Skip to main content



This little mechanical calculator is one of my favorite objects. It came to me with what turned out to be a fouled 10s pin retaining ring; it was a bear to take apart, ‘machine’ a new part, and reassemble. But it works! And I love it!

#CURTA #calculators #MechanicalCalculators #ScientificInstruments



WordPress’s emperor, Matt Mullenweg, demands a hefty tribute from WP Engine, and a battle erupts leaving millions of websites hanging in the balance. Meanwhile, the Internet Archive is under siege from hackers.

Listen to "Smashing Security" in your favourite podcast app:

🟢 Spotify: open.spotify.com/show/3J7pBxEu…

🟣 Apple Podcasts: podcasts.apple.com/gb/podcast/…

Everywhere else: smashingsecurity.com



Takéto citáty nezvyknem postovať, ale tento je naozaj vydarený. Úspech tkvie v dobrej príprave - dlhodobej systematickej práci na tom, aby sme to druhej strane zbytočne neuľahčovali a tiež aby sme, keď už zaznie alarm vedeli dostatočne rýchlo (aj adekvátne) reagovať.

From: @SunTzuCyber
infosec.exchange/@SunTzuCyber/…


"Every cyber battle is won before it is fought" - Sun Tzu, The Art of Cyber War



Buenos días desde la Administración Pública.

Vuelta a la normalidad con lunes de lluvia.



For those who can’t wait for Emoji Reactions I made #Conversations_im available for free on Google Play for the next couple of days.

play.google.com/store/apps/det…

(Note that the Google Play version has some drawbacks like no address book integration and no public channel discovery. For channel discovery you can use search.jabber.network directly.)

in reply to Daniel Gultsch

I can wait, and I appreciate your dedication to the project.
in reply to Daniel Gultsch

Oh nice, really impatient to test it. Side note, do you consider publishing the app on Accrescent? accrescent.app/


“You are using an ad blocker.”

Yes, and the fact that you know I am is why I am using it in the first place.



I love that Metatext and Semaphore are living new and enhanced lives as Feditext and Enafore, respectively. Unfortunately, both of these new projects have long standing, severe problems impacting accessibility, so I'm still using the old, unmaintained projects as my daily drivers. One of these days, I might get around to looking into these bugs myself, but until then, this is how it is.
This entry was edited (1 week ago)
in reply to victor tsaran

@vick21 I actually didn't realise there was a limited free version of Mona to try and I wasn't willing to pay for something I might later decide I didn't love. Might have to try it at some point. That doesn't change things on desktop, of course, where I don't use Mac.
in reply to Jamie Teh

I think you will love Mona. It’s the best thing, IMHO, we have on the iPhone.


I see Microsoft still haven't fixed the crash in Phone Link when you focus the message list. So, here's my app module to work around this and add some other conveniences. I can't be bothered adding type annotations, etc. to submit to NVDA core, nor can I be bothered submitting this as an add-on, so this will require developer scratchpad to be enabled. gist.githubusercontent.com/jcs…

Tamas G reshared this.

in reply to Patrick Perdue

@BorrisInABox Oh oh, yes, I know what you're talking about now. I have calls disabled for Phone Link because it wants to steal calls from my Airpods, but I re-enable calls on demand occasionally if I need it for a specific call, which isn't that often for me. Just in case you don't know, alt+s gets you to settings, which is useful for quickly disabling calls.
in reply to Jamie Teh

Yep. Problem for me is that calls work more reliably than sending messages. And sometimes, new messages won't show up, but I'll get all the pointless notifications like there's no tomorrow. I can turn those off, but then I'd probably get nothing but phone calls. If I turn those off, I'd probably get nothing.
in reply to Patrick Perdue

@BorrisInABox Yeah, it's definitely far from perfect. For me, it's also a hell of a lot better than nothing. And I blame Apple for all of this crappiness, for all the difference that makes to no one.
in reply to Patrick Perdue

@BorrisInABox Anecdotally, Phone Link seems to be more reliable with messages for me in the last month or two, perhaps iOS 18, a Phone Link update or... complete falsehood that I will discover as such just minutes after I write this. Do with that what you will, probably nothing. :)
in reply to Jamie Teh

Oh, cool. I've had pretty much the opposite experience here.
in reply to Patrick Perdue

@BorrisInABox Heh, and now that I've said this, my experience will plummit in 3... 2... 1...


When I want to make a bulleted list using markdown, I use…

  • * an asterisk (0 votes)
  • - a hyphen (0 votes)
Poll end: 1 week ago

in reply to Jamie Teh

@jcsteh It irks me that asterisks look like slightly-off-center bullets.
in reply to Blake Watson

Hahaha. I can't relate to that as a screen reader user. Though in Unified English Braille, an asterisk consumes 1 more cell than a dash. There's some useless trivia to support our preference.


Written by an 18 yrs. old:

“I wish I was around when people had blogs or even myspace. This era was deeply personal and creative. Most writing on the internet was individual, not written in search of “SEO” or profit but driven by the need and want of people to share knowledge–pure curiosity.”

Great blog post re: #IndieWeb #OldWeb

rohan.ga/blog/early-internet/

/via @feedle



Why is it considered socially and legally unacceptable for me to enter a crowded room armed with an aluminum baseball bat, striking at people randomly - a crippled leg here, a destroyed skull & brain there, a collapsed rib cage and punctured lungs...

Why can I not just injur or kill anyone I want with the bat?

I can do it with #covid.



Welcome to the RB family, Weathernaut 🥳

apt.izzysoft.de/packages/me.ta…

Weathernaut lets you check the current conditions, get hourly forecasts for today and tomorrow, or plan your week with a 7-day forecast.

Again, thanks to @bg443 for providing the recipe!

And we just passed the 300th RB app at IzzyOnDroid here: 301 apps / 24.8% are now RB :awesome:

#reproducibleBuilds #IzzyOnDroid



Welcome to the RB family, Jetflix 🥳

apt.izzysoft.de/packages/com.y…

Jetflix is a Movie listing app using the TheMovieDb API which lets you search, filter, select by genre, see cast and crew details, production companies and more.

Thanks to @bg443 for providing the recipe!

#reproducibleBuilds #IzzyOnDroid



I wonder why americans call it "bullet train". I mean they don't kill people and they are hardly found in america. And opposed by most republicans.


When considering the widely discussed ethical issues with generative AI (unrestrained energy use contributing to climate change, theft of creative work, perpetuation of biases, etc.), I wonder if there's a meaningful difference between current-generation models like GPT-4O that can generate image descriptions on top of everything else LLMs do, and earlier, more specialized vision models used in things like the Microsoft Office apps and Microsoft's Seeing AI app.
in reply to Matt Campbell

In other words, if I were to take a stance on not using the former kind of model for ethical reasons, then to be consistent, would I also have to refuse to use the latter (but earlier) kind of model?
in reply to Matt Campbell

The fact that the latter runs on my phone with decent performance, and the former does not, would suggest that there are differences. Whether or not the latter's initial training exhibits some of the same problems as generative models, I don't know, but it's one of the many reasons that just referencing generative models as "AI" is lazy and harmful.

For instance, I read an article yesterday about a publisher updating their copyright statements to include the text:

"No part of this book may be used or reproduced in any manner for the purpose of training artificial intelligence technologies or systems ..."

The lack of nuance immediately made me wonder if I would be violating those conditions if, through my legitimate and unintended actions, any part of the work ever came into contact with any machine learning technologies, e.g. those used for OCR.



If you want a job in the moisturizer industry, the best advice I can give is to apply daily.


Today a man knocked on my door and asked for a small donation towards the local swimming pool. I gave him a glass of water.


11 Best Free and Open Source Firewall Tools lxer.com/module/newswire/ext_l…


At a protest in SF, blind Uber and Lyft users claim the companies aren't doing enough to prevent drivers from turning away passengers with visual impairments (Wired)

wired.com/story/protestors-say…
techmeme.com/241020/p7#a241020…





good morning

Anon: “A computer makes it possible to do, in half an hour, tasks which were completely unnecessary to do before.”





I really like Drew's framework here dividing current AI use-cases into Gods (human replacement, which I think of as still mostly science fiction), Interns (assistants you delegate closely-reviewed tasks to, which is most of how I use LLMs today) and Cogs (smaller tools that can more reliably serve a single purpose, like Whisper for transcription) - more of my own notes on this here: simonwillison.net/2024/Oct/20/…
note.computer/@dbreunig/113330…
in reply to Simon Willison

You sure you linked to the right song? Have you gone through the kind of existential crisis portrayed in that song? I always saw you as being cautiously optimistic about AI, that it can be a useful tool if used well, but not making us redundant.

@forrestbrazeal

in reply to Matt Campbell

@matt @forrestbrazeal oh I absolutely went through these 5 stages of grief, but I've been firmly in the 5th step for over a year at this point
in reply to Simon Willison

@forrestbrazeal I don't think I ever actually felt the anger. And I feel kind of guilty about that, because of course some people have decided to stop there and fight against AI, and they have such moral certainty about it.
in reply to Matt Campbell

@matt @forrestbrazeal I've not felt the anger personally, because I've been releasing open source code for 20+ years so I already default to "I want people to be able to reuse my work as much as possible" - but I absolutely understand the people who ARE angry

If I was an artist and someone trained Stable Diffusion on my work without my permission and then started competing with me for commissions that used my own personal style I think I'd feel very differently about this all

in reply to Simon Willison

@matt @forrestbrazeal I don’t feel like I am stuck in step 4, but I can’t get to step 5 because I do very much believe that (at least in their current incarnation) AI is very much going away. I am worried about the collateral damage that this particular dead tree is going to cause when it smashes into the rest of the ecosystem. E.g.: nvidia currently has a market cap of 11% of GDP, which I do not think is sustainable. But who on earth is going to get a positive ROI on ChatGPT at $20/mo?
in reply to Glyph

@glyph @matt @forrestbrazeal I expect there's going to be a substantial AI crash, but I don't think (most of) the tools I'm using right now will become unavailable to me - especially since I can run Llama 70B on my own laptop now
in reply to Simon Willison

About running models locally, I've experimented with that, but large context windows take up lots of RAM, right? Like, isn't it O(n^2) where n is the number of tokens? Or do you not depend on large context windows?

@glyph

in reply to Matt Campbell

@matt while I agree that the models you are using locally will not go away, I am also skeptical that they will receive much in the way of maintenance once the commercial bubble has burst. Nobody is getting paid when you run a model on your laptop locally, and training them still requires a zillion hours of GPU time and a corpus in the petabytes
in reply to Glyph

@glyph Yeah. Today I went down a rabbit hole looking into what it would take to reproducibly, deterministically build some video assets for a little side project I did. It reminded me of how some hard-core free software folks, particularly the GNU Guix developers, have been working on bootstrapping a whole OS with minimal dependence on existing binaries. We're never getting that for these big models.

@simon

in reply to Matt Campbell

@matt @glyph the models I can run in my laptop today are leagues ahead of the models I ran on the exact same hardware a year ago - improvements in that space have been significant

This new trick from Microsoft looks like it could be a huge leap forward too - I've not dug into it properly yet though github.com/microsoft/BitNet




What a journey this is. Take a listen, preferably with headphones. youtu.be/WrxJKj71c9o?si=0OmZNa…
in reply to Jakob Rosin

Really cool. Thanks for sharing! I felt they overdid panning though! :)
in reply to victor tsaran

@vick21 I think the original was broadcast through multiple speakers in a conference hall, so thats wy panning happened so much.


On this day seven years ago I got A FRICKING GOLD MEDAL for my work on #curl. Real. Proper. Gold. From the hands of the Swedish king.

daniel.haxx.se/blog/2017/10/20…

#curl

Peter Vágner reshared this.

in reply to daniel:// stenberg://

can he hug or is he one of those guys who hit you in the back repeatedly just to let everyone know they are heterosexual?


Just yesterday, I saw a boost of someone suggesting "grand theft autocomplete" as an apt phrase for describing what large language models (such as ChatGPT and Claude) do. But now I can't remember who said it. The phrase doesn't yet seem to be common enough to yield a useful result when doing a DuckDuckGo search for it.
in reply to Matt Campbell

I'm pretty sure it originated from @johl here: mastodon.xyz/@johl/11205983391…


We compared smart ring heart rate data with smartwatches, and the results may surprise you (...Instead, it's the Amazfit Helio Ring that won the hardware accuracy contest. And since Amazfit recently discounted it to $199 and made its pricey subscriptions free, the Helio Ring has become more appealing.
On the other hand, the Helio Ring has unresolved problems like the limited number of sizes (three), short battery life (four days), and lack of automatic workout detection.) androidcentral.com/wearables/s…


Kámo na mňa niečo lezie... celý deň ma bolia predlaktia, pravý kľúčny kĺb na nohe a absolútne nič necítim, ani trávu ani voňavku a mám pocit že na mňa ide teplota. To nič, ráno som fit.

in reply to daniel:// stenberg://

looks like the CFP is not open yet.

I wonder if something #AlpineLinux or #k0s/#kubernetes would be of interest for this event



In 5 Tagen startet die 5. Staffel #StarTrek Lower Decks 🙌🥳

Hier zur Feier des Tages das Staffel 1-Poster. 🖖☺️



in reply to Shay

I ate the discord sound as part of my breakfast, sorry. If it sounds muffled or quieter, it's just cause it's coming from inside my stomach for everyone else now.
in reply to Shay

Yeah I’m not a fan. The incoming call one is fun though.
in reply to Shay

I checked in settings to see if they changed it lol
in reply to aaron

@fireborn I think that sound is only temporary for the season. Cuz if you remember they change it for Christmas, too
in reply to Shay

The sooner they get rid of the message sound the better haha.


Elgg: Set Up a Social Networking Platform on Ubuntu linuxtoday.com/blog/elgg-set-u…


#AndroidAppRain at apt.izzysoft.de/fdroid today brings you 17 updated and one long-awaited added apps:

* Fossify Camera: your light-weight camera app from the Fossify collection 🥳 and it got the 🛡️ (i.e. it is reproducible), yay!

Enjoy your #free #Android #apps and #reproducibleBuilds with the #IzzyOnDroid repo :awesome:



Bluesky: i love this game!!!! so i drew some fanart for it (kinda nsfw)

Mastodon: [Spoiler Warning/Opinion] I love this game! So I wrote an updated graphics driver to deliver better realtime lighting on levels 4 through 8

Threads: Our worst fears have come true. Noah left us. Yes, in the afternoon

reshared this




When I first played #stardewvalley I was overwhelmed by all the tasks you had to do. It felt like the opposite of a #cozygame is there something out there that has the same look and feel but doesn't have the huge list of todos you gotta do?
in reply to Erion

@erion I just remember feeling overwhelmed at the time I started playing it with not knowing what was the best way to go, and having a big list of quests and stuff. I guess for me I didn't take them as optional but wanted to play the game in some min max way. Perhaps I should revisit it as a sort of village sandbox.
in reply to Triptych 🌱

If you do, you will have a great time. When people experience the game for the first time, it's common to have 6-7 saves until you figure out what works for you. Just take it easy for the first few tries, and one day when you are familiar with everything, focus on maxing out things. You will have a great time.