No teď je tady "zima", jak jsem teď na ubytování uvnitř tak to není vůbec problém. Nejhorší byly akorát 2 dny kdy jsem měl ráno pod nulou, teď jsem ve Valencii a je tady cca 20° přes den takže takové naše pozdní jaro.
By popular demand, a single post to boost global switch day across multiple platforms on February 1st 2025.
For those who can no longer get behind Musk or Meta's new policies and feel like switching (e.g. from #Whatsapp to #Signal, from #Instagram to #Pixelfed or from #X to #Mastodon) I made you some graphics to use freely in communicating with your contacts.
A new version of the Toby Doom Launcher is available for Windows. This one hopefully fixes the problems with non-English characters. Also, a new custom game, "Doomguy Verses the Daleks," is available.
For linux-game-manager users, the launcher is kept up to date automatically. Have fun, rip and tear until it is done! 🤘
Thinking about a new Linux laptop to run Linux on? Going with Linux laptop vendors can be a good option as they offer seamless hardware-software integration, pre-installed Linux distros, and guaranteed driver compatibility. Here are available Linux laptop vendors 😎👇
Explore high-quality infographics on Linux, cybersecurity, and networking. Perfect for certifications, technical knowledge, and tech career upskilling.
#AddioTwitter Ho disattivato il mio account su X. Ho scritto e letto su Twitter dal 2009 al 2022, da 2 anni non postavo più e ogni tanto entravo per leggere con un certo senso di colpa. Alcuni account mi mancheranno e sicuramente mi spiacerà non leggere più i post di @orporick, ma adesso per me non è più possibile restare #Twitter #ElonMusk #SocialNetwork @maupao @scuola @informapirata
The pun in the title is that some people pronounce the a11y numeronym as “alley”. That makes the full title sound like uncanny valley, the concept of human-looking things seeming almost, but not quite, human and therefore creepy.
We need a new accessible IM app soon. Help! WHatsapp has been rather meta free so far, but its about to become a whatscrapp mstdn.social/@TechCrunch/11386…
Absolutely broken my mastodon experience here because my client doesn't support threaded posts, so I have to now open the original post on the web just to see the post that probably links to another post. This is the epitome of why I hate quote-posting. It's not twitter. Twitter's dead.
I’m seeing it used quite a lot lately. The “re:last” post situation is equally frustrating, as in some cases, the boost which was re lasted can be far away or filtered out.
The problem with last boost is that clients only show boosts once per x hours, so if somebody else boosted it recently, you have no chance of seeing it.
Quoting is a lot less annoying, because even if it isn't supported natively by your client (it isn't by mine), you can just click a link and you see what the post is about.
@miki Going to a website that I'd not have to go to, if the text was... Just you know, there already, I'm far less likely to read it. Equally, re last: If you don't see it, you're not going to read it either. Admittedly nobody wins, but assuming you do see it because it was properly reshared, you'll know whether you want to continue clicking further. It's not perfect.
@FreakyFwoof @miki I don't know which one I find more infuriating between these. In re last you have exactly the same problems you (Andre) outline in some other ramification here (start of another thread, no mention of original) with the added bonus that you sometimes have to go to the profile of the guy to know what the hell they're commenting on because there is no context at all. Scrap that, I do find re last more infuriating.
Yep, but quote-posting is unofficial, and I reiterate. Mastodon specifically is *not* twitter. Any client that supports it is not adhering to standards. There are forks that do support it, but Mastodon specifically doesn't. Talked about this time and time and time again, so won't, but seriously, bad. No proper attribution when replies come in, IE: original poster is not mentioned, starts a new thread etc, etc.
Here's another really cool recent discovery. The "phonetic punctuation" NVDA addon has been around for a while, and it used to do one thing, substitute various symbols (or more technically any regular expression you put in), with sounds. But, as I discovered accidentally while looking for something in the addon store, it was quietly updated in December, now renamed "Earcons and Speech Rules." And this new version is a huge game changer at least for me. If you ever used the JAWS speech and sounds manager to its fullest this addon is now basically an equivllent. You can give any control type or state a sound, and unlike something like unspoken, these sounds are played in sync with the speech and the addon has no problem dealing with something like multiple links on a line. It also allows you to have NVDA indicate various formatting attributes, like bold, italic and heading levels, with either pitch/speed/volume changes or sounds. The very latest version of this addon comes with built-in heading level sounds that I contributed, and in an upcoming version it will also support assigning spelling errors as a formatting attribute. github.com/mltony/nvda-phoneti…
Phonetic punctuation is an NVDA add-on that allows to convert punctuation signs into audio icons. It can also convert any regular expressions into audio icons. - mltony/nvda-phonetic-punctuation
This seems like it will be neat. I just have to figure out how to turn off the punctuation sounds. I didn't realize just how detailed things would be, so when the pitch went up for an all-caps word, then there were hand claps for quotes, it was all rather startling. I'll also have to see if I can find why I hear "H2" for headings in addition to the sounds.
@alexhall The addon adds its own category to NVDA's settings. All the punctuation rules are under the "text regular expressions", you can simply move down the list and hit Alt+T on each to turn it off. The same goes for the h2 thing, that's under the "numeric formatting" category. That's a leftover from when it didn't support sounds for that and apparently wasn't taken out of the default configuration
@masonasons @jscholes I guess it's because the actual punctuation is technically being stripped. I have all the regex rules off and the punctuation is preserved when speech history copies it.
Not sure how to fix it to be honest. Probably by waiting for the new extension points to land in a stable version of NVDA and updating SH to use those instead of monkey patching. @masonasons
@masonasons @jscholes There is also a way to temporarily turn all the rules off, which is NVDA+Alt+P by default which might be a workaround. And on the flipside I love how the speech history still includes all of the sounds and pitch changes which I didn't expect.
@x0 Yeah the author tried to get to get it fixed, sent a PR even but it wasn't aproved. Perhaps now that the old audio support is on its way out and more people heard of this addon we can get it fixed.
We still need to maintain compatibbility with 2024 though, 2025 is going to be a hard sell even more than usual for breaking every synth add-on ever in what seems like a backward incompatible manner. All the conditionals. It's going to be strange.
@KaraLG84 Ditto, I have 2 drivers and they both behave differently. IBM TTS works, while the latest version of Eloquence threshhold doesn't crash but also doesn't change the rate either.
bringing this thread up again because I just tried the earcons add-on again: Which version of the Eloquence DLLs are you using? I'd really like to get to the bottom of why this crashes NVDA, and it's only IBMTTS that it does it with.
@KaraLG84 I had to remove it. It absolutely broke in chrome, couldn't move in any direction on most pages at all. Very few of the shortcut keys did anything.
@KaraLG84 I've no idea. I don't have any other addons that mess with browsers other than Tyler's thing that speeds up aria events, but that shouldn't cause as many issues as this did.
@KaraLG84 Wait you actually got it to work with IBMTTS? But I thought the indexing in there made it unreliable? At least timing wise. I suppose changing rate and whatnot is just a speech sequence alteration which hasn't anything to do with indexing, but sound playback definitely is. Is it capable of playing sounds without interrupting speech? Cause if it tries to interrupt by inserting the sound in a pause I doubt that works even still. Tony even tried a PR but it behaved weirdly with spelling and introduced more latency to the driver as a whole. I need to see if WASAPI fixed the crackles and whatnot, because if it did I'm just going to strip the damn index buffering and bring back the x0_personal versions if I have to.
@x0 @KaraLG84 Timing is slightly off, yeah, but it does work. Though this addon was the reason I also installed Eloq threshhold even though there's things I don't like about it like how it's randomly saving my voice settings as British English for no reason.
@x0 @KaraLG84 Somewhat unrelated, how spossible would it be to put in a switch that is simliar to unspoken in that , yes I want you to play sounds and such, but I also want you to not alter what you normally would speak? I.E. play the heading sound, while at the same time not midifying how headings are normally spoken, I.E. keep the role announcements? There are practical situations in which having your cake and eating it too proves useful.
@x0 @KaraLG84 Not exactly, because it would still leave the role announcements out of the loop. Unspoken allowed you to have sounds while preserving traditional role announcements.
@pvagner There is a verbose mode you can turn on which I think does what you want. I don’t remember what the default key stroke is, but yeah, I think it’s possible.
@pvagner It doesn't fully disable role suppression, wonder how possible it would be to backport the method Unspoken used to do role sounds while keeping the speech queues. Maybe provide multiprong rule support?
Sounds very cool. For me one of the major advantages of Unspoken is suppressing the role names though, do you know if that's possible as well? It just makes moving around quicker.
It's really a cool addin, but I can't figure out how and what to turn off to make it behave the way I want it too. Right now it's just too verbose for me.
@TSchulte Go into NVDA's settings and look for the earcons and speech rules category, you can turn the sounds you want off from here or add new ones. I agree the default configuration is very noisy, I turned basically all of the punctuation sounds off
First type the curl command into your shell and press enter. What exactly happens then? Daniel takes you through a step-by-step from the command line tool, i...
From the Florida Outreach Center for the Blind: Zoom Workshop on A.I. Technology: Thursday, January 23, 2:00 PM Eastern Time groups.io/g/tech-vi/message/84…
Le chef par intérim du Parti libéral du Québec n’a pas l’intention de quitter le média social X, même si son propriétaire Elon Musk est membre de l’administration Trump, qui menace le Canada d’une guerre commerciale.
what is tought about in secondaire this side of the atlantic. Because were I come from we get taught how he came to power and etc. And what "collabo" and "traitors" are (Pétain, Vichy Govt). And I have been horrified to see it repeating...
Around this time last year, I did a rare thing for me, and wrote/produced a complete song, the first I had done in years. This stemmed from what I call a "syno sequence" that had been running around in my head rent-free since 1992, associated with the word organism.
Being a blind, autistic Synesthete, pretty much anything goes. Words can be associated with indescribable sounds, musical motifs, smells, other words, colors... Ideas and concepts can have that, too. It's very hard to explain this particular brand of synesthesia even to other Synesthetes.
Did you know, for example, that all saxophones are called Steve, but Steve itself, not any particular person, just the word Steve, is usually a pizza box? Well, now you do!
If I took all the filters off, I could write about this for days.
Anyway, in late 1992, during my third grade science class, the topic of the day was organisms, and their different classifications. As soon as my teacher said the word organism, this little four bar sequence in C miner popped into my head, and just never left.
Finally, almost 32 years later, I couldn't stand it anymore, and wrote and performed a song around that little motif. I played all the instruments on this piece, except for drums and bass, provided by @FreakyFwoof, who also helped with the mix, and tightening up my crappy rhythm guitar playing. I'm not great with stringed instruments. I'm much more comfortable with buttons and knobs.
This features acoustic guitar, melodica (actually a melodion, because Hammond-Suzuki calls it that, probably for Hohner trademark avoidance reasons), clarinet, mandolin, piano (Spitfire Labs Mrs. Mills), an emulated Casio VL-tone 1, a doorstop, a Hammond XK-2 pretending to be a B3, drums, bass, and me clapping and yelling random stuff. Except for the drums, bass, piano and organ, everything is acoustic, so about half of it, I guess.
I don't get to play music at home anymore for various depressing reasons. This was recorded in my parents' living room. I can't explain, after years of virtual musical silence, how much joy I got from making this silly thing happen. We could all use some of that right about now, so, in that spirit, I'm sharing this with the world again.
In late 1992, I was sitting in third grade science class. The topic that day was organisms.Just the word organism caused this little musical motif to pop int...
A new BT Speak update is scheduled for release this week. Join us for a live webinar this coming Wednesday, January 22 at 8:00 PM Eastern time to learn about the new features. This webinar will be hosted by Robert Carter, David Woodbridge and Allison Melloy, hosts of the Tech Doctors podcast. You’ll hear from Deane Blazie, as well as Bryan, Stephen and Chris Blazie, along with Steve Clower and David Goldfield. Learn about the update, meet the Blazie family and participate in a live Q&A about the BT Speak. This webinar will also be available as a podcast from the Tech Doctors. We’re looking forward to seeing you this Wednesday, January 22 at 8:00 PM Eastern Time. Here’s the Zoom link to join us. us02web.zoom.us/j/83910762102?…
Topic: The Tech Doctors talk with some of the people who make the magic at Blazie Technologies Time: January 22, 2025 8:00 PM Eastern Time (US and Canada) Join Zoom Meeting us02web.zoom.us/j/83910762102?…
Meeting ID: 839 1076 2102 Passcode: 579635
---
One tap mobile +13462487799,,83910762102#,,,,*579635# US (Houston) +12532050468,,83910762102#,,,,*579635# US
---
Dial by your location • +1 346 248 7799 US (Houston) • +1 253 205 0468 US • +1 253 215 8782 US (Tacoma) • +1 408 638 0968 US (San Jose) • +1 669 444 9171 US • +1 669 900 6833 US (San Jose) • +1 719 359 4580 US • +1 646 931 3860 US • +1 689 278 1000 US • +1 301 715 8592 US (Washington DC) • +1 305 224 1968 US • +1 309 205 3325 US • +1 312 626 6799 US (Chicago) • +1 360 209 5623 US • +1 386 347 5053 US • +1 507 473 4847 US • +1 564 217 2000 US • +1 646 876 9923 US (New York)
Zoom is the leader in modern enterprise video communications, with an easy, reliable cloud platform for video and audio conferencing, chat, and webinars across mobile, desktop, and room systems.
Mógłbym teraz rozpisywać się, jak to twórcy tworzą treści dla samej radości tworzenia, z pasji. Mógłbym wspomnieć, że liczby się nie liczą, bo tak naprawdę liczą się ludzie itp.
Tylko wiesz... To byłaby niezła ściema, ale social media już nie takie teksty przyjmowały 😉
Sam niegdyś pisałem teksty "do szuflady". To dla mnie dowód na to, że lubię pisać. Nawet nikt nie musiał tego widzieć/czytać. Sam fakt napisania np. opowiadania był dla mnie przyjemny.
Istnieją jednak twórcy (i jest ich naprawdę sporo), którzy mają swój cel w tworzeniu treści.
Jeśli widzisz, że np. szef znanej marki odzieżowej, czy dietetyk zakładają konto na Instagramie, czy TikToku, to w większości przypadków możesz mieć pewność, że jest to działanie "profesjonalne i policzalne".
Mówiąc prościej: celem na pewno nie są lajki i powszechne uznanie.
Co do sztucznej inteligencji, to wraz z Adamem Gospodarczykiem i Mateuszem Chrobokiem tworzymy szkolenie AI Devs, czyli jak najbardziej zarabiamy na tym.
Nadal piszę dla przyjemności i frajdy, ale też są z tego wymierne korzyści.
Gdy opublikuję np. opis moich eksperymentów z LLM-ami, to myślę, że dotrę do 10-20 tysięcy odbiorców. Zdjęcie kota z kromką chleba na głowie miałby minimum jedno zero więcej.
Nad tekstem będę siedział 1-2h. Kot to gotowiec z netu ze strony z memami.
Film o np. montowaniu okien trafi do 800 osób, z czego 2 zadzwonią zapytać o ofertę. Film z prankami zobaczy milion osób, a platforma zapłaci Ci za to 4-5k. To prawie tyle, co pół zlecenia na okna do domu.
@ISibboI we count that as a "Linux distribution", not a separate OS. I will admit that the distinction between them is not crystal clear and that shows in the slide...
I have noticed that the quickest way to get information on the internet is to post incomplete or incorrect information. Although in your case I can't imagine anyone considers 106 OS's incomplete.
PalmOS, OS-9, DEC TOPS-10/TENEX, DEC TOPS-20, MIT ITS, Interdata OS/32, Fujitsu-Siemens BS2000, CrayOS (before UNICOS), DEC VAXELN (for turning VAXen into real-time embedded devices).
Ah. Context. You might put that in the image, for easier grasping. The list is so long, it looks like it's trying to be a list of everything... that's an achievement for sure.
@ElPamplina @mdione I have to admit I've grown a little accustomed to learning about curl getting used in many places but still: confirmed used by NASA involved in landing a vehicle on another planet is certainly next-level. Certainly one of the coolest accomplishments so far I think.
Didn't that one probe to Jupiter run curl? IIRC, they slammed it into the core. So for a few microseconds, it was there. Also, what about the one asteroid a few months back?
#cuba #socialism #socialismo Al respecto de esta noticia, yo le iba a sugerir a nuestro Presidente Miguel Díaz-Canel que le dijera esto al gobierno yanqui: "Gobierno de los Estados Unidos de América: Métanse por el culo su lista de países." Pero no sé si sonaría un poco arrogante....
Donald Trump revocó este lunes, en las primeras horas de su Presidencia, la orden del pasado 14 de enero con la que el expresidente Joe Biden sacó a Cuba de la lista de estados promotores del terrorismo.
Hraju si teď s LLM DeepSeek R1 github.com/deepseek-ai/DeepSee… Vypadá to, že tento open-source model, který je dostupný zdarma (chat.deepseek.com a zaškrtnout Deepthink), dosahuje kvalit OpenAI-o1 modelu. Na lokálním stroji ale ten 32B nebo dokonce 70B model (zabírají cca 40 GB na disku) rozhodně nerozjedu, laptop se mně potí u 8B modelů. :) #DeepSeek #LLM
Me na tom trochu desi jejich strategie (a vladni, ktera to financuje). A trochu doufam, ze LLM jako takove narazili na technologicky strop a tohle pomuze demokratizaci podkladovych modelu k dalsim pouzitim a na cenove valce v zasade nevydela nikdo ani v Cine, ani v US.
Free office suite – the evolution of OpenOffice. Compatible with Microsoft .doc, .docx, .xls, .xlsx, .ppt, .pptx. Updated regularly, community powered.
Free office suite – the evolution of OpenOffice. Compatible with Microsoft .doc, .docx, .xls, .xlsx, .ppt, .pptx. Updated regularly, community powered.
I spent few hours hacking together a proof of concept authorization for @panoramax using an existing OAuth2 access token from OpenStreetMap. Basically, I'm already logged in, so why should I log the second time in a Panoramax instance?
Partly solved the issue that @MapComplete had, and circumvented by having a single user uploading everything, adding original authors to an Exif tag.
The idea was: to upload photos from a mobile app, I need to seamlessly login to Panoramax. But I already have an OpenStreetMap OAuth2 token, and I don't...
By Jock Elliott, KB2GOM There was quite a bit of excitement in the ultralight radio community in 2024 when folks “discovered” the Qodosen DX-286 radio. The very first version of this radio was the …
🚨 Exciting news! 🚨 The Linux App Summit 2025 will be held in Tirana, Albania from April 25-26! 🎉 This week, we’re opening the call for papers! 📝 Be sure to check our website for the latest updates and details. 🌐 linuxappsummit.org/
The Linux App Summit (LAS) is designed to accelerate the growth of the Linux application ecosystem by bringing together everyone involved in creating a great Linux application user experience.
A small team of researchers from the U.S. and the U.K. has resurrected the code for a 60-year-old chatbot named ELIZA, believed to be the first electronic chatbot.
Chao-c'
in reply to plactagonic • • •měli bychom s @cobic taky nahrát svoje videa ze #thesuntrip2024 někam na PeerTube 😉
Jinak počítám, že @BajkazylHK ti ty fotky a videa určitě naboostuje, i když jejich specializací je městská cyklistika a ne úplně bikepacking...
plactagonic
in reply to Chao-c' • • •No vzhledem k mé "vysoké" kvalitě videí a fotek tak nevím jestli se mám úplně čím chlubit.
Jen doufám že něco bude z té 360 kamery, to potom budou asi i zajímavější videa pro víc než těch 10 lidí co si pustí ty mé vlogy.
Chao-c'
in reply to plactagonic • • •kvalita je slušná, ale působíš strašně utahaně.
respekt, já takhle v zimě dám půl dne na běžkách nebo na kole a jsem na týden úplně vyřízenej...
@plactagonic@vhsky.cz @cobic @BajkazylHK
plactagonic
in reply to Chao-c' • • •archos
in reply to Chao-c' • • •plactagonic
in reply to archos • • •Jak už jsem psal ono to s tou zimou není vůbec tak špatné jak u nás. Jen ty výkyvy v horách...
Teď jsem v hostelu spal s otevřeným oknem a byla taková optimální teplota.
Archos
in reply to plactagonic • • •@archos@vhsky.cz @xChaos