglists or defuns, they're just lists and atoms, which is an important property of lisp languages; homoiconicity. A minor concession might be made perhaps for strings:
(setq name "John Smith")
<list>
<atom>setq</atom>
<atom>name</atom>
<string><![CDATA[John Smith]]></string>
</list>
Macro stuff, special forms, quoting and unquoting (i.e. ` and , ) might need their own elements too, stuff like:
`(set-age user ,(1+ old-age))
becomes:
<quote>
<list>
<atom>set-age</atom>
<atom>user</atom>
<unquote>
<list>
<atom>1+</atom>
<atom>old-age</atom>
</list>
</unquote>
</list>
</quote>

2025-04-01 20:29:49

- reply

@nprofile…w8cy
names are integral to any level of magickal work in the world
yup, curious they thought only man and god had the ability to name, not even angels did.
personification of the devil
simply the necessary opposite (or shadow) of god
in islam, satan and his jin followers play the main role of psychically "whispering" to people. the human soul is capable of evil will on its own, but satan's jin guide mankind towards temptations and evil, they can enact almost nothing in the physical realm, but every day and at all time they inspire the thoughts in the minds of men, eternal whisperers, kept at bay only through man's fasting and prays in remembrance of god.
iirc muhammad said that satan sits "in imitation of god" on a throne floating over water, and his legions of jin come to report to him, one says "i convinced man to follow his greed today! he stole!" and satan says "go away, you have done nothing!", another comes and says "i made man so angry and resentful he killed another!", satan tells him "go away! you have done nothing!", then another comes to him and says "i never left him, until he separated from those he belonged to!", satan is like you're the goodest boy! and orders him closer to sit by his side, cause he "gets it".
i wonder if his throne is floating over water because god promises a fire below to those he don't like. not sure.
i prefer revering rather than worshipping
not sure what you mean by reverence vs worship.
is it about an internal feeling vs external motions?
deities exist simply to be worshipped
my current think is that deities exist to personify psychological patterns that already exist in humans, because somehow making them into personalities, instead of explaining and discussing them intellectually, inspires more action and understanding in people, and, in a slightly magical way "amplifies" those spirits projected by the human into the world around them.
so it makes sense that a selfish abrahamic god would exhibit much of the same desire
islam god says he don't really care and don't need anyone or anything, plays the tsundere card basically, it's not like i like you or anything, silly baka slave, i just love you, but you're nothing to me! slightly curios that the "tester" god, e.g. in islam, sounds more like someone training neural networks, generations of random "try everything" churn, to pick models that most closely align with whatever the trainer wants. the trained wouldn't see the purpose, nothing would make sense, and to them life would feel extremely oppressive, where you work and work only to be punished continuously and die, but hopefully you managed to exploit the genetic algorithm to replicate parts of yourself into the next generation of sufferers.
just like "to know" in the bible is euphemistic of sex
in quran, the specific act of sex is referred to as "coming unto your tilth", or just "coming unto" women, in general. quran also mentions sperm (or a drop of cum) sometimes, women's period, not sure he'd be shy of saying penis if women were made of it. plus, in arabic the word rib also means "foundation", or the "diagonal" in geometric shapes...
wait... i see what you mean; all of those can easily translate to penis too actually...
you're coming unto something here, more than just tilth i think O.o

2025-04-01 19:10:30

- reply

@nprofile…4duy That bureaucracy sure was a ride, phew~ Congratulations! ^-^

2025-04-01 09:53:52

- reply

@nprofile…w8cy in islam, god made adam to be his "successor" in earth, the angels objected that adam is the kinda creature that will destroy the earth and desecrate it with spilled blood, god basically says "i know what you don't know" and continues anyway. he seems to want "companions" or something, and through the suffering of life he will test the decedents of adam, to find ones who he will enjoy being around. before sending adam to earth, he teaches him the "names of all things", or maybe, /how to/ give things names, which is apparently something crazy cool for all other creatures made by god, including the angels. god tells the angels to name a bunch of things, and they say lord we only know what you taught us, and then he asks adam to name them, and he does, and all the angels are shocked in awe and bow down to adam, except satan, who objects i'm still better than him i'm made of fire! he's made of dirt! i'm most loyal to you but he will be the biggest jerk of all and will do all the bad things and if not you know what i will make sure he will! and god is like okie imma make you immortal until the end of time and you will guide the children of man away from the path to me, except the ones of them who will actually be my most loyal slaves/worshipers (slaves and worshiper is the same word in arabic kek), i will take those, and burn the rest.
so i guess, in islam god doesn't want peasants, he wants slave companions or something.
eve? god just thought adam would feel lonely so he took a chunk of his rib (ouch) and created eve out of it so adam can have some company XD

2025-04-01 09:31:46

- reply

@nprofile…dp7l also, "White Woman Melty Blood" XD

2025-04-01 09:30:26

- reply

@nprofile…dp7l Arcuied in EYE Divine Cybermancy. Massive fanfiction potential there o.o
https://fedi.dreamscape.link/rozenglass/s/160a4cccd885b29113a4f207d9dfb543.png

2025-04-01 09:10:19

- reply

@8qse4sxjxs@minidisc.tokyo the lawful neutral police; 'blkid' and 'mount'.

2025-04-01 09:03:48

- reply

@nprofile…dp7l Tsukihime L4D mod sounds like something the world needs indeed. Thank you for your service~

2025-04-01 09:01:31

- reply

@nprofile…dp7l this has been an account of man. with the rise of man, man had nearly lost his sanity. man realized in that split second, he had made a mistake. man forever will symbolize world war 2 looming.

2025-03-28 17:10:43

- reply

@nprofile…qr5r
throws an Out of Memory error, obviously :3

2025-03-28 05:54:51

- reply

@nprofile…gg8z Indeed. Homebrew was the first thing I installed of course :) What I remember as the most frustrating thing was the window management of OSX. No matter how deep I went down the rabbit hole of gestures and shortcuts, it still didn't click for me. I tried to replace the OSX window manager iirc too. At least had Emacs, where I spend most of my time, and that felt home enough. But at that point, OSX becomes just overhead for my setup of a the tiling window manager, Emacs, and a Linux VM for running the backend software I'm developing.

2025-03-27 14:14:29

- reply

@nprofile…gg8z A lifetime ago, I was an employee at a company, doing web backend development, they handed me a 2012 Mac Book Pro. Thinking it was just UNIX so how bad could it be, I tried to use it for three months, but suffered in pain, slowly replacing everything for even slightly improved workflows. The savior to my productivity was installing a GNU/Linux distro in a VM, making it full-screen on boot, and just using that for everything. Later on, I managed to get approval to install a Linux on the company laptop. That Mac Book to this day has Slackware on it. Was also the first time I hacked on Linux kernel modules to get the buggy WiFi drivers for the Mac to behave. Good ol' times.

2025-03-27 03:44:41

why are poast nitter links in wikipedia's references now...
https://en.wikipedia.org/wiki/Reza_Pahlavi,_Crown_Prince_of_Iran#cite_note-51

2025-03-26 15:07:31

we have to use $latesthypeframework
why reinvent the wheel on solved things
countless smart people already built it for you
next.js critical cve:
you can just say "no plz" to authentication
when will we ever learn.

2025-03-26 14:00:39

- reply

@nprofile…k9rm @nprofile…w8cy if you or someone else might later be curious:
https://libslack.org/daemon/
https://libslack.org/daemon/sources/daemon.c
daemon.c depends on basic functions from https://libslack.org/
nor do i know how it's handling all those features you describe.
chroots
oh, it handles them just by literally calling it. the --chroot option just calls chroot(). changing users, chrooting, setting environment variables, redirecting logs, it just orchestrates those basic functions, does them in the right order, if you request them, and nothing much more.
custom environment variables
again, just convenience, -e adds an env var to the child's environment, because by default, raf daemon cleans up the environment for the child process. you can use -i to pass the entire external environment (all environment variables) to the child if you want, by i don't do usually.
separate tools in daemontools-style supervision
based indeed, but then comes the orchestration of those tools. daemon's man page basically just says: "There are many tasks that need to be performed to correctly set up a daemon process. This can be tedious. daemon performs these tasks for other processes." and that's the goal. it doesn't re-implement chroot(), it just calls it at the right time.
you say log files too; does it rely on a logrotate daemon?
each supervisor instance handles the output of its child, and sends it where you configure it to, either a syslog daemon, or to a log file. i just use plain log files for everything. and no, it doesn't rotate log files on its own, i use logrotate like a dirty whore :3
https://skarnet.org/software/s6/s6-log.html#diesyslogdiedie sums up issues
interesting read, thanks.

2025-03-26 13:28:32

- reply

@nprofile…k9rm @nprofile…w8cy the daemon stores its own pid in a file named by a unique service name, after it locks the file, if a simple getpid(2) call followed by an fcntl lock, then a write(), fails to write the processes's own pid to a file, then... then... you know what? i wouldn't be surprised. this is unix/linux after all, everything is an organically-grown pile of garbage :p
but if it's any indication, i've been running dozens of long-running slackware machines and virtual machines, for years, for personal use and businesses, and i don't know of a single case where such an incident happened with raf's daemon. maybe it's not too bad after all.
thank you for the discussion. i wish you an acceptably good day ^-^

2025-03-26 11:36:44

- reply

@nprofile…w8cy @nprofile…k9rm raf daemon does that, it supervises the child, and restarts it correctly on crash 100% of the time in my experience, even handling crash loops, etc. not to mention setting up output redirection to log files, control commands, chrooting, custom environment variables, etc. raf's code seems to be well concerned with correctness in my experience.

2025-03-26 11:30:56

- reply

@nprofile…k9rm @nprofile…w8cy i see, true. let's say we want to restart a service: svc finds the service by directory name, opens a control file, writes the command to it, a supervisor instance is listening on that file (a fifo pipe), reads the command, sends signal to its forked child; we have a supervise instance per service (started by svscan), each of them forking the single service they oversee.
in contrast to the traditional slackware's raf daemon program, which also has one instance of the parent supervisor per forked child service, but when the raf daemon management command (equivalent to svc) is invoked by the user, it reads the pid of the supervisor (not the child service itself) from the named supervisor pidfile, and sends the signal to that supervisor, which then handles the signal, and sends it accordingly to its forked child process, to which it holds a pid in memory, just like djb supervise.
so, in simple terms, the djb service manager sends commands to the supervisor using text characters over a fifo pipe it finds and opens by directory name, while the raf manager daemon instance sends commands to the supervisor instance using a unix signal after it finds and opens the pidfile by name.
manager -> supervisor, using char over fifo opened in a named dir.
manager -> supervisor, using signal to pid read from a named file.
both afterwards do supervisor -> child service, using a unix signal.
i still don't see it. what problem exactly does djb's method fixes compared to raf's? djb's faq isn't helping, nor reading the code either so far. i would appreciate an explanation of a scenario where it breaks, or a link to something to read on that. i'm curios :)

2025-03-26 10:29:58

- reply

@nprofile…k9rm @nprofile…w8cy
>pid files
disgusting
daemontools-style supervision is the only
i opens daemontools code to check what else he do.
supervise.c basically:
open_trunc("supervise/status.new");
write(fd,status,sizeof status);
rename("supervise/status.new","supervise/status");
so, yeah, just a pid file too, except called status, and has a couple of extra flags.
but you got me though. now i ended up reading beautiful djb code,
and i like the pleasant feeling in my eyes.
maybe one day i will use daemontools, s6, and runit, for extended periods just to see if i'm missing out on anything. trying them on the surface level doesn't feel too interesting, and i have 0 issues running slackware's init setup, so i don't feel the need to change. one day maybe~

2025-03-26 09:59:32

- reply

@nprofile…k9rm @nprofile…3j8n Yes, no. I don't want 1 to 2 million lines of update-required every few months months bloat code running on my personal machines just to route some desktop applications through a different VPN; I suffered enough already maintaining Docker & Kubernetes clusters at work for almost a decade, I paid my penance. Not to mention it would likely end up routing the traffic through the Docker router and Kubernetes networking plugins; a needless overhead and source of the nastiest bugs in my experience. And I only want network namespacing for my use-case, nothing more; I will be fighting against every other feature those container engines provide.
I'll stick to my init script that calls the ip and wg commands a few times, then works forever :)

2025-03-26 07:55:41

@nprofile…xvh7 Cool website, fellow Lainon Architect.
May the soul of the wired reward your handiwork, with most blessed bytes.

2025-03-26 06:30:46

- reply

@nprofile…k9rm Yes. 99% of my init needs are met by running daemon[1] from rc.local at boot; the default method on Slackware. With three lines of shell script I get a named daemon, with PID file tracking, executed by its own user, optionally in a chroot, and with stdout and stderr redirected to log files.
And for the 1% when I want something crazy, like running multiple network namespaces for different services hosted in them, connecting over specific Wireguard networks each, and isolated from seeing the actual network hardware directly, I can modify the simple networking init scripts, and make it happen myself in a couple of hours. I don't even want to imagine what it would be like to try to modify systemd and NetworkManager code to add such a feature.
My experience with systemd was very buggy, in production, causing extended down-times, from bugs left unfixed for years. My experience with OpenRC was alright, but does not spark joy. Only caused production downtime once, due to a buggy interaction with consolekit2, but that's a track record roughly %600 times better than systemd already.
[1]: https://libslack.org/daemon/

2025-03-26 03:00:14

- reply

@nprofile…dp7l Indeed :)
https://fedi.dreamscape.link/rozenglass/s/6c76d18ca50325ab14369663f9aedfda.jpg
https://fedi.dreamscape.link/rozenglass/s/26b561936372e81d2b8822ed3d4cdd66.jpg

2025-03-26 02:48:56

@nprofile…73zh Usernames with formatting characters in them, like @test_one_two@example.com are broken, because one get italicized. I patched this up on my instance, seems to be working well. Here's the patch: https://dreamscape.link/vault/public/debug/snac2/0004-Fix-markup-being-parsed-inside-user-IDs.patch
Here's the same patch in a code block, because why not :)
```patch
From a5888ee164437de3854e360178db0bedf5f0896a Mon Sep 17 00:00:00 2001
From: Red Rozenglass
Date: Tue, 25 Mar 2025 22:39:24 -0400
Subject: [PATCH 4/4] Fix markup being parsed inside user IDs
A user called @test_one_two@example.com would have
the "one" part of their name italicized with an
HTML element, which breaks the ID, and prevents them
from getting actually mentioned; they don't get the
message or the notification.
Added regex to split user IDs during the message
formatting stage. User ID segments do not match any
of the formatting rules, so they get copied as is,
fixing the issue.
format.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/format.c b/format.c
index 755aeb2..dc4c673 100644
a/format.c+++ b/format.c
@@ -91,6 +91,7 @@ static xsstr *formatline(const char line, xs_list attach)
/ split by markup /
xs *sm = xsregexsplit(line,
"("
+ "@" NOSPACE "|"
"[^]+`" "|"
"~~[^~]+~~" "|"
"\\*\\*?\\*?[^\\*]+\\*?\\*?\\*" "|"
--
2.39.5

2025-03-26 02:01:38

- reply

@WitchHunterSiegfried@clew.lol Ah, Blood Angels, the glow on the glassy droplets made me think it's Aeldari. Thanks for the sauce.

2025-03-26 01:26:27

- reply

@WitchHunterSiegfried@clew.lol You are not a waste of space, you are just as necessary to the flow of this story, even if you do not know it yet. Also, the Moon Princess in red Space Marine / Eldar hybrid armor after a meal of blood, is a blessed sight to my eyes, I never knew I needed.

2025-03-25 10:43:20

- reply

@nprofile…dxfh I see, on Slackware this basically would be:
build packages, mount USB drivecp /my/repo/of/updates/*.t?z /mnt/usb/
mount USB to new machineupgradepkg --terse --install-new /mnt/usb/*.t?z
This applies equally to official Slackware packages, and any custom packages, from SBo or otherwise.
Of course on Debian you can do the same, with apt install /mnt/usb/*.deb. I think if you do it this way, the dependency resolver of apt won't go crazy, because it can order the .deb installation to keep dependencies satisfied. Do not try to install the packages one by one though, because you will have to do it in the correct order then.
Unrelated recommendation, if you're thinking about Debian, check out Devuan, for a more sane community, and a more sane init system. Devuan is still down-stream of Debian, so Debian shenanigans might impact them, but at least you have a layer of sanity-checks between you and the Debian project.

2025-03-25 10:33:56

- reply

@nprofile…kzhg Static binaries are great and all, even dynamic ones, as long as they don't rely on 658 dependencies, and don't require you install latest everything, etc. But the greatest way really is to be social, and actually collaborate with distro maintainers to get your software officially adopted by them. And as a bonus, you get your project in order too, following best practices, security, etc. A lot of developers just give up on the idea, and try to provide their packages externally to "workaround" the normal way. People use Linux package managers /because/ they need a 3rd party (the maintainers) between them and the developer of the software, to add a layer of trust in the middle. Trust that is very hard to gain from scratch otherwise.

2025-03-25 10:25:12

- reply

@nprofile…dxfh With respect to the offline mirror, I see that as very possible too. If you want faster setup of machines too, pkgtools (installpkg et al.) support a --tagfile option, in which you can tell them to basically just install a specific set of packages. I have tagfiles for tiny light weight containers, for minimal secure web servers, and other arrangements. All of this is a bit advanced of course, considering you have to know your way around manual dependency management. Some variation of ldd * | grep 'not found' can be your best friend :)

2025-03-25 10:20:08

- reply

@nprofile…dxfh I have a handful of Slackware 15.0 machines running grub, it works for me, including in complex setups where I boot a custom initrd from a software RAID disk, then SSH into the initrd to unlock a LUKS partitions and set up the rest of the RAID arrays, and then boot the actual system. The upcoming Slackware 15.1 will encourage grub as the default configuration instead of elilo / lilo, but 15.0 definitely works well with it in my experience.

2025-03-25 10:11:56

- reply

@nprofile…dxfh I mirror SBo using git. My package management functions are basically some ease-of-life around the core of:
# As normal user:
cat *.info slack-desc README # read about the package
. *.info # source its metadata
wget -c $DOWNLOAD # download the source
md5sum filename.tar.gz # check-sum the downloaded file
echo $MD5SUM # compare to the expected value.
fakeroot sh *.SlackBuild # build the package.
# As root:
updatepkg --install-new /tmp/*_SBo.t?z # install the built package.
chown root:root /tmp/*_SBo.tgz # make it owned by root
mv /tmp/*_SBo.tgz /var/cache/packages/slackbuilds/ # store it in my package store.
I also share the pre-built packages between my machines in general, by mirroring them to one of my servers, and downloading them to other machines from there.

2025-03-25 10:03:42

- reply

@nprofile…3hg5 @nprofile…dxfh The point of Slackware, in my mind, is that it is so simple and unchanging that you can actually own your system over time. My primary reason for using Slackware is that I can fork the whole distro if I really had to, because it is a one-person-sized system. People complain that "Slackware is bloat, it installs hundreds of packages by default", but that is actually the simple approach, because it means Slackware doesn't have repos with 60,000 packages, with unlimited potential combinations. Slackware is rock solid for that. There's only one authoritative configuration, and it is developed and tested for half a decade before each release.
CRUX is very good too, but, I strongly disagree with their "only English" and "delete all docs from packages" mindset. I think CRUX is more invasive philosophically than Slackware in trivial things, yet somehow more disorganized and haphazard at the same time.
How do I do SBo package updates? I wrote ~15 lines of shell and awk script in a shell function, starts with something like "for file in /var/log/packages/SBo" (packages install logs are plain text files, and require a "tag" part in them, SBo is a tag for example). It uses git pull to sync a local copy of the SBo tree. The end result is that it shows me "package x has update entries in the ChangeLog, currently installed version is vx.y.z, here are the 5 git log entries that modified this package in the package tree". I then go over my packages, and decide whether I want to update them or not. I don't update most things, unless I know them to be security sensitive, or I care about new features.
I've made this "update system" in a few hours, fixed a few bugs a couple of times here and there, and used it for years afterwards. It never breaks, it never changes, it gives me all the agency. Well, not really, because it still relies on package maintainers to update their packages, which I don't like. So, I wrote my own RSS/Atom/OtherFormats reader in some 100 lines of shell script, curl, and xslt style sheets, throw it on a cron, to check package releases for critical packages for me, print any updates whenever I start a new shell. I don't rely much on SBo maintainers for packages critical to business production machines (yes, Slackware in prod, serving millions of customers for my clients). Plus, I maintain half a dozen packages on SBo officially, so I have to know about them first :)
All of this is possible because Slackware doesn't have a system. It doesn't enforce things. It trusts you to know what you're doing, and commits to not breaking your work. I rewrote the networking init scripts because I wanted a weird setup with my private networks, and that was easy, and never broke ever on its own. With Slackware I can compound my work building things, I don't have to continuously "churn" trying to keep up. It's what Common Lisp / C / Lua are to the JS+NPM / Rust+Cargo (et al.) never ending slop generators. Even official package updates ask you, with a diff, to judge the new updates, instead of just overwriting your files, unlike some other unmentionable distros.
I worked deeply (build packages, maintain them, etc. for years) with Arch, Gentoo, Debian, and Fedora, worked with FreeBSD, OpenBSD, NetBSD to port software to them, and I've used CRUX for a while. Slackware is definitely the most UNIX of Linux systems. Heck, it might be more UNIX than FreeBSD. I think CRUX, OpenBSD, NetBSD, and Slackware are closely clustered, philosophically, in the idea space.
I maintain some 20+ Slackware machines, for home use, software development, playing video games, including wine / Steam / VR gaming, self-hosted LLMs, business production (web servers, databases, automation workers, etc.), and a few personal machines and laptops for less-techie family members.
If you have any questions on Slackware, I'm happy to answer.

2025-03-25 05:02:30

- reply

@nprofile…fy5w It's beautiful. Craftsmanship struggles :)

2025-03-24 16:17:54

In other news, why did no one tell me about xxd -i before:
-i | -include
Output in C include file style. A complete static array defini‐
tion is written (named after the input file), unless xxd reads
from stdin.
For example, you can add this to your build pipeline, to turn a string to an embedded array of characters. Just #include printf-template.html.h in your code. If printf-template.html contains printf control sequences, like %s, they do work if you pass them to printf(). You can have an HTML file parametrized with printf() for example.
xxd -i printf-template.html > printf-template.html.h
I assume you can embed anything this way; strings, images, 3D models, etc. This seems cross-platform enough. I'm starting to wonder if the new #embed C standard proposal is really needed, or just bloat now. (Even parameterized #embed sounds like something replaceable with some UNIX piping before you pass the data to xxd -i).

2025-03-24 16:06:33

Did some silly overhead benchmarks, mostly with Apache (v2.4.62, with mod_mpm_event), but some Node.js (v18.20.6) and NGINX (v1.26.3) stuff is provided for reference. Tested direct serving of files from disk, as a baseline, compared to CGI, FastCGI, SSI, writing a custom Apache module, Node.js reverse-proxied, and plain Node.js without SSL bypassing Apache entirely, and finally, serving plain files off of disk with NGINX default configurations. In all cases I served a JSON string of around 4KiB in size, pre-serialized. All done on a 2vCPU cloud KVM VM, Slackware, with 4GiB of memory, over the internet, with 75ms to 80ms in ping latency.
The results, and my thoughts, for the curious: https://dreamscape.link/benchmarks/apache.html

2025-03-22 18:07:18

- reply

@nprofile…qr5r @9awy7u3l76@astolfo.social now, if you liked that (you absolute masochists), do it again, but with Vulkan and OpenXR, together.

2025-03-22 17:17:04

- reply

@nprofile…cjya @nprofile…6yr6 As long as it's not chocolate, apparently chocolate is toxic to dogs, and almost every other creature except humans, cattle, and rodents... Sounds like the universe is trying to tell us something... 🤔

2025-03-22 17:12:34

@nprofile…rxcu @nprofile…sf09
Thank you both for the good discussion and perspectives ^-^
I'm likely to sleep soon, so, I wish a good day / night to both of you.

2025-03-22 17:09:22

- reply

@nprofile…r66w For Android, I used to use Librera[1], which is a reader, but also has some nice library organizational features. Combined with Syncthing. But, yeah, sounds like Calibre[2], mentioned by other fedipeers here, is closer to what you want. Well, good luck either way ^-^
May your shelves be organized, friend, and your libraries, comfy :)
[1]: https://f-droid.org/en/packages/com.foobnix.pro.pdf.reader/
[2]: https://calibre-ebook.com/

2025-03-22 16:41:42

- reply

@nprofile…r66w I throw it all in a syncthing folder hierarchy, to keep it accessible "everywhere", but I plan what I want to read a day or two in advance, print it all on A4 paper (yes, including epubs, after I pandoc -> latex them, and fix them up a bit, make the font small, two-columns, fit as much text on one page as I aesthetically can). Then carry the paper in a folder in a laptop bag with me everywhere o.o
now that I think about it... probably not what you're asking for :p

2025-03-22 16:03:25

- reply

@nprofile…xcgm Sorry I couldn't help. I mainly use Firefox currently, with JS disabled, so no animations there. I visited the site with ungoogled-chromium, but I think I haven't updated it in a while, so it doesn't seem to have those animation APIs either; there was no animations for me. I went around with Firefox a bit, and found it a pretty and comfy website. Things seemed to be working smoothly, even without JavaScript, so thank you for that ^-^

2025-03-22 14:53:34

- reply

@nprofile…02dt @nprofile…w8cy I'm running it on a 256MiB of RAM, no swap, proxying all media from other instances, and until today, no browser caching. So I would be playing a video, open another tab to my timeline, it loads a dozen images and gifs, maybe click another video or something, and poof, dead, restart. Happened rarely though.

2025-03-22 14:31:17

- reply

@nprofile…02dt For me, it only "crashed" a few times when killed for out of memory conditions, mostly when proxying multiple big media (videos) through the instance. I don't use any relays though.

2025-03-22 13:49:13

- reply

@nprofile…dklr @nprofile…5k85 @nprofile…e0ug Light initial testing, it seems to be working okay with snac so far, this post is from inside husky :D

2025-03-22 13:42:32

- reply

@nprofile…5k85 @nprofile…e0ug @nprofile…dklr Oh I see, thank you ^-^
Time to try it with snac2, if it works enough, I will recommend it to my wife :3

2025-03-22 13:27:39

@nprofile…dklr hey~ what happened to husky? o.o I just went to install it, but noticed that you were not maintaining it anymore. I haven't been in the loop on fedi for a couple of years, apologize for my ignorance.

2025-03-22 10:53:10

- reply

@9q9fhqziw8u8lh9x@fedi.shorks.gay As I understand, sadly, it has to be a "meta" play; game theory stuff. The automated scrappers are already wasting massive amounts of resources, far dwarfing any resources used by normal users. Proof of work means normal users will waste a small amount of resources, while automated scrappers would find the scaling costs astronomical, so the scrappers /have/ to stop. Normal users end up wasting and polluting a lot, to make the scrappers lose hope and stop wasting and polluting billions of times that "a lot".
I had to deploy proof of work to protect my business clients so many times, and every time it hurts my soul to know I'm wasting resources and human life time by doing it, but I cannot see another option. It's a classic "we can't have nice things" / tragedy of the commons situation. Some people are just jerks and are willing to destroy so much for but a tiny little bit of gain. Now imagine the gains they're dreaming of when they hallucinate about their own personal AGI, and how much destruction they're willing to inflict to get to it. As a sysadmin I have to somehow stop /that/.

2025-03-22 09:31:58

- reply

@nprofile…hmru someone has to run on a platform of "medieval castle for every american" they will get all votes and blow all other political parties out of the murky mud they're in.

2025-03-22 09:21:58

- reply

@nprofile…hmru 10 years later, the average is 4 years later... isn't that the opposite then? houses are getting older, because as time goes forward, their average building year is lagging behind? im confus.

2025-03-22 09:02:25

- reply

@nprofile…agan pretty. good job! ^-^
curious though, you had a whole game in-engine, with tons of animation for everything, what makes this one require special work?

2025-03-22 07:25:04

- reply

@nprofile…jf59 Too many people, unfortunately, felt so little love and support that the idea of hope and kindness is fantasy fiction to them. It's scary really. On the other hand, people don't realize how low the barrier to helping others is. The standards are so low that the smallest act of help can be shocking and life-changing to many.

2025-03-22 07:07:57

- reply

@nprofile…xcgm I'd rather those features be implemented well by the browser devs, than badly again and again by frontend devs, with all the blocking on mainthread and terribly congested scroll events that they are so prone to do. But I also can't help but feel that "fat browser eats all features nom nom, moar millions of liiiinesss" is not the world I want to live in either. But I guess that ship has sailed a long time ago.
Also, I like that disabling JS usually stops pages from animating, because all the animations distract me and hurt my eyes. A quick look at the docs for those new browser APIs don't show anything related to accessibility or the ability to disable those features client side. I hope page transitions and scroll-based animations don't become a required thing in my life in the near future.

2025-03-22 06:58:04

- reply

@nprofile…v27d I don't know what discussion you've been through, but I feel the sentiment. As someone who lived my first 20 years under a soviet-inspired national socialism dictatorship, where kids at school wore military uniforms, and secret police can make you and your family disappear at night if you say one wrong thing, I find the discourse in the US absolutely ridiculous. All sides are escalating endlessly out of /fear/; they attribute the maximum emotional response to the smallest acts of their opposition, and use that to justify even bigger responses from their side, in a vicious, ridiculous, cycle.
It keeps going no matter what side has the power, and no matter what "$current_crisis" is. It feels like the entire society collectively has a severe case of "calibration out of wack" to what evil really looks like, so they think every resistance they face in getting what they want is "real evil". The entire country is having a continuous panic attack, and no one at the leadership level has the incentive to stop it, because if they try to stop the panic first, their opponents would "win". Heck, even the good old trick, of having your entire people focus on an external dangerous unifying enemy monster, doesn't seem to work anymore in the US. Maybe it was used one too many times and people grew a tolerance.
I have my suspicion, and fear, that democracy, anywhere and everywhere, will eventually implode unless making people intentionally anxious and afraid, for views or votes or power of any kind, fraudulently, becomes an outdated strategy that doesn't work from a game theory perspective. How that can be achieved, I don't know. Fear after all is one of the oldest and most powerful motivators of humans, to act together, since time immemorial.

2025-03-22 02:01:09

Added browser caching over my #snac instance, for better experience with all the files being proxied from other instances. Apache configs (requires enabling mod_headers if you don't have it already):
<If "%{REQUEST_URI} =~ m#^/[^/]+/(s|x|y)/.+$# || %{REQUEST_URI} =~ m#^.*\.ico$#">
Header set Cache-Control "max-age=172800, public, immutable, must-revalidate"
</If>
The last .ico part is not necessarily snac related obviously :3

2025-03-22 00:27:39

- reply

@nprofile…mzyr Teaching and mentoring younger people. A lot of old people I admire are kept sharp by teaching young people the things they know, but also the things that are "current". The questions hurled at you from promising young people, will force you to contend with what you know, and try to integrate it into modern times, thus, finding essential wisdom threads linking the old and the new. You will likely find yourself also learning the new things, and building new things, with the people, or for the people, you're mentoring or teaching.
You mentioned physical activity, but also, do not underestimate the effects of diet on the brain. My wife and I have done a lot of experiments over the past few years, and while I don't claim any truths found from the few anecdotes I have, I can only say for near certainty that the effects on the mind, over time, usually months, seem to be major. I recommend approaching the matter with a printf() debugging mindset, minimize inputs through fasting and diet limits, usually starting with some ketogenic (vegan or otherwise) setup, then add things one at a time, ideally unprocessed, wait a week, add another thing, see how things affect you.
Reading is good, but writing is gooder :) Writing about things, or writing what you think, or writing a journal, or writing letters. A plus if you write with pen in hand, helps practice your hands' dexterity. I recommend writing for other people, over writing just for yourself, as the former sharpens the mind more, but a balance of both is good.
Something to ground the mind. We can work the mind in the abstract realm as much as we want, but as we grow older, I feel like the mind becomes more and more "floaty", and I fear it slowly gets "untethered", and that doesn't feel right. Presence, grounding the mind in the moment, in this world, feels important to me. Practices that connect the mind consciously with the movement of the body seem really good for that, like walking meditation, or some yoga-like stuff, or some very light martial arts, or archery. Archery is a good one, but any other "hitting the target" exercise through coordination between the mind and physical movement would probably work.
Sitting with one's own mind, for an hour or so, every day or few days, just us, my mind and I, for an extended time, feels really good too. A garbage collection process of sorts. Calming the mind waves, whether with an intentional meditation practice for half an hour, or just with a cup of hot tea in the park, listening to nature.
And a lot of that can't happen without: Forgiveness. We don't realize how much burdens our minds carry until we let go of the weight, how subconsciously distracting our background processes are, until we put them to rest. It takes a lot of mental work, it can be a bit painful sometimes, but letting go, forgiving, is the true freedom of the mind. Our time is limited, all of us, we won't last; everything else becomes trivia when compared to that.
Oh, and speaking of limited time, don't forget to enjoy some of it with friends and family. It's been a long journey with them perhaps, but it never is long enough. This is definitely one of those "has massive effects on the mind" but "no one really knows what to do about it" things.
But anyway, I'm rambling now, what do I really know~
I wish you the best on your journey of the mind :)

2025-03-21 22:53:39

- reply

@nprofile…73zh The UI didn't have the profile pics of the other person, because I use uBlock, and don't allow 3rd party connections by default. So it is my fault. Although, I might modify snac to make it render something, like an empty square or such, if it couldn't load the profile pic.
https://fedi.dreamscape.link/rozenglass/s/4f830230b350df0a47de9f59c23be60e.png

2025-03-21 22:51:34

- reply

@nprofile…73zh Ah, sorry, I was sleepy, and got confused by the UI. Didn't notice it's something you boosted, but thought you posted it.
https://fedi.dreamscape.link/rozenglass/s/e276493d84fdaa745561a5bd93e52a0b.png

2025-03-21 10:51:08

@nprofile…73zh I hear you're taking a break from snac, to whom should I direct potential future patches?
I have a simpler one this time :)
Allow custom FROM address for email notifications
https://dreamscape.link/vault/public/debug/snac2/0003-Allow-custom-FROM-address-for-email-notifications.patch

2025-03-21 09:34:43

- reply

@nprofile…mtjv I enjoy snac, very light, and works well enough for me. There are a few inconveniences here and there, but I'm slowly patching them up on my instance, and upstreaming them when appropriate. Might want to throw https://git.freesoftwareextremist.com/bloat as a frontend to snac for a slightly more "modern" experience, but I usually just use snac plain. If you face any problems or have more questions, I may be able to help :)

2025-03-21 05:48:38

- reply

@nprofile…46gc @nprofile…azjd preferable to having a rake with DRM that can take me to prison if I lend it to my friends and family, can surprise me with unwanted change anytime it likes, can refuse to work for me, and can work on its own against me in hidden ways. I prefer my simple rake no matter how hard to use it is :3
Besides, I suspect that drawing plain circles is not really a frequently used feature. I never used it much when I used Photoshop, and I don't in GIMP. If I found myself using it a lot, needing to draw a lot of plain circles, I would just code it in myself. I see no one did over such a long span of time, so I assume no one cares enough to build it, nor to donate to someone to prioritize building it.

2025-03-21 05:22:58

- reply

@nprofile…azjd @nprofile…46gc I always drew circles by using the Ellipse Select Tool, select while pressing shift to keep it a circle, then go to menu: Select->Border... set thickness, you got a selection that's a circle, then grab a color and fill the circle. If you want more control, Select->To Path, and from there, you have your circle as a vector path (check the Paths tab) that you can stroke (right-click the path, then Stroke Path...). Once I got used to those, I don't find it a problem :p Although I still use ImageMagick for doing things to many pics at the same time.

2025-03-21 02:14:18

- reply

@nprofile…f6e6 @nprofile…zqaw Fun fact, in Arabic we say "sexual rape" a lot, to differentiate it from the other kind; we use the same word to mean land occupation. For example, you might say "The British occupied Egypt for 40 years", but in Arabic we say "The British raped Egypt for 40 years". And a land that's occupied is called a "Raped Land" (أرض مغتصبة).

2025-03-21 01:43:56

- reply

@nprofile…azjd Yup, my head cannon is that, most likely, that was Atlantis.

2025-03-21 01:41:58

- reply

@nprofile…9a3l My understanding is that we're talking 10,000 years ago, which is likely around the end of the ice age. When the earth starts warming, then naturally, areas closer to the equator warm up faster, and thus become hospitable to life first. But as the temperature rises globally, the perfectly hospitable areas starts going further north and south, away from the equator, and the equator becomes hotter and hotter until it becomes less likely to host forests and massive plant life. Humans can no way cause such massive effects with the level of tech they had at the time. This is my current mental model at least, I make no claims to its truth or accuracy.

2025-03-21 01:28:02

- reply

@nprofile…h6f2 @nprofile…azjd I'd say better gnusterminated.

2025-03-21 01:13:56

- reply

@nprofile…00jw Not exactly. It is the five aspects of the "Big Five" / OCEAN[1]: Openness to experience, Conscientiousness, Extraversion, Agreeableness, Neuroticism. Each plus indicates one standard deviation above average, and each minus is one standard deviation below average. So, C+ means I'm around 1 standard deviation above average in Conscientiousness (i.e. Industriousness / Orderliness), etc. And yes, O+++ means "extreme outlier" in Openness; my mind is so open my brains fell out :)
[1]: https://en.wikipedia.org/wiki/Big_Five_personality_traits

2025-03-20 23:05:24

- reply

@nprofile…fta6 @nprofile…m3qg I wonder sometimes, there's Zen in archery, Zen in blacksmithing, Zen in meditation, what Zen is there in programming? What would it look like for a Zen master to practice Zen programming? Offline hacking on C or Lisp might be a start.

2025-03-20 21:49:05

- reply

@nprofile…h6f2 as a FOSS micro-hyperzealot, I thank the ethereal spirits of the network, every day, that people are actually still /choosing/ to use corporate proprietary software. When people are /really/ at the point of being /forced/ to use proprietary software, I think we should be up in arms instead :p

2025-03-20 21:40:37

- reply

@nprofile…hmru give her some then! what are you waiting for!
(on a more empathetic note, I can imagine that feeling must be horrible)

2025-03-20 21:22:01

- reply

@nprofile…hmru @nprofile…csx6 @nprofile…h6f2 @nprofile…r3jc Intellectual property, just like normal property, does not "exist". It is a social construct; if we all agree that if you made a house you own it, and we will all collectively help you protect your rightful ownership of it, then our society has property rights.
You're writing a great book, and you're working with a publisher to edit it and print it, and five other publishers used their spies to leak the content of the book, then publish it faster through lower quality paper and cover and editing, cornered the whole market for your book, did not share any profits with you, and by the time you hit the market you barely make any sales. If we agree as a society that this is a dick move the needs to be prevented, then we have agreed to intellectual property.
You've been developing, for three years, a piece of software for Aramaic schools teaching English, some people hacked your computer, took your code, slapped a different logo and name on it, and sold your software project to all Aramaic English schools in the country, without your knowing, and before you even consider the software finished yourself. If this is a dick move that should be prevented by us collectively, then we have considered software to be intellectual property.
I think publisher companies leaking your book and printing it is /not the same/ as "I gave two copies of Photoshop to my brothers". To treat the latter as a crime equivalent to the first is preposterous.
I think intellectual property and copyright /laws/, as they exist now, and as they're forced on every country that wants to deal with the US of A through trade agreements, are evil. Kids downloading research to do more research for school, are not the same as companies taking your research and building a product from it and making billions while giving you nothing.
Copyright laws were a social contract to help reward authors, now they're a massive oppressive system to reward massive media semi-monopolies. Those are not the same, no matter how much they want us to be gaslit into believing so.
With the advent of computers with near zero-cost copying, the internet with zero-cost copying across distance, and now the collective infatuation with generative AI with automatic constrained remixing of numerous random sources, I think copyright in any shape is likely no longer sustainable. The only ways to sustain it seem to be extremely oppressive and evil. Thus, even if we collectively would love for the writer and the artist to make money from their works, that ship has already sailed, and it is almost impossible to do in current day and age. All artists I know that live from their art do not rely on the idea that copyright will prevent their fans from copying and sharing their artwork, but instead they build a social relationship with their fans, and gain their loyalty and support, and then make income from in-group signaling status "token" stuff, like buying merch, or subscriptions for early access, or paying to color your chatbox with another color, etc.
You can't just keep taking someone else's art and pretending its yours for too long, because the internet immune system, with its infinite eyes on you, at some point, is going to call you out on your lies, and your social capital with your fans will be destroyed. No need for copyright law or governments to be involved.
Copyright laws by now are making more harm than good, I think. We no longer as a society can justify people in the Middle East, India, and Africa dying from preventable diseases because some multi-trillion dollar US company needs to hold the IP for the medicine indefinitely until it can figure out a way to monetize low-income markets. It's evil. We cannot justify sharing research between researchers being a crime, replicating efficient systems being a crime, speeding up the improvement of life across the world being a crime. It's ridiculous.
Pieces of software are just lists of instructions for fast-running bureaucratic systems, and to artificially limit them from being replicated if they were useful to people is ridiculous. To think that software makers can, and should, make money by hiding the schematics of such systems, thus making them unchangeable, unreplicatable, and 100% extra oppressive, is ridiculous. Software can, and should, be left alone to be monetized through side channels just like art can.
And no, you will not be able to make trillions of dollars out of software if copyright law did not exist, but do we really, as a society, need software companies and people to be making trillions of dollars? To be the highest rewarded type of work in the entirety of society and human existence? Are we really going to, through artificial oppressive ridiculous laws, inflate "made an LLM" or "made an iPod", to be thousands to millions of times more rewarded than anything else that can ever be done by a human in service of human kind in all domains of work? What a stupid bubble we live in. I hope to see the day when we look back at the idea of multi-trillion dollar software companies and laugh our asses off about how silly that time of humanity was.
I used the word too much, but I guess I will say it again: ridiculous.
text wall dedicated
to the memory of
Aaron Swartz

2025-03-20 12:18:57

- reply

@nprofile…fctt @nprofile…ak3x Ah, thank ye~

2025-03-20 12:17:52

- reply

@nprofile…w8cy wireguard and termux, so i can start an sshd on my phone, from which I su to root, and start adb tcp debugging, and then use scrcpy to access my phone remotely. I keep my phone tethered to a desk on the opposite side of the living room, and never take it with me anywhere outside, or to other rooms in the house. This setup helps me when I need stuff from my bank app (in an isolated Android account), or to get SMS codes for my cargo, and other such modern mental illness, while also "not using a smart phone" by default. Back when I still used a phone, my favorite apps were probably unciv and endless skies :p

2025-03-20 09:52:31

- reply

@nprofile…mess @nprofile…6yr6 You don't have to, just toggle things and change things at runtime however you want (press 'h' for help, to find what's available), then press Shift+w, and top will write a config file for you in the correct path, thus saving your settings.

2025-03-20 06:10:46

@nprofile…ak3x Hey sathy, any news on princess gentoo? I see all of her sites that I had URLs for are down, and have been down for days. I hope she's okay...

2025-03-19 10:21:58

- reply

@nprofile…s2gy normalize spending time with one's own mind.

2025-03-15 00:08:21

- reply

@nprofile…hmru @nprofile…h6f2 Had they been more cooperative and social, they would probably be on the main social networks, because their desire for cohesion would override their privacy / security / freedom / seclusion natures; getting along above all else.

2025-03-14 22:32:59

Disco Elysium is just fedi simulator.

2025-03-14 20:10:44

- reply

@nprofile…a0fc hehe, no worries ^-^

2025-03-14 20:04:28

- reply

@nprofile…a0fc The routing table tells the OS where to send a packet depending on the address it's meant to. It does something similar to zip codes in postage. For example, if you have a packet meant for Google's IP, the routing table may tell your OS to send the packet not to Google's IP, but actually to your WiFi router's IP. Thus, the routing table basically helps you delegate the actual delivery of a TCP/IP packet to some other network actor that actually knows what to do next.
I may want for example all my packets meant for the internet to go instead to my VPN server, because my VPN server will handle further delivery. So, I may set my VPN server's private IP as the default gateway in my routing table. But maybe I want to use some local network stuff, and my VPN server wouldn't know how to handle my local server, so my routing table would set my local router to be the destination if the packets are meant for local network IPs, I may have some work network that is handled by a different VPN, so work IPs get routed through that other VPN server, etc.
For your packet to reach its final destination, it is likely being routed by a dozen routers between you and your target server. Each of those intermediate servers is consulting its routing table to decide where to send the packet next, until, finally, the packet reaches a server that has Google's actual IP address in its routing table (think final postman when sending cargo). You can see this happening in real-time using software tools like "traceroute".

2025-03-14 19:42:32

@nprofile…73zh I noticed some posts missing their images. This happens because some attachment images are skipped if the body of the message contained a link to the picture. See this post for example:
https://dreamscape.link/vault/public/debug/snac2/4beb5a0d464e4ad952b192f0403635ea.json
I would rather see duplicate attachments, than miss some attachments completely. This patch removes the duplication check until a smarter detection solution is made (I'm not working on one for now):
Remove faulty attachments deduplication mechanism
Attachments are skipped if they are already included
in the post content, but this sometimes mistakenly
skips images if the post content includes an anchor
that links to the image too. I'd rather see some
duplicate attachments than miss them.
https://dreamscape.link/vault/public/debug/snac2/0002-Remove-faulty-attachments-deduplication-mechanism.patch
In an unrelated matter, I prefer the mentions to look like they do on X, Pleroma, and Mastodon, i.e. at the start of the post, without CC:, and including the replied to user. I wrote the following patch to make it that way. This solves my issue. If you're interested, I can perhaps make it a user option, to allow users to choose their mention style.
Switch to mainstream micro-blogging mentions style
* Mention the replied-to message owner when replying to them.
* Remove the "CC:" and put the mentions at the beginning.
https://dreamscape.link/vault/public/debug/snac2/0001-Switch-to-mainstream-micro-blogging-mentions-style.patch

2025-03-14 18:28:18

- reply

@nprofile…4duy Stardew Valley is best game, end game, and the last word in gaming, no need for any other.

2025-03-14 18:25:28

- reply

@nprofile…agan @nprofile…59py Excuse her, she's reading The Art of Seduction by Robert Green lately XD

2025-03-13 15:41:59

- reply

@nprofile…73zh
Hi ^-^
can't imagine what is causing the spurious unfollow
I have a few more notification objects of such cases, I can give you access to them too, if you wish. They differ in some ways, including the 'to' field, but all seem to share not having an embedded msg.object (but have a link there instead), and all don't include my user in the 'to' field (but normal good unfollows do).
your patch improves the process logic, so I'll merge it.
Go ahead, and thank you for your work :)
The patch seems to have saved me from one unfollow so far, I think, where utype and id both were null, at which case the previous code would have tried to delete the follower, but the new code stops instead. I still need to figure out how to set my followers as followers again on my side at some point, or be a nuisance and ask them to unfollow then follow me one by one :p

2025-03-13 01:34:27

@nprofile…73zh About your comment on small index files[1], taking 4KiB of storage each. Any method relying on file names for the storage itself (maybe as base64url) will now require scanning the names, any method relying on nested directories or file content (current model) requires lots of little 4KiB spaces plus constant opening and closing. An idea came to mind, of using xattrs to store the content, which does not use 4KiB per-file if the actual file content size is 0, but xattrs are likely not very portable and may have file-system specific quirks and limits.
Fundamentally, we need some kind of a tree structure, stored in a file, likely memory mapped. But writing to the file while maintaining the tree will have massive write amplification, unless we manage the storage as chunks, some for indexing, and some for content storage, and keep track of alive and dead ones, etc.
At that point, this sounds literally like LMDB. LMDB, is basically a "storage engine" (used by OpenLDAP) that does little more than that; Copy-on-write B+ Trees in a memory mapped file, in around 4K LoC of C; a much smaller dependency than cURL and OpenSSL, and can even perhaps be vendored in if needed. Its interface is a simple key-value store, with a few little extras, like providing efficient scanning of a subset of keys based on a prefix of the keys, and the ability to store multiple values in the same key and iterate on them, etc.
Do you see LMDB as a potential path forward for snac2 perhaps?
[1]: https://codeberg.org/grunfink/snac2/issues/43#issuecomment-961951

2025-03-12 22:23:01

- reply

@nprofile…q2ey Sounds roughly similar to what the SingleFile[1] browser extension does, which saves the current state of the page, and all of its dependencies, including images, to one "HTML" file (the data is not really directly in the HTML afaik, but a self-extracting ZIP file and its extractor, are embedded in HTML). The extension removes script tags by default when saving a file too, but that's optional. There's a command line version of the tool too.
[1]: https://github.com/gildas-lormeau/SingleFile

2025-03-12 21:04:29

- reply

@nprofile…qr5r What happened? I hope you're okay.

2025-03-12 21:02:02

- reply

@nprofile…4duy You have my sympathies; I see the women around me, and it indeed seems to be a tough life. Hang in there friend ^-^

2025-03-12 20:58:15

- reply

@nprofile…73zh I created this patch[1], which logs some debug info when the utype override happens, and changes the conditional logic to require "msg.object.object" to be equal to snac->actor before deleting an unfollower. I deployed this change on my instance, and tested it with real unfollows from Pleroma, Mastodon, and snac2, it still works okay. I have no way to test it with the "false" unfollows because I don't know what exactly triggers them.
[1]: https://dreamscape.link/vault/public/debug/snac2/debug-false-unfollows.patch

2025-03-12 19:22:29

@nprofile…73zh Hello, Grünfink, I've been facing an issue where friends from different instances (Pleroma mostly?) appear to unfollow me, but they confirm that they really didn't. I suspect commit e9f391e "Added a special check for Undo + Follow.", in which activitypub.c is modified as follows:
if (strcmp(type, "Undo") == 0) { /** **/
+ if (xs_type(object) != XSTYPE_DICT)
+ utype = "Follow";
+
It seems that it treats all messages of type 'Undo', where 'object' is not a dict, to be a utype 'Follow' by default. But that does not seem to be necessarily correct; there are messages that are not like that. This notification JSON for example shows a message that was likely overridden to be utype 'Follow', but originally wasn't. I confirmed with friendo that they didn't unfollow me. My account is not even in the 'to' field, which is a thing I noticed with all such mistaken unfollows (I have some 5 or 6 so far, in a few days):
https://dreamscape.link/vault/public/debug/snac2/1741791872.179863.json
Compare with this object, which is a real Pleroma unfollow I intentionally made:
https://dreamscape.link/vault/public/debug/snac2/1741805474.188476.json
Any thoughts or ideas? I'm familiar with C, but not enough with ActivityPub to know what is happening in this case. A quick solution that comes to mind, based on the handful of objects I've seen so far, is to check whether we're in the 'to' field before overriding the utype, but I have no idea why this whole utype override exists in the first place.

2025-03-12 05:12:47

- reply

@nprofile…j9ru Would you like some sugar, covered in sugar (plus colorants and artificial flavoring) ,with blobs of sugar on top (plus colorants and artificial flavoring), sprinkled with little sugar pieces (plus colorants and artificial flavoring)? We call it cHrIsTmAs cRaCk. Get it? Ha! Crack! Because it's just as good for your health, and just as addictive! /s

2025-03-11 02:28:17

- reply

@nprofile…z4ac They don't need xda-developers, they just made the project so massive and unwieldy to build that almost no one can even begin to hack on it. That, and all the OEM "innovations" in moving critical pieces of the system to non-free closed-source kernel modules. Otherwise, xda-developers is just a place where kids from India get their start with hacking by ricing and modding their devices.

2025-03-10 22:59:18

- reply

@nprofile…ds47 I hear Apple already switched their emoji to the new Syrian flag around three months after the government change. I haven't confirmed that though.

2025-03-10 22:32:00

- reply

@nprofile…ds47 that's the old flag now :p

2025-03-10 20:52:57

- reply

@nprofile…75xr @nprofile…h6f2 I think Facebook is too complicated. LinkedIn made my head spin when I tried it. I think self-hosting my fedi instance wasn't much harder than "learning Facebook", or the half-a-dozen local product delivery service in my country. The bigger barrier is just the network effect, I think. There are a few dozen peers around my mom to teach her Facebook, there are near-zero to teach her about the Fedi.
But perhaps it is better that way. When are we going to learn our lesson, and stop repeating the massacre of the Eternal September?

2025-03-10 20:24:51

- reply

@nprofile…23qj A funny coincidence, I was literally just watching some vids by https://www.youtube.com/@Asianometry about the quantum mechanics of microelectronics. It is not my area of specialty, but I also think it's not outrageous for a software developer to know the general ideas of their dependencies down to the base level of quantum physics.
Every engineering discipline does that, they go down to the bottom of physics when creating things, why should software engineers be any different? It is impossible if you're working in a disgusting swamp of over 10,000 dependency packages maintained by a million different people isolated from each other, so just don't do that. An engineer building a bridge doesn't introduce 10,000 materials that they need to factor into their calculations, instead, they cut out everything but the essentials. If you're working in an intentional, minimal, methodically engineered code-base, those things are not crazy.
I was contracted to debug a software issue once, an issue that perplexed two software developers before me. After days of digging and experiments, turned out to be a hardware-level electromagnetic interference issue corrupting data in-transfer. It's not rocket surgery. Physics exist. You can't just ignore it. It is completely feasible to at least read the general outline of the code of every dependency you use in a project, especially production ones, especially if you're selling to hospitals and schools and public infrastructure. All this cargo-culting, all the illusions people are telling themselves, are because they're lazy, privileged (complaining about the top 1% not knowing that they're the 1% with their $200K+ salaries), don't want to read, write, or think, and most importantly, they block their empathy behind those tech abstractions, to protect themselves from realizing that the slowness of their web app costs 10 million people 10 extra seconds each, per-day, and that amounts to 115 years of human life, per-year. Their stupid app is doing the equivalent of unnecessarily murdering 1.4 innocent people per year, and they feel nothing about it, sipping on their Starbucks, they say smugly: "Developer time is more expensive than machine time", as if to magically absolve themselves from their responsibility, or revel in their complacency in eroding the human soul, because for some reason, developers' souls are more expensive or something.
Aaaanyway, I'm just ranting by now. A decade of work in the industry got me here. But I'm still hopeful, otherwise I would have left a long time ago. I'm hopeful that people would take responsibility, that they would finally honor their craft, and know that behind every system they make are real human souls being shaped and moved. Or put more practically, that the people who see the errors of our ways, would out compete the sloppy masses in the market of ideas and production.
Or maybe, a man can dream ^-^

2025-03-10 18:40:09

- reply

@nprofile…agan That's good! :3
Sometimes I wonder if I'm a bit like you in that regard.
I'm definitely not 100% "typical" at least.

2025-03-10 18:37:33

- reply

@nprofile…23qj Modern devs don't take ownership of code they write, even before LLMs appeared on the scene. See the massive towers of NPM dependencies, all the 100-layer-call-stack-depth frameworks used everywhere, and the infinite piles of copy-paste-with-slight-makeup code from StackOverflow. Modern software development is a disgrace to the human intellectual discipline of engineering.

2025-03-10 18:32:09

- reply

You're feeling better about your body image now? ^-^

2025-03-10 16:38:46

- reply

@nprofile…agan I find it really hard to show any parts of my body beyond head, hands, and forearms, to anyone who's not close and trusted. I wore full-clothing whenever I went swimming for example, I like wearing collared shirts and closing top button to prevent my neck and upper chest from showing. I don't know why that is.
I was going to blame an extreme religious upbringing, but not all family members around me seem to suffer from this as much, and 10 years after I left my religion, I still don't feel okay with exposing myself. It is not a confidence issue either, I think, otherwise I would have cared more about showing to people closer to me.
My suspicion is that it's something about putting up a mask of perfect discipline to strangers, to make it less likely that they'd "fuck with me". Potentially related to growing up in an oppressive socialist government where people wore military uniforms to high-school; orderly clothing and uniforms portray this discipline, thus inspires some tinge of fear, while going around near-naked signals the opposite to most people; vulnerability.
It's confusing to me, and I don't know if I agree with my own opinions about this. But the repulsion and anxiety of exposure is a low-level physiological reaction in my body, not something on the conscious thought-level, so hard to change.

2025-03-09 18:53:42

- reply

@nprofile…3wfk @nprofile…nc0q @nprofile…6yr6 I hear the idea of bottlenecks discussed in regards to typing, but it doesn't make sense to me. It doesn't apply when things are of different nature; i.e. thinking and typing. You think about bottlenecks in something uniform, like code, to know where to focus your efforts for performance improvement. But when discussing typing and thinking, you don't really reallocate your focus from improved typing speed to improved thinking speed instead.
For example: if we estimate, in an average code base you work on, you type for 25% of time, and think for 75% of time, and you worked on the project for 4 hours per-work-day, for a month, that's 22 days * 4 hours = 88 work hours, of which 22 hours are typing. If you double your typing speed, you would save 11 hours per-month, which is 132 hours per-year.
So, even if you invest 100 hours to improve your typing speed, you would break even in less than a year, and then, you'd have 132 extra hours of free time every year from then onward. Plus, the street cred among peers, and you have less chance of typing taking a bit too long and breaking your thought chain, and good typing posture reduces risk of hand injury, etc. etc.
So, I'd say, it does matter. If you can, by allocating a similar amount of time, improve your thinking speed by around 16%, then yup, that would be a better allocation. But there are limits to that, and once you reach the limits of speed improvement for your thinking, your next best bet is improve typing speed.
But who am I kidding, for most working dev people, the thing that would provide most productivity improvement is to get rid of most meetings, and most incompetent managers, thought and typing speeds pale in comparison to those time sinks.
Type on friendos :3