Now in the Decentralised Communication devroom: @cassidy and Anne presenting how Roost helps orgs mutualise their Trust & Safety efforts to keep people safe
Kann es sein, dass eine jĆ¼ngere Generation Autoren sich noch nie wirklich mit den negativen Auswirkungen von DRM beschƤftigt hat und den Behauptungen von Verƶffentlichungsplattformen einfach unbesehen Glauben schenkt?
Unwissend der Tatsache, dass diese ebensowenig das Beste der Autoren oder Creators im Sinne haben, wie das Beste der Konsumenten?
TurtleIsland.social, an Indigenousārun server, has a bigger one, yet itās invisible on Join Mastodon. It meets the covenant and supports safe Indigenous spaces. Help amplify this so it can be found. Boosts appreciated!
Der Digital Independence Day ist eine tolle Initiative, deren Werte wir zu 100% teilen. Die Empfehlung fĆ¼r Signal ist allerdings ein Schritt in die falsche Richtung und treibt Nutzer von einem geschlossenen System in ein anderes.
Zum GlĆ¼ck gibt es auch Wechselrezepte fĆ¼r die Independent Messaging Apps #Quicksy, #Monal und #Conversations_im, die wirkliche SouverƤnitƤt bieten.
Ein Wechsel zu Quicksy ist genauso einfach wie ein Wechsel zu Signal.
Du hast recht, es gibt an Signal das eine oder andere auszusetzen. Aber JEDER Messenger ist besser als WA. FĆ¼r Hans und Erna von der StraĆe hat Signal den Charme, dass + man es unkompliziert einrichten und betreiben kann, + dank Nutzernamen die eigene Nummer geschĆ¼tzt ist, + inzwischen eine betrƤchtliche Nutzerbasis existiert.
Somit ist Signal ein sehr gute und vor allem REALISTISCHE Alternative zu WA. Bevor ich mit der Forderung nach 100% die Menschen verschrecke und dann gar nichts erreiche, gebe ich mich lieber mit 90% zufrieden und erreiche Wandel in die richtige Richtung: weg von WA und dem anderen Sch... von META.
@PC_Fluesterer Quicksy ist genauso einfach einzurichten.
Betreiben kann man Signal allerdings nicht. Der source code ist zwar theoretisch offen aber ich kenne niemanden der einen Signal Server betreibt auĆer Signal. Geht alles durch die USA und durch Amazon Infrastruktur.
Deshalb ist Independent Messaging so wichtig. Da kann man die Server dort betreiben wo man es mƶchte. Sogar im eigenen Keller auf einem Raspberry Pie.
naia, ein Aspekt beim Chat ist der Netzwerk-Effekt. Sogar ich als selbst ernannter Fachmann š hƶre heute zum ersten mal von Quicksy. Damit werde ich kaum WA-Nutzys hinter dem Ofen hervor locken. Mit "betreiben" meinte ich "benutzen". FĆ¼r die breite Masse ist es von Vorteil, NICHT zum Selfhosting gezwungen zu sein. Auch deshalb haben Signal und Threema so groĆen Erfolg.
Signal betreibt nicht mal eigene Server, sondern nutzt AWS. Aber das ist egal, da Signal mit Zero Knowledge arbeitet. Der Client verschlĆ¼sselt, und der Rest ist wurscht. Und falls Amazon beschlieĆen sollte, Signal nicht mehr zu bedienen, kann Signal auch auf jeden anderen Server umziehen.
@PC_Fluesterer @kuketzblog @raanhacht Ob man es Big Tech nennt oder nicht ist mir egal. Was mich an Signal stƶrt ist derene offene Verachtung gegenĆ¼ber der klassischen freien Software Bewegung und Signals Weigerung zum Beispiel F-Droid oder UnifiedPush zu unterstĆ¼tzen. Der Chefin ~$1m im Jahr zu zahlen hat fĆ¼r eine angebliche non profit Organisation auch ein GschmƤckle.
Wir brauchen unbedingt 1000 millionen weitere #soooooovielbessere #digitale #Alternaiven ohne echte Menschen, damit garantiert die meisten Menschen bei #META bleiben, weil dort die meisten Menschen zu erreichen sind! #Dummheit
@DasGehtGleichWieder ich finde es auch Schwachsinn, dass alle zwei Tage jemand einen neuen Messenger erfindet. Deshalb empfehle ich die Lƶsung die 27 Jahre alt ist, sich aber stetig weiter entwickelt und die einzige Lƶsung die ein #IETF Standard ist.
@PC_Fluesterer @kuketzblog Wenn Du mit nicht unterstĆ¼tzen meinst, Signal ist offiziell nicht bei F-droid verfĆ¼gbar, dann hast Du damit natĆ¼rlich Recht. Ebenso mit Unified push. Die Argumentation dazu von Signal selbst, ist definitiv hinterfragenswert. Zumindest gibt es aber Mƶglichkeiten, den Play Store und Googles Notifications zu umgehen. 1/2
Womit wir beim PhƤnomen der Zersplitterung wƤren, dem Quell des regelmƤĆigen Misserfolgs alternativer Bewegungen. Zu jeder Alternative gibt es drei bessere Alternativen, und zu diesen noch-bessere Alternativen, und die ganzen Alternativen nutzt dann jeweils nur ein HƤufchen eingefleischter Fans.
Ein soziales Netz lebt nicht davon, dass auch noch das letzte SchrƤubchen perfekt justiert ist. Sondern davon, dass mƶglichst viele sich beteiligen.
@PC_Fluesterer @kuketzblog @raanhacht SelbstverstƤndlich ist WhatsApp noch viel schlimmer. Aber das macht ja Signal nicht automatisch zu einer guten Alternative.
Die financial statements von Signal sind ƶffentlich einsehbar.
Since 2013, the IRS has released data culled from over 1.8 million nonprofit tax filings. Use this database to find organizations and see details like their executive compensation, revenue and expenses, as well as download tax filings going back as fā¦
@PC_Fluesterer @kuketzblog @raanhacht Was meinst Du mit āWeigerung, F-Droid zu unterstĆ¼tzenā, denn Signal gibts doch Ć¼ber #fdroid via Guardian Project?
Dear developers, as we all care about privacy I would recommend to publish TextSecure on F-Droid (with having updates) as some people don't use Google Market. edit: The discussion on F-Droid can be...
To understand how we can replace Google push notifications (FCM) with something open source and decentralized, we need to understand how they work and why they are needed in the first place. This talk explains the mechanics of push notifications and why, despite their potentially bad reputation, they are a more elegant solution than having every app maintain its own persistent server connection.
While open-source tools like microG can remove proprietary Google software from your Android phone, the actual notifications are still sent via Google's servers (Firebase Cloud Messaging).
UnifiedPush is a framework that allows push notifications to be delivered in a decentralized manner or through self-hosted servers. Numerous open-source Android apps already support UnifiedPush, including Tusky, Ltt.rs, Fedilab, DAVxāµ, Fennec, Element, and many more.
The presentation ends with a short demo on how to use UnifiedPush on Android.
I am getting ready for the second day of #FOSDEM! The first day was a blast! It was great to meet old friends and new people. I am loving the new location of the #XMPP & Realtime Lounge, that is now in the AW building. If you are around today, come say hi!
correctie, ik was vergeten dat ik naar de main track k building ga voor het Fear and Loathing in the App Stores: when FLOSS principles collide with the Gatekeeper interests
#MilionChvilek svolĆ”vĆ” demoÅ”ku. Kdybyste se nĆ”hodou v nedÄli v 15:00 nudili... pro mimopražskĆ½ je to asi na prd a narychlo, ale tÅeba by nÄkdo vĆ½letil...
just can't believe I extended the frame struct. Something I was so so afraid for weeks to do. Feeling accomplished. Great way to end the night for me. V5 DSP, future improvements to the extended frame struct are now a breeze, OMG
I think about opening a PR to add a section that suggests people to bring an oxygen bottle to get fresh air on those packed rooms that no one is able to open a window. But @autkin has air quality monitors this year
it's releasable for sure. But I don't wanna draft another release right after we made the last one. So, for the rest of you, these sliders will have to wait. For those of you willing to compile from Master, they are there for you now. Speed quotient (voice gender): slider 100 Creakiness (laryngealization): slider 0 Breathiness: slider 0 Jitter (pitch variation): slider 0 Shimmer (amplitude variation): slider 0
wow, we now have more sliders, and finally Speed quotient (voice gender): slider 0. that's right. you can now change voice gender breathyness, something we lacked and now have. PR has been merged into Master, bumping DSP to version 5 and Voicing tone to version 3.
Have you ever tried to parse an 80 gigabyte JSON file?
That's a thing my team had to figure out when I, regrettably, worked for an American health insurance payor. This was as a direct result of malicious compliance on the part of other insurance companies, in reaction to attempts to use market-oriented regulation to improve healthcare costs in the American system.
Anyway, I can't recommend trying to handle 80 GB of JSON.
I also can't recommend market-oriented regulation of healthcare costs.
I recently had to parse a few terabytes (sharded into multi-gb gzipped shards). Multiple datasets, each in a different order. On a machine which had less disk space than the whole dataset. We only needed to extract a subset, but the fact every dataset was in a different order complicated things. Not fun.
We should explore adopting or integrating the MidCamp Live Captioning project as part of FOSDEMās accessibility efforts. https://github.com/MidCamp/live-captioning?tab=readme-ov-file#live-captionin...
āThe reason why RAM has become four times more expensive is that a huge amount of RAM that will be produced was promised to be paid with yet to exist money to be installed in GPUs that also have not yet been made, to place them in data centers that have not yet been built, powered by infrastructure that may never appear, to satisfy AI demand that does not actually exist to obtain mathematically impossible profit.ā
25% of the population has a disability. 100% of the population has temporary or situational barriers which needlessly stop them from accessing information on the web. The FOSDEM community isn't doi...
A silly finding with NVDA Composer by @FreakyFwoof is that the more notes you enter, the slower it gets to process each one. I have no idea if this is deliberate, and wouldn't be surprised if it is, as I'm sure on some toys from days gone by this is a real thing as it keeps having to move more and more notes around to make room in memory for new notes. When playing real music this probably hardly matters, but in this example I'm deliberately keysmashing while in QWERTY and chromatic mode so what I'm composing is pure nonsense. Anyway I ended up entering 542 notes, and as you can see, at first it's really fast but then it gets slower and slower, and if I'd kept doing it, it'd get even slower. Note that I was banging on the keyboard fast during the whole recording. Unfortunately I do not have a project file or exported MIDI as I cleared the whole thing out immediately.
Did you update so the export dialog works properly now? You can export to mid and two wav sounding things. Sine wav, and Nokia wav (which isn't really that) but sounds closer to what the composer window sounds like.
Okay. Seriously. I know this is beyond what this addon is supposed to do. But I would love, love, love a program that would take a MIDI and render all the notes, not just monophonic notes, as that type of square wave. Bonus points if I could change the ADSR parameters of that square wave. Also, did you figure out how to create that type of square wave? I know it's not *really* a square wave in the scientific sense, but it doesn't have that annoying raspy/scratchy sound which as I understand it comes from aliasing, since a true square wave has harmonics going to infinity, whereas a wav file has to stop somewhere.
Oh yes, I can import mid and export wav. But as per the design of this addon, it makes the MIDI monophonic. What I'd love is a program that does basically the same thing, rendering the MIDI as that type of square wave, but with all notes left intact, rather than squished down to one note at a time.
Oh I see I see. I'm not sure how it would distinguish export from playback, might get a bit complicated. Sounds like we need to spin off the import/export into it's own thing somehow.
Have you tried importing .rng yet? There are a few in the demo folder reached with CTRL+Shift+O, and of course at least a hundred in phonetones project.
It's fun that you can share projects with people just via pasting the clipboard as text. It gets rendered back into music at the other end. If you copy/paste the text below, you'll get a little Waltz thing I started.
Imagine, you could start something, like a little musical version of 'Once upon a time, there was...' Send that to people and they add a little bit more. 'A house in a deep, dark well. Why the house was in the well, nobody knows...' They send it back to you and you add, etc, etc. All via clipboard so they don't have a cchance to hear it first. Musical chainletter.
Another thing I think it'd be cool to be able to generate from MIDI files are monotonic audio files, only one note is playing at a time, but where the real MIDI has multiple notes going, it rapidly switches between them. There was a DOS program from the mid to late 80's called Pianoman which actually did that, you could compose supposedly multi-voice music, but since almost no one had any kind of sound card at that time, all the program had to work with was the PC speaker which, unless complicated programming was used, could only play one note at a time. So it rapidly switched between the different notes, which had its own unique sound. I posted a few pieces from Pianoman sometime last Summer.
Ironically, one strange use of this addon is to make a MIDI monophonic, which, for example, makes Rush E sound entirely different. To do this, import a MIDI file, then immediately export to another MIDI file.
I couldn't do what you wanted with this addon but maybe if I get it to build the import/export into a stand-alone app, it would be possible. Did just add some new functionality though, including reverse. Redownload when you get a chance. I'm quite happy with the new features.
creaky voice phonemically is used in languages like Danish (stĆød), Vietnamese (some tones), Jalapa Mazatec, certain Zapotec languages... We need it. If we are going to be multilingual, we can't go without creaky voice, no matter how obscure it is.
Here's everything coming to Perspective Intelligence in version 1.4
- Standard On-device AI - IOS Now, everyone can experience Perspective Intelligence even if you do not have an Apple Intelligence capable device. - Web search for All Access subscribers - Perspective Intelligence can now search the web to get more up-to-date information - RSS Feeds - You can now add RSS feeds to Perspective Intelligence to get the latest news. (1/2)
I've notice that the HTML & CSS for https://fosdem.org/2025/ is a little out of date. Wrong doctype Use of obsolete elements like Metadata not compliant with modern standards Repeated use ...
I am going to break the frame struct for the Danish support, I think. We will add a new extended frame struct or things like Shimmer as well. This opens possibilities. Thank you Denmark. Thank you for your strange and amazing language.
@Bri No, I install rustup from the official source and then just rustup update pretty often. That way I can install stable and nightly easily and stay up-to-date. It all manages itself.
Wow! I apparently am going to need a larger internal hard drive. Giggles. fedra Size: 4.9 GB, Attributes: ----D-------, Modified: 1/31/2026 11:33:15 PM
Democrat Taylor Rehmet is the victor in a special election for the Texas state Senate, flipping a reliably Republican district President Trump won by 17 points in 2024. texastribune.org/2026/01/30/teā¦
With ballots tallied from all but a handful of voting centers, Rehmet had 57% of the vote, besting the 43% for his GOP opponent, Leigh Wambsganss, who had a massive spending advantage.
Hey @FastSM First of all, thank you for the fabulous app. So lightweight and so functional! Amazing! Is it OK to submit feature requests here or should I go to Github?
All right. It would be great to see the status of the post on the timeline, e.g. was it boosted, quoted, a private reply, etc. Otherwise, all posts look like they are āpublicā. What do you think?
NV Speech Player 2.01 Release Notes This release focuses on stability, correctness, and cross-language improvements. The primary goal was to address accumulated issues in the synthesis pipeline and bring several under-maintained language packs up to a proper standard. Synthesis Engine Fixes Phoneme Tuning: M (Bilabial Nasal): Adjusted formant values based on acoustic phonetics research. The antiformant (cfN0) is now set to 950 Hz to properly cancel F2 energy, eliminating the overly "stuffy" quality in words like "same" and "memory". P (Voiceless Bilabial Stop): Reduced frication amplitude and adjusted formant values for a flatter, more diffuse spectrum. The F2 locus is now at 850 Hz, appropriate for bilabial consonants. This addresses the excessive "puffiness" reported in words like "quote" and "paper". K (Voiceless Velar Stop): Reduced frication amplitude and parallel formant values to soften the release burst. Language Pack Improvements: Danish: Resolved a significant intelligibility issue. eSpeak-ng outputs glottal stops for Danish stod, which were causing vowels to be interrupted and producing speech that was difficult to understand. Danish stod is currently stripped entirely rather than rendered as creaky voice. A proper implementation would require synthesizer-level support for laryngealization. We are exploring this for the future. Spanish is improved! github.com/tgeczy/NVSpeechPlayā¦ github.com/tgeczy/NVSpeechPlayā¦ github.com/tgeczy/NVSpeechPlayā¦
Spanish really improbed from the last version I checked. L, y, M and N consonants have smoother transitions to vowels. I also apreciate the different G phonemes as in gato or agua. It seems that phonemes related with R hare hard to replicate.
@muchanchoasado yep, R still needs work. words like "pero" and "madre" sort of work now, but I think it'll get some more tuning in the future. I'm just glad Spanish is now understandable, it was a big mess before. Happy to make it better.
Briš„°
in reply to JamminJerry • • •Briš„°
Unknown parent • • •Briš„°
Unknown parent • • •Briš„°
Unknown parent • • •Tj
in reply to Briš„° • • •