more uf3 fonts, this time trying to rasterize a vector font. This is my rendition of Instrument serif
Once, someone asked what was the weirdest platform you run GtS on. Well...
This message was sent from
#GoToSocial running on an
#OpenBSD #arm64 board (#ODROID HC4).
@joel how the experience?
@radhitya pretty good so far. Not seen a difference yet from previous amd64 VPS installation. But that’s only 1 day history :)
I think this #GoToSocial instance ran on #OpenBSD #arm64 using #ODROID HC4 is doing pretty well while eating (only) 5W.
@joel Do you run of SD-card or something more reliable ?
@santi it initializes the boot (using u-boot) from the SDcard but the whole system lives on a SATA drive.
https://www.tumfatig.net/2026/running-openbsd-7.8-on-odroid-hc4/
EuroBSDcon 2026's call for papers is open, see https://2026.eurobsdcon.org/cfp/
Here is a direct link to the submissions form https://events.eurobsdcon.org/2026/submit/ew426G/info/
@eurobsdcon #eurobsdcon #openbsd #netbsd #freebsd #conference #development #devops #sysadmin #freesoftware #libresoftware
I don't like liquid glass, but it is not that bad after all...
is it just me? are there any operating systems for google pixel with polished/minimalist ui?
@hi no, that’s not just you. I have a GrapheneOS thing for work, and… let’s just say it doesn’t diminish the rate at which my hate for everything Android other then its system shell (*g*) grows.
Just booted my SE (2016) again to also update it and charge it up again, and, even the SE (2022) is such a downgrade, but still pleasurable compared to the Pixel 9a, both hardware‑ and software-wise. (With the Pixel, I cannot tell which way is up. At all. All the time.) It helps they’re still on iOS 18.
I just don't want to touch android phone, so bad it looks and feels to me...
@hi @mirabilos this sounds like a mobile phone withdrawal. I envy you. You'll get over it, hopefully, and start to see the world again ;-)
@hi not that I don't see the same stuff on the stock Google's Android.
The most useful app on my phone is Termux - a terminal emulator.
I never owned or used an iPhone. I probably should stay away from it, given how addictive these ugly inferior Android phones still are, it'd probably qualify as a hard drug ;-)
@hi having gone back and forth a few times from stock android > iphone > calyxos/lineage > iphone again...
no it's not you, unfortunately. i wish the open source phone was calmer and more pleasant to use but it's just not.
phone bad!
On a networking event they prompted us to display our LinkedIn QR codes. For easy connecting.
I refuse to have LinkedIn, so I made an alternative. What do you think? Any suggestions? Clever ideas? https://geff.re/cv/
@geffrey wow!!!
Open source?
@theodorus_75 It’s all there! No backend required. Go ahead and copy to your heart’s content. 😌
The contact page itself has some obfuscation going during the Eleventy build. If you like that too, I can share that for a while.
In #openbsd what is called #vi is actually #nvi, and when you install #nvi from ports, it is #nvi2.
To danes with our national characters æøå we need the latter because #vi prints two byte hex sequences for æ, ø, or å. #vi handles them correctly except for the printing of them.
#vi and #nvi use the same #man page.
Update:
On #netbsd #vi is also #nvi. Here you must install #nvi2 to get #nvi if you need æ, ø, and å support.
Best wishes from T. R. Dane (The Real Dane ;)
TIL about SQLite .www mode:
$ sqlite3 my.db
sqlite> .www
sqlite> SELECT * from mytable;
Opens the result of select in a browser and shows the table!
now we can confirms all 8,000 bunnies can fit inside a single #uxn vm.
p.s. in the background, ps -o %cpu,rss runs in a loop
Graphically visualizing the erasure of a 16kbit #EPROM with a UV source in 16 seconds!
This time lapse shows all of the bits being flipped from 0 to 1 over about a 24 minute period. The first bit flips at 8 minutes.
What's fascinating to me is the bits "twinkle" as they flip back and forth before settling.
@TangentDelta are you manually doing ^C to exit?
@snufkin_vc Yeah... That's on me though, as I had some of the logic related to exiting an Uxn instance reversed and was having to manually ctrl-c out. It's fixed now.
So Anthropic employees are using Claude Code to contribute AI-generated code to open source repositories and hiding the fact using their own internal “undercover mode”.
Totally trustworthy people.
(Any open source project that at the very least requires disclosure of AI-authored contributions should immediately ban Anthropic employees on principle.)
@aral Honestly I don't actually hate this.
It's a tool. The _user_ is responsible for what they're submitting. It's putting code generated by them in their name. I think this is actually good.
@aredridel @aral I really can’t agree with this, because it’s a question of accurate labeling not of “responsibility” or “authorship”. co-authored-by is perhaps the wrong method for labeling such things, but consider raw milk. ultimately, it is indeed the producer’s responsibility to ensure their product is free of contamination. but disclosure of its method of production is explicitly the kind of requirement that allows consumers of said product to make safe choices
@glyph Yeah, I disagree. Code isn't ingredients and it's not “contamination" any more than you should label “I used search and replace on this”
What you want to know is whether it was well engineered or not.
And in fact, this is almost entirely orthogonal to "safety”. This is an engineering product. The safety comes from processes and whether or not _anyone checked the work done was right_, not the inputs.
@aredridel @glyph It is ingredients. It's not search-and-replace. It's literally incorporating parts of an unknown set of almost-surely-copyrighted works, without license or attribution, into the submission the person is misrepresenting as their own.
@aredridel @glyph What "AI coding tools" *should* be putting in commit messages is:
Co-Authored-By: An unknown and unknowable set of people who did not consent to their work being used this way and to which there is no license for inclusion.
@dalias Morally arguable but not actually true under the copyright regime that exists.
At what point does learning from others constitute their authorship?
@aredridel @dalias it is true.
And LLMs cannot learn. They are merely a lossy compression/decompression thing. They regurgitate a somewhat averaged completion of the prompt from the other works they ingested.
@mirabilos
Not, strictly, true, though I get what you're going at.
There's a few phenomena going on that shape these tools beyond that.
- Emergent complexity
- "Memory" records
- Embedded context
- Incorporating social inputs
I still think they are strictly tools, but ones that can self-adapt with alarming power if you configure them right.
They have a medianizing effect on a lot of their output (that's actually one of the reasons they're good at code. We generally want code to be "normal". It's one of the many reasons it's pretty bad for more artistic creative work, morally and technically.)
But that's not the same as only repeating the median. The temperature, the randomness injected in makes them actually jump to stuff that is at times nonsensical but also at times clever. It's just randomness, but then with a heap of context and congruence applied that is rather interesting.
@aredridel LLM slop is nothing like "learning from others".
But if you recall, we even took precautions against that. FOSS projects reimplementing proprietary things were careful to exclude anyone who might had read the proprietary source, disassembled proprietary code, worked at the companies who wrote or had access to that code, etc.
@dalias Yes. Do you know why?
@aredridel So that it would be abundantly clear, in any plausibly relevant jurisdiction, that the work was not derivative and not infringing.
@dalias @aredridel A test which LLMs fail by the very virtue of their functioning mechanisms.
It's all fundamentally derivative of the training dataset and it has been exposed both to AGPL and to proprietary datasets.
@lispi314 Has any legal authority weighed in on that claim yet?
@aredridel @lispi314 The facts of the matter are completely and utterly obvious.
Now, we live in a world where legal authorities are under complete capture by billionaires pushing this drug, so I am not going to make any predictions about how courts will rule. Even if they do rule in favor of these companies, those rulings will not be treated as precedents that benefit us.
And they will not be accepted by our communities.
What defines FOSS is not whether a court says it's non-infringing, but whether our communities agree that it was made respecting the intent and consent of the authors who licensed it.
@dalias Have you checked with the Free Software Foundation about that?
(Seriously, if it's a moral argument you're making, it's way stronger if you actually make it!)
Now "respect the intent of the author" is a fascinating concept and one worth examining!
@aredridel The FSF is a fan club for a sex pest, so no, I have not checked with them. I am speaking for the communities I would want to be a part of.
@dalias Right. You're appealing to a definition of "FOSS" that isn't entirely clear what it is. And the people who do usually have (some) claim to that authority, the common uses of it, are not the ones you're using.
I'm sympathetic to that but I can't tell what it is in an appeal to an unstated norm for a community that I can't quite identify.
@aredridel @dalias that’s just deflecting, asking dalias to define something that is not even important to the point he’s making
@aredridel @dalias
> but not actually true under the copyright regime that exists
Under the copyright regime that exists in the US specifically, the generated code is at best not copyrightable at all (and therefore cannot be included into any projects with licenses relying on copyright).
Of course maintainers of said projects might decide to yolo it, but also they might decide to not; and in this case, the intentional deception by antropic becomes even more significant fraud.
@IngaLovinde @aredridel The ruling you're talking about was a case about actually *generated* code, before "gen AI" was a real thing, not obviously-derivative transformations of a corpus.
@dalias @IngaLovinde @aredridel AFAICT it merely confirms that AI output cannot be copyrighted as new work of its own (naturally, as the human creativity aspect is missing and it’s merely an algorithmic transformation on a deterministic machine, PRNG inputs notwithstanding).
It does not reduce the claims of the authors of the works it ingested to regurgitate the output.
@mirabilos @IngaLovinde @aredridel Exactly.
@dalias @mirabilos @aredridel which still means that at best the generated code cannot be copyrighted, and at worst it violated copyright and license terms of the original authors (whose works were ingested to train the model). In both these cases, the resulting code cannot be incorporated into any FOSS project with any license.
Typically when people submit code to FOSS projects, they also (implicitly or explicitly) claim that they hold the copyright on the submitted code, and agree that this code will be licensed under the license the project uses (which they only have power to do if the first claim is actually true).
When LLMs are used to generate code, the first claim is false, and it _is_ a contamination.
@dalias @IngaLovinde @aredridel if something cannot be copyrighted and no others’ rights apply, then it is in the public domain. For LLM output, which has been proven to vastly resemble existing code under copyright, that’s not the case.
@mirabilos @IngaLovinde @aredridel Indeed it's been demonstrated that you can "coax" LLM chatbots into emitting large parts of their training corpus nearly verbatim, so it's clear that the works from the corpus, with minor degrees of lossiness, are contained within the models. And when they output something very similar, it's ridiculous trying to argue that the output isn't derivative too.
@dalias Have you seen how people perform on similar tests?
@aredridel @dalias people are still humans, not machines.
Are you a TESCREAList?
@mirabilos Not even remotely TESCREAList. However, I think it's a fair question to ask: why are we drawing lines how we do? Especially when comparing work product.
@aredridel on the most basic level because copyright mandates human creativity, the expression of human personality
@mirabilos The thing is that actual use of these systems tends to involve LOTS of human creativity and attention. Lots of video and bits get spilled on hierarchical autonomous agent models and the hype, but real use? Much more hands on. The "I don't write code by hand anymore" people aren't just a minority but an extreme minority.
@aredridel the prompt is but one of the many inputs that go into the regurgigated thing, but a minority compared to the "training data" *shrug*
@dalias Got examples to show to support that position?
Remember copyright is a _legal_ regime and the legal regime seems quite oriented toward that NOT being the case.
@aredridel @dalias the Text and Data Mining exception to copyright law only applies ⓐ to models for analytics (discovery of patterns, trends and correlations; §44b UrhG), and ⓑ to works whose right holders didn’t opt out (ibid. p.3); there’s absolutely no basis on which “genAI” even could be considered permissible, as both the reproduction (§16) and the right to make changes, editions and other derivatives (§23) are protected by law, by default, and always require a licence.
So, thrice denied, by existing law.
@glyph @aredridel @aral This. It's critical to understanding the risk presented by a PR to know that it's not authored by a human, but extruded and then at most reviewed by a human.
From a quality perspective, and to your point earlier, code review is insufficient to find flaws in extruded code. While the legal risks are likely small, as per previous discussions, they're not strictly zero, nor are they well-tested in courts.
@xgranade @glyph @aredridel @aral I’d even go so far and say that you cannot rightly publish the result of slop under a FOSS licence because you lack what in the Linux community is known as the DCO.
@glyph @aredridel @aral I approach the above from what I consider a deeply unethical viewpoint, namely one of asking about the practicalities of slop PRs instead of the ethicality, in part because that's the only axis along which there's any nontrivial question whatsoever.
Along an ethical axis, the onus is very strongly on OSS projects to reject slop PRs, at which point the Claude Code prompt is even more odious — it demonstrates an intent to lie in order to bypass codes of ethics.
@aredridel @aral it’s not a tool.
@mirabilos @aral Wow I straight up think this concept of 'tool' is not good. The stick, the bit of whatever to open a bottle? That absolutely is a tool. To have it have a purpose in advance is such a restriction on the idea of a tool.
An enabling technology. A way to extend ourselves. That's a tool. Whether bag or bicycle.
@aredridel @aral no, it’s not a tool. It’s a… I’m not English, so I don’t have all the words, best I can say a quick hack barely good enough for the use, but not designed and engineered to help the use. (I don’t want to use the word “hack” here because it’s worse than that.)
@mirabilos Quick hacks are tools! That's the beauty of being human. We turn all kinds of stuff into tools.
Whoever did the ◀ ◀ REW graphic for Apple's homepage today, has never used a one-bit display with Chicago.
Besides the arrows being impossible to render, the glyphs in the text didn't have the standard 6:9 ratio. Also lots of anti-aliasing which is kind of hard to do when your only choices are black or white 😉
I pretended to be Susan Kare and fixed it, bottom is the original, top is my interpretation.
Just yesterday, I was working on the Easter Egg in Notchmeister.
In a one-bit Chicago kind of mood, I guess. 😉
More about Notchmeister in the App Store: https://apps.apple.com/us/app/notchmeister/id1599169747?mt=12
@chockenberry Kids these days. Sheesh.
3 things i have learnt about nvi today:
a) There are _number_ buffers that hold the nth last yanked/deleted text (that included newlines): "3p pastes the 3rd most recent thing you deleted (this one applies to vi and is thanks to Learning the Vi Editor by Linda Lamb);
b) nvi faithfully implements a second u undoing the first u, therefore making a 2-cycle undo/redo loop; but, a . after a u undoes older and older edits! (thanks to a random blog post);
c) the :script sh ex command is like emacs subshell-buffers, except without the convenience of editing with emacs bindings. Surreal. (thanks to unearthing ancient pty calls in the source code)
@drj I've known/used the first two for a long while, but the 3rd one is not only new to me.
The man-page mentions nothing about :script, and the source-code are pretty lacking/opaque in what it does. I can't seem to get it to do anything useful (it seems to take a filename and dump it, then put your shell-prompt in the text?)
What does your book suggest that it's supposed to do?
@justine I prefer nvi2 on FreeBSD to vim, it's simpler and more responsive I think. It gets some getting used to though, things like undo are u followed by . to undo more, then you can press u again followed by . to redo a couple times.
That's one reason why the often quoted reply https://stackoverflow.com/a/1220118 is so good: It makes it clear that many "features" or "improvements" of Vim are actually not improvements at all, but rather workarounds if one does not properly understand how to do things using what is already given.
One of my favorite examples is visual mode, which I have used for years before I noticed I don't need it at all; I just failed to properly learn the respective ccommands for moving and marking.
Of course, the next logical step after moving from Vim to vi would be to move on to ed 
@thorstenzoeller @RussSharek @justine @clf
/me doffs hat with a warm smile and welcoming gesture
One of the cool things is even if you go back
I bet you will have a lot more of the vi
primitives in your fingers.
Like you can learn ALL of vi.
You can't say that about emacs or vim.
@mrmasterkeyboard There's a discussion on misc@ about this type of commits https://marc.info/?l=openbsd-misc&m=177507171827044&w=2
@horia I don't care for their responses. They still use genAI. genAI is not an auto-complete like they seem to say from my quick read, far from it. that's like calling a book a deadly weapon because it can hurt people by throwing it at them. yes, genAI can steal/generate code but it cannot do it in the same way an autocomplete can.
Folks trying to de-slopify your tools. What are you doing when your favored programming language/compiler starts accepting llm-generated code contributions? #NoAI
Switching compilers? Switching languages? Sighing in despair but continuing to use it?
@trashheap currently option 3 as no one is working on option 4 (stay on the last slop free version of software XY)
After a long and annoying journey, my new music player is a used 6th gen iPod classic with a new 128GB SSD harddrive. I installed Rockbox, a replacement firmware that bypasses the need for iTunes and allows custom skins for the iPod interface.
#ocular #gear
https://ocular.nchrs.xyz/feed.html#ipod.jpg
If you need a laugh, you can read up on my escapades of trying to get this to work on my Linux computer:
@rostiger hm, good idea. i’m still using my ipod nano, but dread updating the library… itunes
@tovabele If it's a 1g or a 2g (I don't know how many g's there were), because Rockbox supports those. With a windows machine, the process is fairly simple, too. And then you can kiss iTunes goodbye foreva (or rather boot it out the door).
Late at night
@hi you don't. I kinda managed to distract the kiddo with an old EEE PC, but that didn't work for long.
@hi at that time, I had to split days. Computer was mostly done when elses were asleep or watching TV :)
It's been a while, but I tried to knock out housework with them around. That helped them see that the house didn't magically clean itself nor did meals magically appear while they napped. That left their nap time¹ for me to do work-/hobby-related things.
Additionally, we don't watch much TV (no streaming services, just the occasional DVD from the library), so that frees up a fair bit of time after they go to bed.
Finally, we also later enrolled them in preschool (we opted for 2days/wk when they were 2, 3days/wk when they were 3, etc) which also availed a few hours for work/hobbies.
⸻
¹ this later morphed into "Quiet Rest Time" where they had to spend 60–90min in their room doing something quiet—read, nap, play with toys, whatever as long as it was quiet…at 11 and 16, they still do this on weekends, and it helps with *everybody's* sanity 😆
no screens, no nap, so it has been a hundred hour work week with no vacations for two parents :)
amazing time, but not so much time for hobbies.
"quiet rest time" is a good idea! thank you, Tim
@hi around 5 AM on saturday and sunday morning before they wake up. Im a morning person so it works out and then maybe 1 day a week after bedtime. But that doesnt happen that often because Im tired.
Added a secret journal to my site.
It's password-protected in a fun way — decrypted as you type. Get the password right and *poof* the words pop out.
I'm going to post little updates once in a while for the humans in my life, away from the prying eyes of LLMs.
Would you like the password? Have we ever DM'd or talked? Then probably! DM me :)
And — 1000 internet points if you can crack the encryption by any means. I think it's pretty tight, but I'm still a baby.
Boosted cos this looks like fun.
And it's my brother's friend's birthday.
Unfortunate birthday really but he didn't get to choose.
@spiralganglion we should pool our resources to solve this. My contribution is: the password is not "gruntfuttock".
@spiralganglion so cool! nice that you encrypted it paragraph by parapagh.
we kinda can see the shape of text instead of just pure chaos
@pdro one fun little detail — i inject pseudorandom spaces into the encrypted text, so the lines wrap like words do.
@spiralganglion I would *love* the code you wrote for this ... I was thinking of doing something similar !! but not sure how to do it
@alifeee sure! this first bit is in my website build script. it reads markdown files from a `journal` folder and encrypts their text (but not formatting) with a password from a `.secret` file, both excluded from git. the result is saved to my `source` folder and then compiled as if it's a normal page on my site.
https://github.com/ivanreese/ivanish/blob/274b38155a9eb6ed42b610121fd4324e0783556d/Cakefile#L436
and then here's the client, which does the decryption
https://github.com/ivanreese/ivanish/blob/main/source/script/journal.coffee
i tried to add lots of comments for future me, but if anything is confusing let me know!
Theo de Raadt has introduced a new hw.blockcpu sysctl to #OpenBSD -current to offer more control over which CPU core types (Performance, Efficiency, and SMT) are available to schedule processes on. Modern Intel (and ARM) CPUs additionally have slower LP-E (low-power) core that severely hinder system performance.
deraadt@ modified src/sys/*: Some new intel machines have a new 3rd tier of cpus called LP-E which are E-core (Atom) without L3 cache. These cpus are Lethargic, and it sucks when processes migrate to them.
This introduces sysctl hw.blockcpu= which takes a sequence of 4 letters.
S (for SMT), P (regular performance cpu), E (efficient cpu) generally 80% to 50% as fast), and L (lethargic cpu) which are even slower.
By setting this, you can select cpus to kick out of the scheduler. The default is SL.
The hw.smt sysctl remains for now but we will eventually delete it.
hw.smt changes and follows hw.blockcpu=S.
ok kettenis mlarkin
Would I then hw=blockcpu=L
To enable smt ??
Thanks in advance and apologies if I'm misunderstanding this.
@justine Yes, as I understand it. That would block only the LP-E cores.
hw.smt is still supported in 7.9-beta but will now print a deprecation warning.
hw.machine=amd64
hw.model=Intel(R) Core(TM) i5-10505 CPU @ 3.20GHz
hw.ncpu=12
hw.byteorder=1234
hw.pagesize=4096
hw.disknames=cd0:,sd0:5b94487c8aa38d41,sd1:d37010def7c156f7
hw.diskcount=3
hw.sensors.cpu0.temp0=30.00 degC
hw.sensors.cpu0.frequency0=4400000000.00 Hz
hw.sensors.cpu1.frequency0=4400000000.00 Hz
hw.sensors.cpu2.frequency0=4400000000.00 Hz
hw.sensors.cpu3.frequency0=4350000000.00 Hz
hw.sensors.cpu4.frequency0=4300000000.00 Hz
hw.sensors.cpu5.frequency0=4350000000.00 Hz
hw.sensors.acpitz0.temp0=27.80 degC (zone temperature)
hw.sensors.nvme0.temp0=28.00 degC, OK
hw.sensors.nvme0.percent0=0.00% (endurance used), OK
hw.sensors.nvme0.percent1=100.00% (available spare), OK
hw.sensors.softraid0.drive0=online (sd1), OK
hw.cpuspeed=3201
hw.setperf=100
hw.vendor=Dell Inc.
hw.product=OptiPlex 3080
hw.serialno=2GLK4K3
hw.uuid=44454c4c-4700-104c-804b-b2c04f344b33
hw.physmem=34078810112
hw.usermem=33578160128
hw.ncpufound=12
hw.allowpowerdown=1
hw.perfpolicy=high
hw.smt=1
hw.ncpuonline=12
hw.power=1
hw.ucomnames=
@brynet are power-savings roughly inverse-proportional to the speed, so if one wanted to maximize battery-life, one might set the sysctl to SPE and only get L CPUs?
And a piece of me is curious what happens if I were to do something dumb like pass all the flags and block all the CPUs 😆
sysctl hw.blockcpu=LEPS
Is anyone doing okay anymore?
| I'm doing okay: | 206 |
| I'm not doing okay: | 426 |
| Other: | 106 |
@eniko
Not great not terrible
@eniko I have set the bar pretty low for myself, but I'd like to think that I'm doing okay as long as I can pay the rent, which thankfully is the case at the moment.
When the work year began on Jan 5, I had one full time and one part time game dev job.
As of end of day today, I'm unemployed.
@eniko I oscillate between ok and not ok, currently in the not category, but hopefully it doesn't last long.
@eniko I'm doing OK financially but I wish the world was a better place and I could help out more, so I marked it as Other
@eniko I answered Other since the way your question is phrased, assumes at some point I was doing okay, which I'm not sure was ever the case.
@eniko "What do you mean, "bad"?
Dr. Egon Spengler: Try to imagine all life as you know it stopping instantaneously and every molecule in your body exploding at the speed of light."
any recommendations?
@hi There are many fanless cases if you care to build your own, look at quietpc.com for inspiration:
https://www.quietpc.com/systemchooser?quietness=fanless&fanless=selected
They are UK based, so you will get hit with taxes if you let them build it for you, but they use standard components that you can source yourself.
https://pcpartpicker.com/ is also helpful to make sure everything fit together.
404 Energy Supply Crisis Not Found
@flaki yay! Do you have an overview of your setup written down somewhere? Would be interesting
@hi @asciijungle I don't, at least in part because it keeps evolving almost every day.
I have 12 solar panels on the ground, almost flat, about 5kWp in total, with two chargers and various batteries and power stations. I am building a new battery bank from prismatic cells soon which will take over the hodge-podge of the current setup. I have about 4kWh of capacity which is plenty for everything I need overnight (e.g. fridge), and can cook and even run a small electric heater.
@hi @asciijungle the panels are actually installed rather securely on mounts intended for flat-roof mounting and concrete slabs, if you discount the issues of e.g. missing grounding (soon...) they are actually fairly safe there, I run a long length of 6mm2 DC cabling to the "cottage" where all the batteries and chargers are, which is not very efficient but they produce waaay more than I can handle now anyway so it's not very important.
I plan to co-locate a solar hybrid inverter/battery box next to the panels and just run a single AC wire to the house to power everything from regular wall sockets (which will also get rid of the excess heat and noise).
https://flaki.social/@flaki/116284256507976531
It's completely bonkers to me that 12 solar panels I bought from a supermarket last fall, just plopped on the ground and hooked up to a hodge-podge of off-the-shelf parts (solar chargers, inverters and batteries) provides my small off-grid cabin/tiny house with the same amount of electricity my apartment uses on average (~7kWh).
In Estonia. In March.
Like sure there is the "ThEsUnDoEsNtAlWaYsShInE" crowd and a couple of caveats but, like how is this _not_ a NO-BRAINER for everyone? Like, housing associations of soviet era buildings like my apartment??
@asciijungle the pricing on the raw panels* is bonkers, if you have the space for them and can source them without needing to order like a whole palette they make amazing value and return-on-investment. That worked even pre-oil-crisis (and particularly if you drive an EV) but they are even more of a no-brainer now. And even up north.
___
* (professional) installation will cost you at least 1-3x the price of the raw panels and that's even before full system buildout with wiring, inverters, batteries etc.
@flaki yeah that is true. And I think its only getting worse.
In germany there is a regulation for "Balcony power plants". I think its time for me to get into that.
i don't have batteries yet, but if i did, my household would already be fully off-grid as of march, covering everything from car charging to heating and cooking.
@asciijungle @hi yeah have looked into LTO (Lithium-Titanate) because we did need heaters over our cells in the winters but they are too expensive for any significant amount of storage.
We don't/can't feed back into the grid because we are still waiting for our grid connection to be installed... we only got a 10A grid uplink though because basically the only thing I foresee needing it for is "heating and car charging in the winter".
@hi yeah I still want a small bank for testing at some point.
I think, like for all cells, your best bet is Ali Express. Some vendors have European stock (usually in Poland, Germany or Spain) to ease and expedite shipping. Some established sellers also have their dedicated shop(ify). I ordered my cells from https://hakadibattery.com who seem to have a good track record across the various stores they sell in, but can't vouch for them until I have received the cells myself.
| yes, for myself and other people: | 12 |
| yes, just for myself: | 32 |
| no, but maybe in the future: | 38 |
| no: | 38 |
Closed
@hi ah, I question I just cannot resist!
I have my own instances of #Gotosocial, #snac2, and... #honk, all for my own use.
All running as jails on a FreeBSD box with an 8 core Ryzen 7 and 32GB memory, but those 3 jails, plus a couple of others only really utilise about 4-5GB.
@hi running GTS on a little VM on some old hardware I already had.
It’s super easy to maintain and so reliable. Highly recommended.
@hi I run a slightly modded standard Mastodon instance (mostly just adjust the post length). It runs in Docker on a dedicated Debian VM along a bunch of other services.
All my services run on my home server, but I have a cheap VPS to act as an "exit node", so I get static IP without dealing with the dynamic IP shenanigans or relying on Cloudflare/Tailscale etc.
i couldn't figure out yet what to do if my #snac2 instance goes down...
@hi this :)
https://xkcd.com/303/
If it's a small downtime, fedi, just like email "catches up", I had a few hours/a day of downtime here & there, I don't care/mind much to be honest (but it's rare).
Now that I have the "cottage" I have unlimited 4G so I should set up a failover, but unless DDoSed deliberately, issues are so infrequent that I don't really bother (and since this is a personal instance, the only person to upset here is me :) )
@hi I have migrated a couple of websites to ssg7!
Love the builtin compression, the removal of files and selective updates. Very nice!!
Fucking tmux is AI slop. #openbsd #tmux #floss https://github.com/tmux/tmux/wiki/Contributing#use-of-ai
EDIT: There has been a fair amount of discussion about if this applies to upstream openBSD, and how the AI code policy is interpreted.
An example can be seen in this commit by the lead developer landing in openBSD attributed to "claude code" https://github.com/openbsd/src/commit/9c2b8e445a0bdfafdd6148b1760f00aa5429627b
@trashheap they forgot to update the difference section "GitHub holds the portable tmux version. There are a few minor differences, mostly for portability." They shall be more soon…
(T_T)
i really hope that slop doesn't make its way into #openbsd's base. lately it feels like i'd have to fork everything i care about...
@hi @trashheap
Seems to me a bit more nuanced:
Use of AI
Code written with the assistance of AI can be acceptable. However, the question of ownership and copyright of AI-produced work is not yet well-defined in law.
Given this, in order for code produced with AI to be accepted, it must either be trivial enough to be not copyrightable (basic refactoring, one line bug fixes), or there must be a public statement available from the AI publisher showing they do not assert copyright over the work.
@ParadeGrotesque @hi @trashheap
That last part shows a profound misunderstanding of the problem.
Why do you say that?
Seems to me a cautious statement, in line with the rest of the position.
You could even understand it as a way to say: release the code or state clearly that it is free of problematic license.
Which is another way of saying: we won't accept your code unless you can vouch for it. Which no slop shop is able to do.
@ParadeGrotesque @hi @trashheap
I don't think anyone is worried that Anthropic et al are going to assert copyright; it's more that copyrighted works were used to train the model so the generated code may or may not fall under the original authors' copyright.
How can you vouch for code you didn't write?
The various other AI concerns are externalities I guess.
And I believe that's exactly the point the tmux author is trying to make: you did not write the code, you cannot vouch for the code, therefore I can't accept your slop generated code.
(Unless it is completely trivial, in which case your code is probably not needed)
@ParadeGrotesque @FritzAdalis @hi
You could even understand it as a way to say: release the code or state clearly that it is free of problematic license.
Take a look at that bit of text again, specifically the text which FOLLOWS it.
It is followed by a listing of LLMs who have issued statements from various LLM vendors that they assert no copyright on the output; and are therefore "fine."
@ParadeGrotesque @FritzAdalis @hi
And I believe that's exactly the point the tmux author is trying to make: you did not write the code, you cannot vouch for the code, therefore I can't accept your slop generated code.
As an example you can see a fairly lengthy pull request "co-authored with Claude" here: https://github.com/tmux/tmux/pull/4744/changes/b700e9ce219cae63988c4287fd3cde41a6a6f8c4
AND then you can see it landing in upstream tmux in the openbsd source tree, attributed to the original author and claude here: https://github.com/openbsd/src/commit/9c2b8e445a0bdfafdd6148b1760f00aa5429627b
The ultimate committer of that code being Nicholas Marriott, the lead developer of tmux.
@trashheap
Ooh! How'd you achieve the red text?
@dick_turpin My instance runs a mastodon fork called glitch-soc, it supports markdown syntax for block quotes.
@trashheap Brilliant. The best I can achieve is Bold, italics, etc., using the Unicode Text Converter.
@ParadeGrotesque @FritzAdalis @hi YEAH, it stings; tmux has been part of my personal stack for a long time.
@ParadeGrotesque @trashheap @hi
It is the only change in openbsd-src that mentions Claude. (I think Xavier Claude doesn't count here.
@FritzAdalis @ParadeGrotesque @hi Agreed. Though the whole proccess of tmux portable pull requests getting filtered through nicm if/when they land in openbsd; makes the whole thing hard to track.
AND it looks like there are a couple of recent pull requests in downstream portable tmux that mention Claude; which are just newer, and have yet to filter up stream. (Assuming they are accepted.)
@hi as I understand it, OpenBSD already said AI code was not acceptable (introducing notion of copyright). The tmux project that accepts AI is the portable one. I don’t expect code from there to go back to OpenBSD src. Unless it is something trivial and acceptable from their POV.
@joel @hi Here is a commit crediting claude code landing in the openbsd tree upstream. https://github.com/openbsd/src/commit/9c2b8e445a0bdfafdd6148b1760f00aa5429627b
@trashheap 🤷♂️ maybe it was something identified as acceptable. I only have uneducated user opinion.
@trashheap Should any tool, or anything, that helped in in some way produce the code get attribution? Currently it seems like people are giving the company/product free advertising.
@aslakr For those who would like to use software, made entirely by human ingenuity; such labels are valuable; and seem to be the "middle ground" a number of large FLOSS projects are settling on, as an example the linux kernel's draft policy requires labeling.
#RunBSD Recent obsession is about reusing #arm64 boards that sleep unplugged.. So today, here are two sets of notes regarding
#OpenBSD support.
Running on the #ODROID HC4 is pretty decent now. It already runs the DHCP, DNS and the WireGuard gateway. More is to come, if everything goes well.
https://www.tumfatig.net/2026/running-openbsd-7.8-on-odroid-hc4/
Running on the #Raspberry Pi Zero 2W is fairly unstable as-of now.
https://www.tumfatig.net/2026/openbsd-7.8-on-raspberry-pi-zero-2w/
sd2?# dd if=u-boot.bin.sd.bin of=/dev/rsd2c bs=512 skip=1 seek=1
# dd if=u-boot.bin.sd.bin of=/dev/rsd2c bs=1 count=444
@hi short answer is "because Mark Kettenis told me to" :)
From what I understood, it grab two parts from the u-boot image, the boot-loader and the firmware. I'd rather have grabed file and do regular copy/paste on a filesystem but that doesn't seem to work this way.
@alderwick yeah, I once learned the consequences the hard way 🤣
I guessed as you say, it just overwrites the necessary bits. That’s why this part is only done for installation. And then a full dd is done when the sdcard just needs the u-boot stuff.
How many computers do you have within arms reach right now? (If you have to ask: yes, it's a computer)
| 1: | 109 |
| 2: | 118 |
| 3: | 108 |
| 4: | 77 |
| 5: | 50 |
| 6: | 37 |
| 7: | 18 |
| 8: | 17 |
| 9: | 8 |
| 10+: | 41 |
Closed
I'm at 10 computers within arms reach if I'm sitting at my desk at home and I need validation that that's not abnormal.
So far it's not looking good.
Make that 11. I just found another.
Nevermind. 12.
Uhhh. 13.
Ok final list.
Wait that's 14.
@jp Four cellphones, three SDRs, two Pis, a couple of HackRFs (more SDRs?), two laptops, a Sony Vita, a Nintendo 3DS... probably a couple of other things under the piles that I've forgotten about. I may have a problem.
mysterious MNT Pocket Reform mainboard 2.0 is real, as evidenced by this JLCPCB confirmation photo :0 hope i didn't make too many mistakes and the bringup will go smoothly once it arrives...
@mntmn me too, I think I hosed mine :/
@grimmware oh no? how?
@mntmn it started with a keyboard flashing accident (I ran an older script and it accidentally wrote the sysctl to the keyboard) and then I spent ages trying to get the sysctl to power on by sending it serial messages but it never seemed to come up, so I’ve reflashed the firmware (it comes up just fine as a mass storage device) but I’m not getting any output over tio :/
@grimmware reflashed the firmware from another computer? and is the programming switch turned off afterwards? otherwise it won't run the fw and always come up in usb bootloader mode
@mntmn
Neat, is that a compute module socket like the Jetson or Turing Pi 2?
@FritzAdalis nope!
@FritzAdalis maybe i misread your question. so the answer is kinda, but not compatible with those you mentioned. this page has a table https://mntre.com/modularity.html
44 today!
Yesterday at the playground, my daughter gave me some flowers 🥰💖
This weekend has been full of arm64 joy. I’m happy!
@hi haha, nothing terrific. Just ran OpenBSD on my ODROID HC4 and Raspberry Pi zero 2W and they seem to work pretty well. Not sure why it failed so badly a year ago.
@passthejoe more or less the same way as previously https://www.tumfatig.net/2023/running-openbsd-on-raspberry-pi-zero-2-w/
Gonna write a new post so that I remember how I did. When GtS will be finished compiling.
Here is the deal, it's become apparent that I can't get rid of #AI slop code from out of all the software I use on a daily basis. Even if you avoid it in your kernel or your init system, it's in mesa; it's in the python programming language. It's coming for a dependency somewhere in your stack..
However, this is the new line in the sand for me. I don't support you anymore when you start embracing AI slop code. I don't donate to you, I don't take time to learn how to use your project.
If alternatives exist, ill pivot on the work I do, on the choices I feel free to make. When no such alternatives exist, and you embrace AI slop code your project becomes just an unfortunate dependency Id rather not think about.
If you want to attract new users to your software, if you want my donations; if you want me to champion your project, file bugs, if you want me to learn your programming language; just reject slop.
Thats the deal.
@trashheap do you say all code generated with AI is slop, or only certain code?
@dentrochat My big pet ppeve issue with LLM generated code, is that it's derivative code of the FLOSS code in it's training set AND since no current LLM respects the license of the FLOSS code in terms of attribution or copyleft; it's all a license violation and useless, regardless of quality.
So I would say all code, as it is generated today.
@trashheap @dentrochat the technology is genuinely useful for certain circumstances. If a theoretical new llm & training set comes out in years to come with ethical data sourcing and attribution, would you support it at that point?
And people have been reading FLOSS code for years, and applying what they have learned to other applications without attributing the source of their knowledge? Would you call this out as unethical? Because a machine learning from FLOSS code isn't all that different.
I'm more concerned of slop that comes from over engineering, lack of testing, disregard for security, and unreviewed code. I don't have a problem when llm generated code is used for hacky scripting, targeted functions, or as a learning tool.
Curious on your thoughts...
Yeah the differentiation between someone manually recreating floss code or via an LLM probably yields quite some discussion.
@dentrochat @alanxoc3 I am afraid the idea that LLM Chatbot learning is "like" human learning is a bit fallacious. I have described the error in thinking at length here: https://tech.lgbt/@trashheap/116318493939878305
AI and Logical Fallacies (multipart) [SENSITIVE CONTENT]
I want to talk about the fallacy of division.
Fallacies can be thought of as common logical traps, people fall into. They appear to be in the form of a rational or logical argument; but it's a bit of reasoning that is faulty or unreliable; and isn't necessarily true.
Logicians and philosophers have cataloged and named fallacies. The fallacy of division is what happens, when you attribute something which is true, of the whole to some, or all of it's parts.
Examples:
Congress is dysfunctional; Therefore every member of congress is dysfunctional.
Free and open source software is better than proprietary software; therefore every piece of free and open source software is better than it's proprietary equivalent.
This cake is delicious; therefore every morsel of it will be delicious.
Notice this last one, is potentially true. A logical fallacy is a fault w/ the reasoning, not the conclusion. Fallacious arguments have POTENTIALLY false conclusions.
(continued 1/3)
uIf you're an nvi user what other tips or hacks are there that I should know about ? I don't suppose there is some magical way to theme it ?Undo the last change made to the file.
If repeated the u command alternates between these two states.
The . command when used immediately after u causes the change log to be rolled forward or backward depending on the action of the u command.
@justine not sure if nvi have registers
https://codeberg.org/ditchgithub/templates_and_code_snippets
vi/nvi has "registers" (sometimes confusingly called "buffers" which leads to semantic-overloading since "buffer" can also refer to the document-storage in RAM), so you can do things like
"jdd
to delete the current line into the "j" register/buffer, and then
"jp
to paste the contents of that "j" register somewhere else. Likewise, you can execute the contents of a register as a macro with
@j
though note that it's a bit trickier to populate them with useful content because of line-endings (if you yank/delete linewise into the register, you'll have a trailing newline in your macro; if your macro includes newlines and you don't want the trailing newline you have to take extra pains to includ just the ones you want)
https://pubs.opengroup.org/onlinepubs/9699919799/utilities/vi.html#tag_20_152_13_43
@justine I feel like u for undo has been in vi for as long as I can remember...
Though my memory is rubbish so I am most likely wrong.
@justine OpenBSD vi should be nvi, or derived from it, at least based on my recollection and the vi man page:
The ex editor first appeared in 1BSD. The nex/nvi replacements for the ex/vi editor first appeared in 4.4BSD.
My understanding is that original vi only supported one level of undo, and that the "." operator didn't perform additional undos going further back in time. The POSIX standard¹ seems to back this up, not listing "u" among the repeatable commands.
However, multi-level undo is quite useful, so most later versions of vi/vim have included some way to undo multiple levels back…as you note, vi/nvi use "." to keep going back while "u" undoes the undo; meanwhile vim defaults to "u" going back multiple levels of undo, and control+r redoes them (and that doesn't get into the whole undo-tree going back/forward in time). That said, vim allows you to change to the POSIX way if you prefer:
:help undo-two-ways
:set cpo+=u
⸻
¹ https://pubs.opengroup.org/onlinepubs/9699919799/utilities/vi.html#tag_20_152_13_35
@justine vi in OpenBSD *is* nvi, I thought. It's hard to imagine them digging up Joy's code and including it for fun.
But of course if they've done something interesting there, I'd love to know about it.
@RussSharek @justine you beat me to it. :))
@mischa @RussSharek @justine To quote from my wishlist[^1] of things I would like to have:
"A minimal vi(1)-like editor. What I have in mind is pretty much plain vi/nvi, just a little less minimal – something like vi + Unicode support + multiple undo + syntax highlighting (the last one is debatable). In particular, it should be a lot more lightweight than Neovim or even Vim; I don’t care about extensibility, for instance."
I have used nvi more or less exclusively for a couple of years; it is just a little (though really not much) too minimal for my taste.
modern vi/nvi ticks most of those boxes (more minimal, something like vi, Unicode support) just missing the syntax-highlighting. That said, if you want ephemeral colorization, you can pipe your buffer to bat(1) or pygmentize(1) to display particular regions like
:'a,'bw !bat -l awk
:'a,'bw !pygmentize -l py
@gumnos @thorstenzoeller @mischa @justine
The pygmentize trick was one I tried, though it hadn't occurred to me to try it with marks like this.
I am giving life without syntax highlighting a go, and so far it's working better than expected.
Still learning to think of the editor as a shell tool, rather than a sort of meta shell on its own.
BTW: There is a fork called neatvi which apparently added highlighting as a feature. I've not played with it though:
CC: @gumnos@bsd.cafe @thorstenzoeller@exquisite.social @mischa@exquisite.social @justine@snac.smithies.me.uk
I tested it with nvi on Slackware using source-highlight from SlackBuilds.org:
:'a,'bw !source-highlight -s sh -f escHowever nvi doesn't render ANSI escape codes in its output pane out of the box. The issue is in filter_ldisplay() in ex/ex_filter.c — it sanitizes output through KEY_NAME() which converts ESC to ^[.
I'm working on a small patch that detects ANSI sequences and switches nvi to ex screen mode before writing, so the terminal renders colors correctly. Early results look promising!
CC: @thorstenzoeller@exquisite.social @mischa@exquisite.social @RussSharek@mastodon.art @justine@snac.smithies.me.uk
Hah, I was just in the middle of composing a reply to similar effect.
I just discovered that vi/nvi seem to consume ANSI-sequence output from executed programs.
I usually use bat/pygmentize from within ed(1) and assumed they'd work the same, but testing them just now, it was an exercise in frustration watching bat/pygmentize emit the right things (I could colorized pipe output to hexdump and see the right output data, but when vi/nvi received it, it stripped the ANSI sequences off).
Sorry for the minor boondoggle! (at least until your patch is in place ☺)
CC: @r1w1s1@snac.bsd.cafe @thorstenzoeller@exquisite.social @mischa@exquisite.social @RussSharek@mastodon.art
@justine @gumnos @r1w1s1 @thorstenzoeller @mischa
I see my being slow to tinker has saved me some headaches. :)
Thank you for the heads up.
CC: @thorstenzoeller@exquisite.social @mischa@exquisite.social @RussSharek@mastodon.art @justine@snac.smithies.me.uk
Happy to help!
Truthfully, it feels really nice to know that I'm not the only one who's seeking out and enjoying simpler tools.
Speaking of help, @gumnos also offered me a lot of help recently, including with a mapping for opening markdown links in splits. Creep back on my posts a couple of days and you'll find it. :)
Want to see some cute gadgets?
Come to the #CLT2026 Plan 9 booth!
We have a PinePhone (Allwinner A64) running 9front, and a little Lichee RV (Allwinner D1) that you can cpu into!
@a @jrsharp That one is running Linux with https://github.com/u-root/cpu - on top of oreboot, that is.
I looked a bit into Moody's WIP port of 9front (branch `riscv`), and dug it out again on Saturday. I'd love to get it running on the D1 at some point, but we're not there just yet, sorry.
Wanna help out? It's one of my favourite platforms. The 512MB vatiants works well with upstream oreboot, which is my personal main focus.
@justine I've stuck with the built-in vi on OpenBSD for most tasks. I miss some vim-isms (especially things like ci") but I don't think nvi2 would help with that.
I'd like to use emacs more (heathen) but I find that I end up tinkering with my setup more than doing productive work. The only time I've bemoaned the built-in's lack of utf-8 support was when I tried futzing around with my .XCompose file, but ibus has been a buggy pain in the ass, so even that's not much of a loss.
@justine For quick edits, base nvi is enough for me. I switch to Emacs for heavier work or prose (where I need Unicode occasionally), but if I was without Emacs then nvi2 would work just fine. I've kicked the tyres a little (debating a move myself) and I like it.
@justine That doesn't look like my boring cwm setup. Is that sway-something?
@justine was about to ask the same! It looks somewhere between the shiny of Hyprland and the simple of Sway; looks great!
Hey uh, can you all help me with a very important debate?
Thanks!
Which is the best canned fruit:
| Canned Pineapple: | 4 |
| Canned Peaches: | 5 |
| Other (Please Comment): | 1 |
:(
i use openrsync, but still we don't want to lose corner stones like rsync, curl, and imagemagick...
@hi what happened with curl?
https://git.sr.ht/~r1w1s1/code-notes/blob/main/notes/Using_NVI_in_2026.txt
In case you're not already in each other's circles, @RussSharek has been recently sharing adventures in #nvi ☺
@TangentDelta at Discord has UF #forth running on the #m5 cardputer using #uxn
Now that web search engines are full of LLM-written articles, are human-curated Yahoo!/dmoz/Google Directory/awesome lists-style web directories the future? 😀
@bogdan_ov this sound super cool. I didn’t see a .tal file in the repo. Am I missing something?
Backup all the things with #plakar and #OpenBSD !
https://x61.sh/log/2026/03/25032026151800-plakar.html

Devils on the Moon Pinball is out!!!!
What a time to be alive!!
get it here: https://play.date/games/devils-on-the-moon-pinball/
Only on
@playdate
#pinball #indiegames #pixelart #playdate
testing: if you are a mastodon user, can you see the poll? does it have six options? can you vote for all six options at once?
| option 1: | 25 |
| option 2: | 25 |
| option 3: | 25 |
| option 4: | 26 |
| option 5: | 26 |
| option 6: | 25 |
Closed
Do you also sometimes obsessively work towards realizing an idea/project despite it negatively affecting many other aspects of your life for a while? If so, have you found ways to make that burst of creativity more healthy?
@gustav very often, it helps to have people around you that ground you and force you to do things you may not want to do while high on your own brain fumes. Also deadlines can be a good motivator to snap out of it. Still, sometimes the hyperfocus happens and I have no control over it, sometimes I forget to eat until the end of the day, at those times I try to go with the flow and be kind to myself.
@hi Awesome. Are you using a mastodon client such as phanpy or elk? Is everything mostly working smoothly for you?
i used to use ivory app on ios, but it was a bit buggy. everything else is working smoothly.
on openbsd don't forget to increase open file limit; 4096 is working for me.
@hi Thank you for the feedback! Yes, I've increased the open file limit :) Maybe I'll learn to love the native web UI!
@grunfink Hello! I've set it up yesterday and it's been super easy (fastcgi behind httpd), but some things that feel like they should be working from third party clients don't work, like attaching images to posts. So I'm wondering what setup people use!
@grunfink Hmm, my issue with posting images from Phanpy (or other Mastodon clients) is with the GET endpoint, after I submit a post with attached images, the GET request to `/api/v1/media/post-d7891f3f5d40f8dca38f3581784e1bac.png` returns 404 (even though the png and txt with the image description do exist on the server, eg have been uploaded properly.
@grunfink I mean, it does include the host of the instance, so `https://social.gosha.net/api/v1/media/post-d7891f3f5d40f8dca38f3581784e1bac.png`. What I'm trying to figure out is why snac is responding with 404 to that.
his vim classic is exactly how software should be built: no new features.
meanwhile i'm switching back to vi ❤️
Posted a long rant about #Apple and how they lost me after being a user for 26 years:
Explicitly non-stick pans, as opposed to seasoned cast iron or whatever, are…
| …wonderful.: | 0 |
| …uniformly trash.: | 4 |
| …mostly great if you buy the right kind.: | 1 |
| …more hassle than they’re worth.: | 3 |
Closed
@a I think you need a “contain dangerous chemicals” option
@TracyTThomas Yeah, I think that’s mostly a subset of the 2nd or 4th set, possibly depending on the specific type of non-stick.
@joel @jp There is a custom open source firmware if you are not aware yet, looks promissing: https://github.com/crosspoint-reader/crosspoint-reader
@fredy_pferdi yep, it's what I installed on mine!
@joel just looked up that book, how had I not heard of this series?
@theTangentSpace I had it in my collection for a while because of a Humble Bundle, decided to start it on this!
Beware of spoilers btw
@hyde @joel I just uploaded one here!
https://polymaths.social/@jp/statuses/01KMG3MG2EA0V2JD6G1XK89FP1
@joel @jp I love the small size of these. I have also seen people attach them to the back side of their phone. I'm not sure how practical that is, but I at least like the idea of it. Haha!
I definitely plan on picking one of these up at some point. I looked a while back but I could not find a good US distributor without a ridiculous markup.
should i get one too? no
i don't read from screen that much and i have an embarrassing amount of unfinished projects already (~.~)
WELP I can no longer say that #FreeBSD's draft policy on LLM code contributions was leaning towards #NetBSD's way of thinking of banning the slop code entirely. I was going off what was said at last year's BSDCan.
Apparently it's shifted since then. https://reviews.freebsd.org/differential/changeset/?ref=1487182
Hopefully it shifts back, before it's finalized. I feel a bit crushed. I've been talking about that prior draft policy as a positive indicator for months. Im realizing I had pinned a lot of hope on it.
I've been trying REALLY hard to stay away from software with LLM generated code. (sigh). Ive dived back into FreeBSD hard after many years away, and have even been trying to contribute some stuff to ports.
I guess if this goes south, ill be re-evaluating things. Maybe NetBSD? Though I know it's missing some things from pkgsrc that will hurt to do without. #bsd
@trashheap There's certainly things missing from pkgsrc, do you have specific things in mind ?
I'd think the things a FreeBSD user would miss most on NetBSD would be jails and related, or possibly Wayland or the more recent OpenZFS features
@tfb Honestly number one my list is the lack of signal client.
@trashheap Oh ouch, that's probably not a fun thing to get ported, being electron and all. I've managed to largely avoid Signal, but that depends on who you're communicating with of course.
I have no idea if the Linux version runs under binary compatibility, but that could be worth exploring
@trashheap before this heads off on rampant speculation, the Project doesn't yet have consensus on how to handle this, & we are working towards one. There are 2 drafts in circulation that I'm aware of.
Within the Project there are at least 3 groups:
- No AI (absolutists)
- Pro AI (accelerators)
- everybody in the middle
Yesterday's core meeting (which I attended) agreed to:
- publish a pragmatic policy for the interim
- consolidate the existing draft options if possible
- if not possible, at provide 2 options that can then be voted on in our next formal Elections, and discussed prior
There is no question that AI is very controversial, and also that FreeBSD the Project has committers who will have differing views across this space.
From the wider environmental picture, through concerns about copyright issues & licence erosion, and to the very real impact on code review effort from AI-generated or AI-assisted systems, and how AI is being used extensively by potential newcomers to the project, it's very easy to pick a side, and then be outraged that the Project hasn't taken *your* side.
@dch @trashheap different policies for different parts of the source? a pretty conservative stance for kernel/drivers, and a lean one for tests/docs, etc.
@charlesrocket that’s a good idea, but could a single license still cover this? The license is a critical part of BSD history and our culture @trashheap
@dch there's already a separate licence for FreeBSD documentation:
@grahamperrin it’s still the BSD 2-clause version tho @charlesrocket @trashheap
@dch @grahamperrin @charlesrocket Ive missed the thread on something. What does the license have to do with AI/LLM code-commit policy?
@trashheap If software is missing in pkgsrc, then contributing recipes would be both rewarding and improve software availability for everyone 😉
Ok, now that Beastie accepts AI generated code (once reviewed by a mentor yada yada), the question is how do you actually identify AI code. I mean, once it works as expected. How can you tell an AI code from a newbie code - that did a bunch of copy/paste/adapt from Stack Exchange for example?
Build once run
Everwhere™
#uxn
The game feels more "real" only because it's not on my laptop screen.
"We use debian, that should be age verification enough"
@h3artbl33d @eris2cats old Puffy
@h3artbl33d @eris2cats not calling you Puffy. I am calling you old Puffy! 
doing a brutal last-minute change for MNT Reform Next: primary M.2 SSD now full-size (2280), secondary M.2 SSD (2230/2242) now on the back of the mainboard. microSD Express gone, but could still live in an adapter in any of those 2 slots if needed.
this is a cost-saving measure on the one hand (2280 are more broadly available and affordable than the small sizes these days), but on the other hand allows you a lot more flexibilty when choosing fast storage options. you'll also be able to do RAID stuff between the 2 SSDs, of course.
now the bad news: i didn't anticipate the recent SSD+RAM apocalypse last year. the 32GB RAM+256GB eMMC are safe because i bought them early. but i didn't buy all SSDs early enough: even with the switch to 2280/being able to combine stock of different sizes, the SSDs at promised capacities would cost us 33514 EUR over budget. we don't have this extra cash, so the only option is to reduce the SSD capacities: we can ship 1TB instead of 2TB and 512GB instead of 1TB. i'm extremely sorry for this.
everything else about the Next is going well though. we fixed many small papercuts that i experienced while daily driving the laptop and have many parts already stocked or ordered. i have a draft of a more detailled Crowd Supply update that i'll finish over the next few days.
for reference, SSD price development graphs @ PC Part Picker https://pcpartpicker.com/trends/price/internal-hard-drive/#storage.ssdm2nvme.2000
i forgot to say that with this new plan, the half capacity SSDs are still more expensive than the original ones (approx 5158 EUR over budget).
@mntmn thnx for all the work; I too am flex in regards to the SSD if it helps you to sustainably deliver the next without jeopardizing your company/team
@mntmn ouch!!
@mntmn i remember hoping that 2020 would be the last supply crunch ;;
@whitequark yes me too. last year i was like, finally things are getting better, no new crisis in sight, and then *bam*
@mntmn
Yeah, no one thinks you’re immune to the rolling catastrophes of stupid. We all know you’re going to make the best judgments.
@mntmn As long as you don't solder the SSD on the mainboard and encapsulate the wohle mainboard with 1 cm thick epoxy, I don't mind and looking forward to the Reform next. :)
@mntmn omg love it 😍 it's so great that's the only huge point I was worried about not getting an affordable m.2 with much capacity but with 2280 it will be much simpler.
You and the whole mnt team did a great job
curious about automating...
@hi @justine I have one script that starts with Xenodm. It checked which monitor is connected and activate only the laptop one (if external is not connected) or the external monitor, when it is there. I can also call it with a keybinding in case I disconnect the external monitor for some reason. xrandr is a great tool!
@hi or use hotplugd if you can identify some device from your monitor. Mine has a USB Hub so I get a lot of messages when monitor is (un)pluggued.
EDIT: like so https://www.tumfatig.net/2024/automatic-display-switch-for-openbsd-laptop/
Cc: @justine
@justine I too am wondering about monitors, though in my case I’m suddenly yearning for a small-ish 4:3 monitor instead of my 27” 4K monitor. (And not just because it’s impossible to use when installing new operating systems and whatnot.)
Aside from not running macOS, the Framework laptop I have only beats out my MacBook Air in the display department, because “tall and narrow” beats “short and wide” imo
I'm looking for full-time work!
I work at the intersection of social and technical systems, and specialize in building up people, programs, partnerships, and organizations around open source.
I have a deep track record in complex community relations, am fluent in the nuts and bolts of many technologies, and have spanned governance, org development, nonprofit and people management, comms, marketing, events, and beyond.
Let's fly! 
i turn 40 today
please use this pii to theft my identity, and when you find it please let me know where it went, the little scamp :)
@spiralganglion Happy birthday. I believe at this age you are no supposed to completely reinvent your identity, anyway.
(Aside: I think 40 was also when I gave up any pretense of hiding my birthday from online posts like it was some sort of secret password. Still working on the mid-life identity overhaul.)
@AmeliaBR thanks amelia! i'm spending the day in edmonton (going to the citadel, etc)! thanks for keeping the city cool while i was away :)
@spiralganglion I hope you had a good one. The weather wasn't too bad for March.
Did you do Wizard of Oz at the Citadel, then? How was it?
@AmeliaBR Yep! I loved the costumes and dance routines in particular — fantastic work. The rest was fine, nothing spectacular, but still enjoyable.
Sums up my experience growing up
@ilovecomputers I feel like the core of what we loved about the internet is there - it's hiding in chat rooms, little closed member forums, hand crafted websites. It's not gone, just harder to see, but if you dig through the muck, you find yourself in a small meadow with a few other folks who might share with you something good.
Any recommendations? :)
I agree that the web is usable and pretty nice using RSS to follow blogs. I also love blogs with a comment section with the same community of people discussing the topic.
What I miss somewhat are dedicated web forums that are active. Something like Head-fi and Steve Hoffman's forum for music.
I wish for forums like that for other topics, as well.
@mutkitta @triptych @ilovecomputers Here’s a ton of resources for discovering indieweb stuff out there on the web - https://shellsharks.com/indieweb. Happy surfing!
@shellsharks @mutkitta @triptych @ilovecomputers hum, cool and thanks 🙇 I've just finished update my website ( https://benjamim.neocities.org )with humans.txt and robots.txt
BLACKS RULE but unironically and said better » 🌐
@albinanigans@blackqueer.life
https://ourfavoritevoid.club/directory is a webring for smolweb peeps that like to blog (or like black cats, or both)! Check out some of the folx there! We've also an old-school phpbb forum, but unfortunately would probably not qualify for your definition of "active."
https://32bit.cafe is just fantastic, too. Lots of tutorials and smallweb denizens to connect with.
@mutkitta Look up:
https://melonland.net/ and its forums - plus all the "handy links" on the page
https://www.naiveweekly.com/ (yes I know it's based on Substack)
https://goodinternetmagazine.com/
Browse some webrings: https://brisray.com/web/webring-list.htm
@sarajw
Adding to this: https://ooh.directory/
@mutkitta @triptych @ilovecomputers
@ilovecomputers And the reason for all of this is "Monetization".
Things used to be different because people put stuff on the internet because they wanted other people to see it, not to sell it.
People using their real names and faces
Governments wanting to control it
Corporations flooding it
Just a few things from the top of my head that made it all go to shit
It was certainly a group effort
@ilovecomputers I feel this in every fiber of my being. I weep for every generation that has come after me that they will never know the peace of just getting to be without being tracked, monetized and fed into a perpetual rage machine.
https://social.lol/@triptych/116276727579512933
To be less of a downer, I want to pin this reply thread on this post as it contains links to indie websites and communities that continue to live on. It’s not just nostalgia; even amongst the next generation, there’s growing enthusiasm for slow tech.
@ilovecomputers I feel like the core of what we loved about the internet is there - it's hiding in chat rooms, little closed member forums, hand crafted websites. It's not gone, just harder to see, but if you dig through the muck, you find yourself in a small meadow with a few other folks who might share with you something good.
@ilovecomputers fr. that said - look up web revival 😍
@ilovecomputers This is the experience of the new "Lost Generation".
What's more crazy is knowing how deeply the media we consume shapes and informs us, and thinking about what teens and 20s today are growing up with. X_X
(I guess to be fair... What all of us are continuing to grow up with... But yeah, earlier formative experiences, chances to connect with people who expand your horizon, etc... dang)
@ilovecomputers Yup. Mine too. The fediverse gives some hope that we might get some of that old vibe back.... But it won't ever be the same.
@ilovecomputers
i just can’t convey the frustration and sorrow that it’s been to grow up at first without the internet and then watching it bloom into this useful, fun, connecting force you sometimes spent time on, only for it to degrade into this constant oppressive waste of time and energy where people are constantly pumping out algorithmically designed content for max algorithmic appeal and even the most simple search generates either no results or an unimaginably bad ai generated slop none of which is usable or correct. we briefly had a library of alexandria and then fed it into a paper shredder so advertisers could sell a random mass of pulp back to us at a premium.
It’s almost like you wanna go back to a library. You know, people building stuff they like or create and for local business
It really calls for a municipal utility network that serves the residents of a location of a city or town, that provides Internet access, web posting services and a data center. Librarians would be recruited to provide information services to structure things usefully.
Each city could then federate with other cities to create a larger network.
@cassolotl TY! Searching for a tumblr post is such an utter pain.
@ilovecomputers You're welcome! I just google an interesting phrase from the post in quote marks and it usually pops up in the first three results. :)
https://eldritch.cafe/@cassolotl/116288484685754164
A big kudos to this person for finding the source
I know the guy who invented it gave it to us all with good intent and selflessness, but had he kept a la Jimmy Swales and prevented it from being colonized by entities hellbent on making money off it, we'd be better off.
Maybe those companies would have created their own facsimile. That would be fine. Have a profit-driven internet separate from the humanist internet.
Holy shit my IRC pebble intergrations I wrote actually work??? Woo!!
(it get a notification when my name is mentioned)
@angelwood OMG PLEASE SEND THE CODE
@ohnoitsnoah it depends on the IRC client/bouncer, but im using a little tiny python script running ntfy + a super thin soju client. and then pebbleos just intergrates perfectly with ntfy.
but any irc bouncer should have a way to hook into it, and the ntfy api is just one curl call!
i would recomend having a irc bouncer so it can run 24/7 ^__^
Home-made passport photos for citizenship application
“I won’t be on Bluesky, but my book will.”
“I have to find readers and reviewers. There are readers and reviewers on Bluesky.”
“Since I’m using my book promotion as an example of my marketing skills, it needs to be where industry people are.”
This is how I’m coping with something I really don’t want to do. So, tell me, will I be wrong for posting my book promotion over there?
EDIT: For that 2nd option, maybe think “It’s not ethical.” There’s bias in how I wrote it.
| Gotta do what you gotta do: | 61 |
| You’re a traitor to everything: | 3 |
| It’s complicated. Let me explain 👇: | 7 |
Closed
I struggled with this a great deal myself. I quit major platforms for my values, and haven’t regretted it. I left a lot of folks behind. It took me a very very long time to build up a community to replace the ones I walked away from.
And all I can say is not a single person from those platforms followed me over to my new homes. Not. A. One.
Intransigence is real. Inertia is real. Social media fatigue is real. The sad reality is the people you are trying to put your stuff in front of won’t be coming to Mastodon for you.
Having a presence on *any* platform isn’t as effective as it used to be, for lots of reasons.
I use Bluesky because I know (thanks to our capacity to view the last time each of our followers were last active) that half my followers stopped using Mastodon. So I’m here for different reasons.
I also know that Mastodon is a much more global platform, which doesn’t do me any good if I’m trying to sell my art as I can’t sell it internationally from the US right now. So again, I’m here for different reasons.
Finally, for the most part, most Mastodon servers emphasize sustaining community over growth & reach. Other platforms have different goals.
My biggest advice for you is to continue recognizing why you use a platform over another and why you don’t. That will make you a better user. Lean into that. Don’t try to make it something it will never be or apply expectations it will never fulfill. Embrace what it *is*.
40 pull ups at the gym this morning for my 40th birthday.
@neauoire Happy birthday! Disclaimer: I'll surely not manage 50 for my coming 50th. 😅
@dwardoric yeah, I don't know how sustainable a job that is to match year and reps X) That being said, I couldn't do pull ups 10 years ago.
@neauoire A couple of years ago I met a guy at the gym who was 82 and did pull ups like a sewing machine. So it seems possible. 😉
@neauoire damn, gotta work out even more to keep up your pace
Happy birthday! And Congrats! :)
@neauoire happy birthday! Welcome to the club, I am just 2 year ahead myself. Well, nearly 3 I guess. And yes, the pressure to keep fit does feel real at this point 😁
Happy 40th and I look forward to hearing that you manage 50 pull-ups for your 50th...I'll be 81 if I make it that long but I'll still be envious of your youth. [hugs]
@ccohanlon I might not, this might be peak for me. I don't think I could have done 30 when I was 30 X).
Wait, you're not 80!? Are you seriously 80? If that was not a joke, you look way younger.
@neauoire NO. I will be 80 when you turn 50 😆. I'm 71.
@ccohanlon Ah! damn that makes me sense, I was starting to think I might just know the perkiest 80 year old.
@neauoire 40 pull-ups straight? 🫨🫨
Did two more sets, so I'm at 60, and it's only 11am. I wonder if I can get to a hundred.
@neauoire how long did it take to learn how to handstand? You inspired me to try this at home.
@alex27 this is probably going to sucks to hear, but.. three years.
@neauoire not at all, I've decided to do that, and only then checked internet and saw numbers: 780 days, one year, and so.
It was crazy from the first point of view, but I'm writing a bird each day for 3 years already, and planning to do that 7 more years, so feeling of time shifted a bit. I hope that I will be able to do handstand at least in 4 years.
Thank you a lot!
@alex27 that's such a good way to think about this, it just takes time, lots of time. But the process is fun, so, if you can enjoy the process. You'll be able to handstand before you know it ✊
eighty
hundred!
Probably was a stupid idea,
let's find out how stupid tomorrow morning.
Not as sore as I thought, although, somehow I hurt my foot with these shenanigans.
@neauoire that’s how it starts! One day you’re out on your skates, doing 100 pull ups, then the next your foot hurts, or is it your shoulder that feels weird? Wait what’s that tightness in my lat, oof gonna need a minute to rest that one…
Happy birthday!
@neauoire happy barfday !
@fleeky couldn't do a muscle up for my 40th :(
@neauoire as long as you are still pushing for it is the more important thing !
i finally started training for planche handstand , i am not even close but i think of you and the muscle ups often for inspiration! "other people are working hard and i will too!"
@neauoire I went the opposite way in number a few years ago: a one-arm pull-up for my 50th :) 🎂
@jack goals, right there. I find that if I stop doing daily reps, I lose it SO FAST. I really hope that I can do a one arm pullup when I'm 50. It's so easy to get hurt with those.
Reached that bit of gamedev where I write a program to play the program.
My mind is blown. When I first got this HP hybrid tablet, one of the first things I tried was installing OpenBSD. However, the installer (7.8 amd64) would never boot all the way, rebooting the machine around the time it was detecting the CPU cores. I tried a lot of different Linux distros and other OSes, and eventually settled on Fedora KDE as it had complete support for the device, including the touch screen and digitizer pen, the detachable keyboard and its battery, and so on.
Well today, in a fit of boredom, I flashed a USB drive with the OpenBSD installer and booted it on the tablet, and inexplicably it finished booting and got me to the installer prompt! Of course I immediately blew away the Fedora installation and installed and configured OpenBSD to my usual desktop, Xfce. I'm typing this now on the tablet, everything is working except it doesn't detect the secondary battery or the digitizer pen. I can live without the keyboard's battery gauge as it is always secondary to the tablet's battery anyway, and I never use the pen so that's no loss. No Bluetooth either but that's a given with OpenBSD on any hardware.
Needless to say I am thrilled and amused, and I'm going to keep OpenBSD on this thing out of pure spite.
Have you ever picked a lock?
edit: a practice lock is any lock that you were picking purely for practice which was not securing anything. a lock in the wild is a lock that was in place to actually secure something (love locks count)
| yes, a practice lock: | 376 |
| yes, a lock in the wild: | 296 |
| no: | 441 |
| show results: | 13 |
Closed
@eniko yeah, a (really really cheap) safe I had some documents I needed for the next day and I couldn't find the key, time for practicing the thing I watch on youtube! So I guess it counts as a practice lock.
@eniko
I get paid to pick locks in the wild. I'm essentially a professional burglar (locksmith), just call me Bilbo.
@eniko A coworker managed to lose the key to his office drawer inside the drawer itself (I don't remember how), so I tried to pick it and succeeded (it was a fluke, I never managed to do it again even though it's a very low quality lock).
That time the card reader on the office maglock malfunctioned and the backup control was inside the office was more interesting. I had to cut a wire to get in (I chose a red one for tradition), don't know if it counts as picking a lock or not.
@eniko 100 years ago, @steggy was doing geology field research (ok, she’s not THAT old, but it was a long time ago). There were these sample stations in a state park and we were supposed to drive up an access road to get to them and collect some readings. The park officials knew all about this (it was in conjunction with the local university). They were supposed to leave the gate unlocked at the beginning of the access road. They didn’t.
So it was either a several mile hike on foot, leave and come back later, or I could just pick the master padlock on the gate. It’s a master padlock. C’mon. So, yeah, a few seconds later we had the gate open and we went and gathered the readings. Left the gate locked, just like we found it. This was ok because it was in the name of science! 😛
@eniko Does exploiting a weakness in a suitcase's combination lock count? (Reducing the complexity from trying 10^3 to 10*3 things to try.)
@eniko not pin and tumbler locks, but I’ve successfully extracted the code from two different multiple-dial locks, one a Master padlock and one an official Xiaomi scooter lock. Just close your eyes, apply some pressure, and think really hard about what you’re feeling.
@eniko In school I once used a ruler to open the lock on a window in the second floor because the teachers did not allow ventilation of those rooms in the summer.. not sure if that counts as "picking" as I'm unsure wether that counts as "securing" something if I can open it with a ruler :D
@eniko does finding the combination of a lock by feeling count? I have done this with lock around the house, including my father's briefcase.
@eniko
I can pick one in the wild with my leather man screwdriver. It's an old one on family property but it is locking something up so it still counts.
Hi #fediverse. We need to talk about something.
While talking to a colleague about lockpicking it came up that they have never picked a lock. Like, not even once in their childhood.
Another colleague listening in admitted they also have never picked a lock.
My hypothesis is that most people have at one point in their life picked a lock.
Have you picked a lock?
Please boost for scientific accuracy.
@eniko I suppose using an angle grinder doesn't count as 'picking'? Then no, I haven't picked a lock.
@eniko I can't remember any instances of picking a wild lock, but I did cut (with a small file) a new front door key (5 lever lock) from memory and a bit of trial and error.
@eniko I'm voting practice locks, although some of that practice was on a locked door I had the key for
@eniko Yes, and I (we) succeeded.
My nephew had somehow locked an old door, so we took two pieces of scrap metal and fixed them inside, and after about 45 minutes of trying and even practicing on a sibling lock, we managed to turn it twice and it was unlocked. We even tried (while the door was ajar) to open it again, and learned more about the mechanism and succeeded in that too.
I might have done something similar earlier in life, but I don't remember.
@eniko We've touch decoded a code lock we thought was holding something but turned out it hadn't ever been set up and just had attachment brackets inside. That feels like an edge case on both halves of your question.
—🌔
I'm the lockpicker of the building when neighbors lost their keys.
First they seem a bit scared. Now they'te grateful
@eniko Where does "my lock inside my home, picked because it was easier than locating the key" fall?
@eniko I did to one of my wardrobes. Using two hairpins. it was cool to do it for the first time.
you're all delinquents!
i mean so am i but that doesn't get you off the hook
@eniko I wish I was that cool, but I did it with the validation of the owner of the lock and of the thing it was securing (they had lost their key).
@eniko I've opened locks without proper picking, the usual super basic ones that use flat edge things to turn a barrel. So uh... not sure how *that* counts
@eniko next to my work is abandoned facility. Nothing interesting at all, also i worked there many years ago, so just "hold my beer" moment. Those locks are so beaten that i used just random keys to unlock.
@eniko i learned lockpicking for the sole purpose of not paying actual money for the laundry machines that were present in my apartment at the time lol
@eniko does taking an iron bar and just forcing it until it breaks count as picking a lock?
If not i think it should
@eniko A combination lock that I forgot the code for which wasn't securing anything but it wasn't for practice.
@eniko I grew up in a small town with an abandoned military base. Once my friends and I found a locked filing cabinet so naturally we picked it, hoping to find some exciting classified documents or whatnot. Nothing so cool though, I don't even remember what was inside other than disappointment. So yeah that's the one time I picked a lock in the wild.
@eniko Have you ever encountered door combination locks using touch screens? They're really practical. You never have to remember the PIN as the fat stains of peoples fingers are there to guide you every time. 👍
@eniko I haven’t picked a lock. I have opened doors by sliding a plastic card in the lock, though.
My doors. I’d locked myself out.
@eniko Does slipping a latch count?
@eniko picked/bypassed a couple of friends lockers after they lost the keys and picked my own garage gate lock because my key had gotten damaged somehow and wouldn't work. It can be a very useful skill to have even at a lower level like me.
@eniko It hardly counts, as I was a teen at the time and I didn’t have any skills, but I popped an automobile’s door lock with a slim knife blade. Inserted blade into lock and turned. I figure I was just lucky or the lock wasn’t pinned
@eniko
When I was a kid I used to have this metal cash box in which I kept my valuables, like candy, cool rocks and sticks I had found, plastic miniature soldiers etc.
As it happens I lost the key so I had to pick the lock. Did it in a couple minutes with a butter knife.
@eniko I picked my front door lock of my old house back when I was a kid. I had to go there for whatever reason I don't remember now and I forgot the keys, so I looked around and made a pick from random things I found around... and somehow it worked.
@eniko I lost the key to the padlock for the little in building storage unit for my apartment, and I taught myself to pick a spool pin on my practice lock so I could open it.
@eniko
How would you count locks being used as game mechanics in a live action game?
Real picking? Just a practice lock.
Disassembly, removing some pins, and then putting it back for an easy pick later, that was decades ago
@eniko I have never picked a lock, but at about 10yo I "unlocked" the door to my parents' bedroom where my younger brother had locked himself in by pounding on it in a strategic way to pop the lock.
I also rescued my child from having accidentally locked herself in my bathroom by removing the doorknob entirely.
@eniko Not picking the lock, but I used to shim classroom doors in HS if the teacher was taking too long to get to class. I always just "found the door unlocked". 🤷♂️
@eniko i am looking for a decent set of picks though, available in Europe, not too expensive… Got tips?
| stock iphone: | 18 |
| stock android phone: | 10 |
| grapheneos: | 6 |
| some other kind of phone: | 3 |
| i don't have a phone: | 1 |
| see results: | 1 |
Closed
@hi lineageOS with microg
Tried iodeOS, e/OS/ etc. But they add too much
Never tried graphene, but then I don't buy the phones it's targeted towards so a non starter for me
If you have a long term partner that you live with, do you usually sleep in the same bed?
| yes: | 319 |
| no: | 61 |
| other: | 22 |
| results: | 51 |
Closed
@eniko we do sleep in the same bed, but we bought a king size one a while back and I couldn't go back down size.
Also I understand that it is not uncommon to sleep in separate bed, room or both. I know 2 close friends couple where it is the case (for various reasons)
@eniko we shared a bed for about a week after moving in together and then switched to separate bedrooms and stayed that way for over a decade
@eniko My wife loves me more than my snoring is unbearable.
Same is for my father which sleeps light, and mother that snores a lot.
@eniko yes, and once you have kids cats the number of beings in the same bed tends to go up, especially during winter
@eniko Separate because my wife goes to bed and wakes up much earlier than me, and if I wake her up in the middle of sleeping, she has a very tough time getting back to sleep.
@eniko wife and i like to have lots of space...so we got two large double beds. also, she still claims i snore...
@eniko Had one before where we didn't because chronic pain and insomnia with separate sleep "schedules" made it difficult for either of us to get more than a few hours of sleep. As far as I can tell, it didn't effect the relationship.
@eniko Same bedframe, different mattresses and sheets
(they require extra firm concrete and I desire clouds)
@eniko my husband and I are both extremely light sleepers and wake each other up constantly, so we usually sleep separate.
@eniko same bed but two different mattresses (200*80 cm each) and sperate blankets.
Every time we're traveling and have a bed with one mattress and one blanket we are grateful to be home again 😅
@eniko I don't really differentiate between friend or partner, but with a very close friend of mine we sometimes sleep in the same bed, but most days I want to sleep alone.
@eniko I mean, yes but it's one of them German beds these days where it's separate slats mattresses and duvets even though it's the one frame. It's a bit weird, though no more fighting over the covers, I guess?
#openbsd
Date: Sat, 21 Mar 2026 11:59:02 -0600
From: Theo de Raadt <deraadt@openbsd.org>
To: Renaud Allard <renaud@allard.it>
cc: tech@openbsd.org
Subject: Re: [patch] ext4fs rw
In-reply-to: <2c8df0cc-938e-4036-a628-4c7f69874e0a@allard.it>Renaud Allard <renaud@allard.it> wrote:
> Maybe it should be made clear on the website that OpenBSD will only
> allow new code made by a human. Because I feel there might be more
> requests like this and there is no point in repeating the discussion.Yes.
alas, sometimes it's about doing things slowly, eventually correctly, and with insults and hard work 😑
(as felt from the perspective of one who recently reported a bug¹, got insulted with a "why tf would you do something stupid like that"-like response, only to have someone else follow up a few days later having confirmed the bug and providing a patch²)
o_O
you cant hide app text labels in the default launcher
the notifications screen is combined with the control center
some ui elements dont fit properly in their containers (e.g. truncated text like "dont distu..."), and spacing can feel off
there is unnecessary reflow, opening a simple app like the calculator i can actually see the ui rendering piece by piece
the interface sometimes feels like its built from mismatched lego blocks that dont always fit together cleanly
adaptive brightness reacts too quickly, with noticeable abrupt changes
scrolling feels less natural (though i will probably get used to it in a few days)
built in apps lack consistency, there are too many different implementations of the same ui elements, like context menus
i get that the android ecosystem gives developers more freedom, and that is exactly why im using grapheneos, but the trade off is a less cohesive, consistent ui
Scoresub for #picotron is out!
Getting it to work with my MAKETEN clone "TenSum" was surprisingly easy.
I wasn't able to claim the top spot on the highscore board yet, but I'll keep trying.
Can't wait to play around with scoresub some more to create all kinds of online-ish projects. Asynchronous multiplayer games, highscore hunts etc.
in terms of tidiness, how ready is your home generally to receive unexpected visitors?
| it's always ready: | 70 |
| i can get it presentable in a couple hours: | 277 |
| not ready at all: | 228 |
| other (comment) & results: | 18 |
Closed
@eniko I have to vote "other" because my house is so spectacularly bad that I will not allow anyone in under any circumstances.
Look, I have extenuating circumstances... 
@eniko Always ready. The trick is not to feel obligated to tidy up for other people. When we tidy up it's for ourselves :)
I've been happier when it's always fairly ready. not "photoshoot for a magazine" ready but "everything is clean enough that I wouldn't mind a boss or first date seeing it without notice". Turns out I don't want to look at that stuff either.
@eniko Unexpected visitors? Y'mean invaders?
Gimme a mo', I should be able to get something ready >:(
@eniko it's between "not ready at all" and "presentable in a couple of hours", depending on how the chaotic phases of my partner are going
- I'm too ill to keep up with cleaning
- I know zero people that give a shit about not spreading covid, so I have been alone 24/7 for 6y, no visitors
@eniko between first and second. There always couple things doing whatever they want; can be fixed in 15 mins.
Also i really do not afraid ("do not care" not really fit, yet it almost it 😉 ) no visitor.
@eniko me: It's always ready
My wife: Argh, we need a day to clean!!
I try to convince her to clean *after* guests leave, but it never works.
@eniko I definitely vary between always ready and can get it ready, but at this point in my life it is probably closer to 30 min to get ready not a few hours.
It's take a LOT of work to get to this point.
@eniko can get it presentable in a couple hours if they don't need to visit my cav.. i mean, office.
@eniko it's a shitty chaotic mess, but I don't really give a crap so if someone wanted to visit that's fine... so I voted that it is always ready.
@eniko its generally acceptable, but i usually spent 15min collecting dirty laundry / dishes and emptying bins beforehand
@eniko it's not at all ready, but I decided a long time ago that anyone visiting can just deal with it. I've got too much going on to make my house a pretend level of neatness
@eniko "Always", but I voted "other". If someone comes to my home, they should know that a child and at least one person with executive dysfunction lives here and they should adjust their expectations accordingly.
I can always push the clutter aside and brew some tea. 🍵
A big part of the reason we like to host friends weekly is that it forces us to make sure our home is in reasonable shape. Like there are massive amounts of toys on the floor but they are in one contained area
@eniko I'd like to say I keep it tidy for my own sake, but truth is, other people coming over is a nice kick in the butt to do what I should do anyway.
@eniko This is a standard, long running joke in our house after our son once memorably remarked 'You only ever hoover when someone's coming round'.
@eniko It's always ready because if people visit unexpectedly, why should I pander to their needs?
If you need to sleep, well, did you bring a mat and sleeping bag?
And if you don't like the state of cleanliness, or dresses I can wear one more time over piled chairs or things like that, well, there are quite a few hotels in the neighbourhood.
I might clear some tablets e-readers and computers from the sofa, though, so you can sit.
@eniko Oh, I misread. You ask "In terms of tidiness"... so then, I don't care. But in general terms, I'm just not ready.
@eniko It probably needs a good 4-6 hours of cleaning, but I can only manage at most 5 minutes a day, and not every day. And of course the mess accumulates over time.
@eniko I sometimes have to gently remind my wife that at the part where we say "Oh, we have company coming over, we really should tidy up the house", our house is already more presentable than literally anyone else we know makes their house after they *finish* cleaning
Could someone please tell the maintainer of https://openports.pl (#OpenBSD Ports search) that the cert has expired? 😥
I just realized something..
I come from making games with algols where queuing animations used this awkward kind of setup:
.then(() => {}).then(() => {})
Even back then it felt shitty to work that way but it was all I ever known, and over time I'd read about continuations, or stringing along an env variable, fussing with callbacks, it was just all bad, but you learn to live with it.
In catlangs, animation is absolutely seamless:
this then that
It is a valid way of sequencing animation functions because the atoms of a catlang can be concatenated. It's the ultimate scheme for doing gamedev animation in my opinion.
Anyhow, You wouldn't NAME a VARIABLE.
@neauoire a while ago I made a little toy language for HTML interfaces in a forth like, and it was surprisingly intuitive.
Totally! here's a little demo https://arcades.agency/ICBINF/?loadfile=https://arcades.agency/media/demo.icb
And here's the "docs" https://arcades.agency/icbinf.html
@neauoire A timeline editor with keyframes and curves or hang me now.
@neauoire any language with coroutines becomes very natural for this too. Eg our dialogue system has single scripts which wait for responses (no function boundaries) and this was extremely writer friendly.
I wonder about the similarities and differences vs catlangs
You can't concatenate them in most languages but you can just call them one after the other (or wrap them in a function which does that)
@maxc as long as they're not a special form, they'll feel something like in a catlang, if APL have continuations, I'm guess the tacit nature of the language will make them feel quite similar to catlangs; otherwise, it's context switching and noisy syntax just like in JS. In scheme, which I find quite beautiful, continuations are super slow and look like a stain.
@neauoire in lua it's cooperative multithreading basically. coroutine.yield and you can manage the calling yourself. Since it's just a function call it can be put wherever to build something that doesn't have to deal with the nuts and bolts when you're being creative.
So in Arco, dialogue:line("tizo", "yo") makes the character tizo say yo, and waits for player input before continuing the script. Extra nice that prompts can just return whatever was picked. It can take as many frames as it needs and the rest of the game can keep in trucking.
Not as elegant perhaps when you drill down inside I guess, but it's explicit, so you get to say when interruptions can happen, and once you're in a coroutine anything using that functionality is just a normal function call.
@neauoire can I see a snippet of code on how this works ? I’m curious !
@mario_afk Sure! For example, here's the event that triggers when the last card on play is picked up.
the code: https://paste.sr.ht/~rabbits/11a769a87150e38a3936f592bc340f37d563ace6
down-animation.start() plays the pulldown animation
hand.new() draws a new hand while the cards are offscreen
back-animation.start() brings them back
The important bit is that there is no special syntax for async code even tho the program moves between a button event, to a screen animation, and back.
in context: https://git.sr.ht/~rabbits/donsol-hd/tree/master/item/src/donsol.tal#L346
@neauoire it looks great! Thanks :)
@neauoire Some years ago, a friend of mine made a JavaScript based version of a language that had Orc's parallelization operations included. Your example reminds me of that a little.
Of course, the focus was different, but the UX nonetheless was better.
Imagine a block in which things get done in parallel, until the first one finishes. The others get cancelled. E.g.
firstOf {
Get(first_url);
Get(second_url);
Get(third_url);
}
The "then"...
@neauoire ... part is just what follows the block, and so is just as intuitive as if I/O was blocking.
My TL;DR is at any rate that there are a lot improvements to language UX that we need to experiment more with.
@jens Your friend would have liked Occam on the Transputer : )
Almost up... sun-3/280 booting but not mounting /home or allowing login yet... (Realized my server rack is of the exactly appropriate age. Ha ha.)
@davefischer The hell is that Ethernet port? 😁 I've used coax connectors, and RJ45, but never that ...
@duncan_bayne AUI. A transceiver plugs into that, which converts to 10BT (RJ45) or thinwire (coax), or fiber optics, etc. OR... a cable from that to a vampire tap which clamps onto "thickwire" ethernet. Early 80s. (Vampire taps are crazy. You DRILL into a big fat cable and clamp onto it. !!!)
@rostiger fantastic! Great work!
@rostiger Congrats on this!
Let's say I just wanted to create a series of image galleries, and maybe didn't care about a comment system, would okular still be a good choice?
Also, a great next step would be to host a web-based tutorial (if there is one, i didn't immediately find it).
@exquisitecorp Sure! ocular has no comment system per se, uploading to Mastodon (which is completely optional) serves that function.
You can have multiple galleries by creating multiple project directories where the script lives in each of them. You can't build different galleries with one command though, you would need to build each gallery separately and manually.
The setup and usage instructions currently live in the readme in the repo. Is there anything you are missing? Or do you think better communication on where to find the instructions would be enough?
@rostiger thanks, that's helpful info re multiple galleries.
In terms of install and running okular, WHERE do i put my photos? does okular resize them? do i need to name them anything in particular? how do i add captions?
i'd add the output of okular -h to your README. on looking at the script it seems like maybe you use okular -a and then add a single image at a time? i'm not certain.
i guess i was expecting i'd drop images in a folder, have a text file with captions or titles, and then run your script. i'm thinking that assumption is incorrect.
@rostiger basically, i had assumed it was a static site generator for image galleries, something like faircamp-ish. but it sounds like you individually add photos.
@exquisitecorp Ah, that's valuable feedback, thank you! I need to clarify the intention of ocular: it's more of a smallweb replacement for platforms like instagram or pixelfed. Unlike many gallery systems out there where you take a bunch of pictures and it creates a gallery, ocular is for sharing pictures one by one, creating a feed of images. Instead of uploading an image to a platform, you add the image locally, build the gallery and upload the whole thing. With each upload, the feeds from the webring are pulled and integrated in the gallery.
This is also why you don't put any images anywhere than were they already are on your computer. ocular will create a database with the absolute paths to the image, which can be anywhere on your computer. It creates differently sized previews and stores them inside its project directory, but the originals will stay untouched.
@rostiger that logo is real pretty :)
@rostiger awesome! Congratulations.
Does it publish every new image? Can I publish only some of them? If I edit an old image (is that a thing?), would the Mastodon post be edited as well?
@bouncepaw No to most of these. :) It's a fairly simple system that only ever posts the last added image to Mastodon. If you delete or edit it and want the changes reflected on masto, you will have to do that manually.
@rostiger it looks very nice. i've been enjoying following the RSS feed of the entire webring. i was thinking of setting up something similar myself, but i think the digital journal i made myself is filling the niche for me for now...
@palomakop Thanks, glad to hear it speaks to you. Totally vaild to run your own setup, too! If you are interested in participating in the webring, you can still do that, even with your own structure, as long as the rss feed accommodates the webring conventions:
@rostiger yeah, maybe i should make a pics-only version of the rss feed for this ... since i can have more than one image per journal entry, but i have the images saved separately in the db. something to think about :)
@rostiger i had to go into my journal code to fix images from my phone coming out upside-down... so i'm going to take a stab at the new images rss feed ^-^
@palomakop Cool! You could set it up that when you add an image to a post you could give it a special tag that your generator parses and builds it into the rss feed along with the link to the post. This way, when someone clicks the image in the webring it would lead directly to your post.
@rostiger well, i was thinking of adding a text link to the post in my cdata, hmm, so many ways to do it haha... once i have something working, could i ask you to double check and maybe give feedback on my implementation of the webring patterns before i join?
@palomakop Sure, I'd be happy to! I don't think it's necessary to put it in the CDATA, as the minimum requirements for the webring will probably be lowered to this:
https://forum.merveilles.town/thread/114/subversivepics-aggregating-image-galleries-29/
@rostiger ok, here's my new feed: https://journal.palomakop.tv/images.xml just wanna double check that i read the info correctly and my feed is OK before i make the pull request to join! btw i also made a webpage version https://journal.palomakop.tv/images :)
@palomakop Nice! This looks spot on, well done!
@rostiger this is really great. i have been looking for a way out of corporate image hosting and sharing. i read through the repo docu, it all seems doable. i have a webserver, just need to check how to access it without “their” online toolset.
one big question though: depending on where i am, i use different machines (linux laptop, osx macbook and win desktop)
Do you think this works with posting to the gallery, or are there local dependencies that need to be synced?
@tovabele Nice! Happy to hear you'd like to give it a shot.
I have only tested it on Ubuntu, where all of the dependencies are already pre-installed. Theoretically it should be possible to use on a Mac, since afaik MacOS also comes with commands such as mkdir, cp, rm, rsync and curl preinstalled. I have never tried it, but I'd be curious if it works for sure!
As for windows, I have no idea what would be needed to get it up an running, but I'm pretty confident that it won't run out of the box.
Anyway, if you do try it, let me know how it goes or if you run into any issues. :3
@rostiger @tovabele I'm trying to get it going on macOS and have found a couple of stumbling blocks, things like using xdg-open in the script and an issue with the BSD version of realpath not supporting the -m option.
I'm currently working to get it usable and then I'll submit a PR for the changes if that is ok?
@rostiger Right now I'm getting this when I try to add an image so I'll be digging into it to see what is going on.
I can potentially solve the realpath issue with grealpath which is installed via 'brew install coreutils', you can add the gnubin folder to the PATH to force your computer to use grealpath for realpath but that might break something if the user is relying on the macOS bundled version somewhere else.
@drisc I'm considering to do away with realpath altogether. I changed the code recently so original images are copied to src/img/ instead of storing the path in the database because it would cause issues if the original file was moved or renamed.
The Lua code currently takes care of copying images, right after adding everything to the database. I could copy the image earlier, in the bash script. The only downside is if the script were canceled prematurely, the image would still be copied to the src dir.
@drisc Ok, I found another solution that should be compatible with macOS. Instead of realpath, you can use
absPath=$(cd "$WD_PATH/$(dirname "$1")" && pwd)/$(basename "$1")
This changes the working directory to the directory of the image, prints the working directory path and concatenates the filename resulting in the absolute path of the file.
Can you test if this works on your machine?
@rostiger Yep that worked for me, I was able to add a title and other metadata, though I am getting this from the conversion process (it still works but all 3 emit the warning):
WARNING: The convert command is deprecated in IMv7, use "magick" instead of "convert" or "magick convert"
I assume the default for Ubuntu LTS is an earlier version.
@drisc Ah, yes, I'm on an older version of imagemagick. I will write a check for this. Besides the warnings, is ocular working for you now?
@rostiger It's working now, the changes I had to make were adding "local keyword =" before kw on line 533 of main.lua to resolve the const error that cropped in in my screenshot and adding a new function to the bash script to use the correct "open" command on each platform, I've included it below:
function openFile() {
case $(uname) in
Linux) xdg-open "$1" ;;
Darwin) open "$1" ;;
esac
}
Then replace xdg-open with openFile.
The only other thing I noticed is that the style and theme css, the favicon stuff and the profile/banner image are supposed to be in a folder called media but the build just puts them in straight into dst.
@drisc Are you referring to the buildKeywordIndex function? I've got a couple of local changes, so line 533 is something else for me, hah
The issue with the media stuff is that I think cp -r isn't supported on macs. I'll fix this with the next push.
@drisc Ok, I pushed all the most recent fixes, in the bash script as well as the Lua scrip. Could you pull them and check if things work?
@rostiger Yep that version is working for me though the media folder in the src folder doesn't appear to be getting copied properly. It's making the media directory but the contents are still ending up in the root of dst. Give me a few and I'll try and puzzle out what's going on.
The imagemagick error is also still there, likely because the command still works and returns a version instead of erroring out.
Currently doing some simple complex bash scripting on OpenBSD because I can’t find the courage/patience to (re)start learning C from the beginning. The learning curve is so much tinier when it comes to integrating with shell scripts.
That reminds me of the time I was scripting things with dialog 😆
@ojs yes! :)
I generally use ksh and try to keep POSIX compliant. But in this case, I have to export functions. And that doesn't seem to be possible using ksh(1); from what I saw in the manual page and a bunch of try&fail. I could probably use ksh93 but bash is already installed by some other software so I just use it.
That said, when the bash scripts will work, I will probably try again to have all working in ksh.
@mirabilos well, right know, looks like it does. At least, the export -f on_click works as expected :)
@mirabilos I have 0 understanding of all what that means 😅
@mirabilos hum… ok… still I don’t do or expose BASH over HTTP 😅
@mirabilos I don’t do visual effects for movies! Kidding.
I don’t expose bash or ksh on the web :) My stuff is script for local desktop stuff - cf the wmaker screenshot posted yesterday.
Also took the opportunity to update my #tmux statusbar. Because, why not.
And yes, this one is running ksh(1) you weirdos>>> ;-)
@joel Nice color scheme. I might have to nick that. Very ham radio.
@pesco haha. This is called afterglow.
@joel oh you mean it has more than those two colors? ;)
@pesco yeah, I am a color fan :)))
How would I differenciate INSERT/REPLACE/SEARCH mode in vim if there were not 🫢
@nathanael why not just pkg_add?
Hi.
About verify yt-dlp, why dont you use correctly the option `-C`
?
as instance:
```
$ sha512 -C SHA2-512SUMS yt-dlp
(SHA512) yt-dlp: OK
```
Another tip:
```
for file in yt-dlp SHA2-512SUMS; do https://github.com/yt-dlp/yt-dlp/releases/latest/download/"${file}"; done
```
;) :p
Pretty much done implementing gameplay mechanics, now onto my favourite bit, interface animation 😈 Just adding a soft cursor lerp movement makes it all feel already smoother.
Just wondering, do you prefer reading #Fediverse …
| Boosts without further comments.: | 12 |
| Quotes with a comment.: | 9 |
| Posts with content and the URL.: | 9 |
Closed
@joel depends on the content:
@joel You forgot to add "cat pictures" as a fourth choice.
@joel a reply to the original toot containing one's thoughts if applicable plus a boost to bring the original toot into the timeline of a follower for ease of visibility.
I may be an edge case
@paul that makes me wonder if boost and quote are visible the same way from the original sender POV. I’m thinking of the case where someone would like to publicly emphasis someone’s problematic stance without having to interact with them - kinda name&shame.
@sotolf I assumed we were talking of post quotes. And it works on GtS but… maybe we should "define works". Attached are screenshot of the same post quote done with iOS Ivory. The first is seen from Ivory using an spare Mastodon account. The second is the GtS timeline as seen on the GtS web interface. Given the first use case, I consider GtS Post Quote to be available and working. 🤷♂️
@d6 yeeeee! i wonder if this will run on my remarkable >:3
@nilix if you try it out let me know how it goes! i've only tested it on this one machine so far -- in theory it should auto-detect the input devices/events but it's possible i'll have to add more code
@d6 cool! Is it a module you compile into the kernel?
@mccd no, the program runs in userspace. basically you log into one of the virtual consoles and run `uxn-lfb <your-rom>`
@d6 (Forgive a Linux newbie) So this will run without X like SDL relies on? I think small LCDs from e.g. Sparkfun can show up as framebuffera, right? And the display on MNT computers?
@gustav right, you don't need X, Wayland, SDL, or anything else.
i'm not sure about sparkfun but a raspberry pi, mnt reform, and most other linux systems with attached displays should provide framebuffers.
An a real alternative (not a front-end): Peertube.
Full disclaimer: I am the founder of Exquisite.tube.
If you do want a front-end: selfhost Invidious, Piped or Cloudtube. Might want to refrain from sharing it with too much people, as Google severely ratelimits it. It'll become a cat and mouse game before you know it (been there, done that, gave up).
but feels like if i'm looking for mainstream content (e.g. movie trailers) there isn't no second best. youtube is the only place
@hi I honestly found nothing :(
So I added some YT synchronisation (based on yt-dlp) to my Peertube instance just to watch things from there. But there are a few glitches so I’ll probably go back to deploying Invidious somewhere.
Hi #fediverse. We need to talk about something.
While talking to a colleague about how I recently learned most people have never sat on a cow it came up that she has never sat on a horse. Like, not even once during childhood.
Another colleague admitted they also have never sat on a horse.
My hypothesis is that most people have at one point in their life sat on a horse.
🏇 🐎 🐴
Have you sat on a horse?
Please boost for scientific accuracy.
| Yes: | 8004 |
| No: | 2295 |
experimenting with a new enemy in City of None ...
(haven't done the creature art yet, not sure what it's gonna look like)
More #DecemberAdventure stuff, making a little UF2/UF3 font parser so that I can use the fonts from @neauoire and @rek in my Oni programs.
@hi I also wish I had written down everything, the best I have done is some few random notes about something that doesnt make any sense today. Since I have been tinkering with Linux/Solaris/BSD since around 97 it would be fun to be able to look back at that but there is nothing to look back at, just because I cant get myself to take notes 😂
When you wake up what do you do
| get out of bed in 5-10 mins: | 211 |
| fuck around on my phone for half an hour: | 148 |
| fuck around on my phone for an hour (or more): | 99 |
| something else: | 70 |
Closed
@eniko if I did start playing with my phone I'd likely wake up my wife, so it's better to just get up.
Is there a way in #OpenBSD to stop charging battery from the OS ?
I know there is "hw.battery.chargestop" but it doesn't work on my laptop.
TIL that, if you have a USB soundbar connected to your #OpenBSD laptop, it is better to use sndiod -f rsnd/0 -F rsnd/1 than sndiod -f rsnd/1 (to force USB only). The first one works nice. The second seems to bring high latency on mpv and weird sound rendering on vlc.
Been working on a small decentralised music search tool https://squirrel.band/, initially indexing sites using faircamp (by @freebliss).
Currently grabs the sites listed on https://simonrepp.com/faircamp/ and in the webring (https://faircamp.webr.ing/) using each pages available RSS feeds, keen to add more sources if anyone has any suggestions.
Default administrative users on a system should be named
| root: | 15 |
| Administrator: | 5 |
| baron: | 9 |
| supervisor: | 2 |
Closed
I’m attempting to use Zig on OpenBSD-current. I have to perform seemingly every action, from viewing zig build —help to actually building a project, as root. This hasn’t been a problem with make, the uxn assembler, etc. What do I need to do to make this easier?
(I am using zig-0.15.2 from the package repo; maybe I should be using the latest snapshot? I saw there were some OpenBSD-related changes made not that long ago but I figured the version available from ports would either work out of the box or give me instructions for making it work.)
After around 23 years, I think I finally made a decision to switch from #Linux to #OpenBSD. It gives me this fantastic feeling of discovering new horizons and seems much more coherent. Also has no #techbros behind it.
I might even remove Linux from main SSD and leave it only on my USB disk to play one old Windows game that I have.
It seems like the only thing I will miss is battery life on Linux. But I'm ready for this tradeoff.
now running the coolest operating system on the coolest hardware - I've got no idea what I'm doing 😅
more importantly - which stickers should I get? thinking maybe a custom "oem sticker" (is there a proper word for them?) and have any of you done any crazy mods to your eee pc (or other laptops) that you'd care to share? I hollowed the dead battery case so could fit something fun there maybe.
#openbsd #bsd #eeePC #eee #retrocomputing #diyprojects #diy #pcmodding
OK, let's try the experiment two years later.
I've created a Fediverse account with a non-ASCII username.
@你好@i18n.viii.fi
Please try to follow it and then answer the poll questions.
If you can click & follow, which app and service are you using?
THANKS GANG!
| I can click on the username.: | 11 |
| I cannot click on the username.: | 407 |
| I can follow the account.: | 6 |
| I cannot follow the account.: | 150 |
I've fully moved my website to @OpenBSDAms at last! Very excited to make this transition to #OpenBSD and httpd(8) for web hosting.
https://ratfactor.com/openbsd/blog-13-moving-to-openbsd-dot-amsterdam
Hey there, some personal news: I've decided to move on and am therefore looking for a new job starting July.
I loved serving as a program manager with open source projects involving some of the most brilliant software developers out there.
In a future job, I'd like to continue working near the intersection of open source, funding, business, policy and the societal impact of digital technology. But that's a small niche and so I hope that this post might open some doors through my contacts here. Are there OSS foundations or (non-) government orgs looking for staff?
Being a software developer at heart, I have been excited about tech all my life and later found myself involved in various software projects and business ventures big and small. I speak both fluent Nerd and fluent Business and frequently helped translating between the two during my career.
I'm aware that my requirements are a strong filter: Part-time and remote, German timezone. I'm also fairly unenthusiastic about the current generative AI / LLM hype and would prefer to work outside of that bubble.
If you have possible leads, I'd be grateful to hear about it. Thanks in advance and thanks for sharing.
Hello 👋 I'm looking for some software engineering work at the moment. I have extensive experience in frontend, typescript, react.js. Contract would be ideal. I'm based in Belgium.
Boosts welcome, thanks everyone :)
🎉 picoCAD 2 is out! 🥳
Model, texture, and animate - in a single lo-fi, easy to use package.
GIF, GLTF, OBJ, and sprite sheet export.
Available now on Steam and Itch!
Check it out! http://picocad.net
Thanks for boosting! ❤️ 🙏
I wonder what it would be like to design a phone keyboard specifically to be used with vim
If the computer is a prosthetic (and it is) then this means we legitimately have to be careful about what software we run, because we are deciding what to make part of our selves, part of our minds
@mcc I sometimes find dreaming to be frustrating, because my computer prosthetic doesn't function very well in dreams. I can imagine that it responds to my commands, but those responses don't meet the standard of quality I expect from my computer. In dreams, my computer is just as forgetful as I am, and I often find myself going in circles trying to accomplish tasks that would have been trivial with an actual functioning computer.
@mcc I had this half formed thought about what pisses me off about AI yesterday, which your point frames better.
I just don’t want computers to work like this. I don’t want weird black boxes, that are almost certainly highly unreliable, that are plumbed together on the fly by some “agent”.
I hate that, and I think it’s shit.
I believe computers can be highly reliable, robust, transparent, and understandable.
I desperately want them to be like that, especially as an extension of myself.
The AI approach seems to be entirely conceding that aim and embrace “it doesn’t matter how it works if it usually gets the job done”.
I don’t want that.
TIL that the trunk(4) #OpenBSD manual page has an EXAMPLES section that describes how to "(...) set up roaming between wired and wireless networks using two network devices (...)". Which means you don't have to access the OpenBSD FAQ to do it. And that's fscking great when you brought your laptop to the and forgot to configure the automatic switching!
:open youtube.com and let me know if webkit crashes.Edit
I think because I'm on current and vimb is version 3.7.1 but needs to have the following dependencies.
@nina_kali_nina also you are porting to OpenBSD. Their C library does not put up with a program doing ill advised bullshit and it will throw aborts sooner than others. I love it as a development platform and anyone working in C or C++ should make sure their garbage works on OpenBSD. It will surface bugs you won't find easily on other platforms. Better: it fails faster so it's easier to debug problems that are subtle and difficult to find elsewhere.
@hi I have a WIP patch that prevents crashes - https://github.com/LibreSprite/LibreSprite/issues/542#issuecomment-4063449431 - now I need to figure out how to test it, and possibly make a PR...
Going over some of the finer points of #picotron accounts. podnet:// folders can only be listed by the owner, and so filenav reacts to that state change after switching accounts. podnet and scoresub() are arriving Monday-ish, along with PUT and POST fetch() requests. Get your 16x16 user icons ready!
Nothing like a Sunday morning to move all your stuff to #GotHub #OpenBSD
https://x61.sh/log/2026/03/14032026191148-gothub.html

OpenBSD adventures, day two. My MacBook with arm64 is running fine under OpenBSD, but there is no video acceleration, so it can't play full-screen videos.
I started to think what I can do about it, and I realised that we had a few e-waste Chromebooks bought for $20 apiece. It's 1.5GHz Celeron, and it is as dodgy as laptops get: it is spray-painted, it is made of cheap plastic, and the keyboard and the touchpad are both kind of only look like real ThinkPad but there were so many corners cut making it that I can't type "root" without it missing a letter or two every other time. This is what kids apparently were using in schools ten years ago or so?
Some things are glacially slow, but Xfce4 is quite usable, and it can play YouTube in 720p. Everything the laptop has to offer seems to be working (even webcam).
It works incredible for an ultra-low-end device from 2013.
LibreSprite on OpenBSD works*, wow**, and it was easy*** to make it run
*until it crashes; gotta figure out what it's unhappy about
** it's an old version
*** one compile time error resulting in a one-line diff, plus an hour of time