I guess it's Xmas Eve. Please note, it's okay to feel other emotions at this time, besides utter glee and happiness. Xmas is difficult for most of us. It's never really been my happiest time of the year... and I'm okay with that. For me, what helps is talking to people on the phone and stuff.. Heck, I spent Xmas on my own one year and it was glorious! Another year, friends forced me to go to a local Chinese restaurant and man, the food was good. Then we went to a few pubs and I was amazed at how many people were out, just to hang out with other people. Anyway, just remember that if you are feeling alone, that you are not. Especially if you're on Mastodon.

reshared this

in reply to The Matrix.org Foundation

> Write a blog post, tell TWIM, tell the world, tell us what we can improve

Well, I can tell you what improvements I would like to see from a perspective of a person who has been trying to write some maubot plugins.

I'd welcome improvements to HTML rendering in messages in all clients, because right now literally every client has some issues in this area and this makes it a pain in the ass to create message layouts that look good. Among those problems are incorrect <table> element rendering, missing newlines after some tags, <details> tag not working at all, issues displaying <sup> or <sub>, weird behavior around inline <img> tags or not displaying them at all, and many more, in every client there are different ones. Oh, and it would be nice to get some control of column width in <table>, pretty please. It's not in the spec, but would be great to have it. Element Desktop has the most correct implementation of all this but it's also not without problems. Why is this important to me? Because I want to create messages like the one on the screenshots I attached (made in Cinny), something similar to Discord embeds. The subset of HTML at our disposal is already very limited, so it's even more annoying when it doesn't work the way it should.

Other than that, I would appreciate fixing spoiler support in Element. Right now it's half-assed on desktop, and nonexistent in Element X. Have you ever tried to discuss movies or books without properly working spoiler tags? I'd rather not discuss them at all.

in reply to Korba

@korba well, we can work with that. the original choice of HTML for rich text in Matrix was to give us the flexibility for pixel-perfect copy-paste in future, but then we fell down a rabbithole of dealing with Markdown dialects and haven't got back to figuring out how to do richer text while sanitising styling against attacks. Have you already filed a bug for this somewhere ooi?

Friend: I don't know what to do!

Me: (I explain what he needs to do.)

Friend: That sounds like just what I need. I'm going to do that!

(a week later)

Me: How did it go?

Friend: Oh that? I didn't do it.

Me: Why not?

Friend: I don't know. I just didn't.

Me: Okay. Will you try again?

Friend: Probably not.

Me: So you'll just ignore the problem then?

Friend: Oh no. It's still a problem.

Me: But I gave you the solution.

Friend: And a great solution it was.

Me: And yet...

Friend: Exactly

reshared this

Dear Amazon, if you're going to use AI enhanced synthetic speech for Audio Description on your Prime TV series, the least you can do is check it for pronunciation quality. The voice being realistic isn't enough.

In Talamasca: The Secret Order, AD mispronunciations include "Surré" instead of "Surrey", "alyus" instead of "alias", and "one hundreds" instead of "hundreds".

#Amazon #Prime #Talamasca #AD #accessibility #MustDoBetter

in reply to Austin Nix

Naturally when I mentioned The Life and Adventures of Santa Claus earlier it's because I was reading it. and now I am not because I've finished it. I've read enough of the Oz books to know that I probably would have guessed this was written by the same author even if I didn't know that up front. Baum has a very particular style of writing that you can always tell it's him when you read one of his books.
in reply to Austin Nix

This has nothing to do with Christmas, but I thought I'd mention it since you mention Baum. He wrote a ton of other stuff besides the Oz books, and it did sell quite a bit. I think the one which sold best, I'm not sure whether sales were better than the Oz books but am sure it didn't last as well, is his Aunt Jane series. It's an interesting glimpse of the time, I'm not particularly into that myself but did read the first one. Anyhow, it's on Gutenberg, I believe, as well as being in the usual Delphi collection which they call, with some justification given that it contains, at least, the vast majority of what he wrote, the complete Baum delphiclassics.com/shop/l-fran…

I love this recording of the Nat King Cole trio performing The Christmas song in 1946. The version you always hear with the strings is lovely and everything but this stripped back version with just bass, guitar and Nat's piano and vocals is something else.
youtube.com/watch?v=5QJNhCi3Ay…
#Music

Good grief I wish someone would create a way to do a mass migration from #lemmy to #piefed. Lemmy is by far the most annoying software I have the displeasure of maintaining. In today's adventures, the Lemmy developers finally created a #postgres index in 0.19.15 that I'd created myself ages ago for performance reasons. But the database migration fails if the index already exists. Even if it's the exact index it wants to create, created exactly the way the migration is doing it. And no, you can't skip a database migration, and no, it's not smart enough to check and skip the migration by itself. So you have to log into the database and drop index idx_post_aggregates_creator, and let Lemmy recreate it. Yes, this takes a while, for absolutely no reason. And if you wanted to do it without downtime in an offline copy of the database, I guess you'd have to do a bunch of fiddling to trick Lemmy into thinking the migration happened when you went to upgrade production. There had to be a better way to do that! But it's not documented and I have no idea what it would be.

André Polykanine reshared this.

“What makes me transgendered is that my birth sex—which is female—appears to be in social contradiction to my gender expression—which is read as masculine. I defend my right to that social contradiction. In fact, I want to live long enough to hear people ask, ‘What made me think that was a contradiction in the first place?’”

(Leslie Feinberg, "Trans Liberation")

#trans #lgbtq #lgbtqia #nonbinary

Super huge panic. Took left Cadence out of backpack, now cells across two sides of it won't raise or lower at all. Gosh. Did me dropping my bag on the car floor ruin the display? (another reason I wish cases existed for it.)
I ran the dot calibration tool, nothing- those dots, 4 cells on either left and right sides of the left display, appear dead to it. Wow, just wow. What a holiday.
Still no luck. Now only 4 cells actually try to pop up when I use it. Have a feeling something did get dislodged inside and now will need new one sent.
This entry was edited (19 hours ago)

We are grateful to everyone who has become a #FriendsOfGNOME. Your donations help us plan new programs for 2026 and strengthen the future of #GNOME. We also thank all our contributors across many roles, without whom our desktop would not be possible.

Happy Holidays!

#Linux #OpenSource #FOSS #GNOMEFoundation

donate.gnome.org/

Emmanuele Bassi reshared this.

I have to give @Tamasg for making the initial version of the FlexVoice 3 NVDA addon! I like it, and there is another version of that synth you can try and make an addon for, it is version 2. I believe according to a sample I heard, that I kind of like the intonation of version 2 more than version 3, and it contains other .trv files for the voices. Now you kind of know why I've been waiting for this TTS to speak my own text for so long (it has been about a year since Rommix found it), it is one of the most unique speech synths, and that's what I'm looking for! Anyway, here's the .zip of FlexVoice 2 for anyone who is interested:

datajake.braillescreen.net/tts…

in reply to Alex Krier

I guess the problem with that zip is that it's not a full SDK like with FlexVoice 3.0. That one was so easy because all the header files and calling conventions were there for me to ingest and quickly understand the order of arguments functions take. For 2.0, it's mostly a guessing game, so I don't suspect a driver for it would be ready for awhile unless some type of headers or deeper docs emerge around the API it used.
in reply to Tamas G

I will be patiently waiting for the FlexVoice 2 addon. FlexVoice is one of my dream retro synths I have been waiting for since circa 2019. It is using a hybrid synthesis method, combining diphone concatenation with parametric LPC. That's why it sounds like that. The voice I am using in the current addon is Kim, since I am biased towards female voices. I will be getting later addon builds when you let me (and others) know through this Mastodon account. Again, thank you so much for making one of my retro TTS dreams come true!

Today's #FreeSoftwareAdvent is paperless-ngx, a key part of keeping us, well, paperless.

It is a document management tool, but I use it in a very basic way: it is hooked up to our scanner, and anything we scan gets automatically converted to PDF and OCRd. We then shred the paper. I try to scan, and shred, everything on the day that it arrives.

It is particularly useful around tax return time, as it means I can easily get the information I need from stuff which people have posted to us.

github.com/paperless-ngx/paper…

#FOSS #SelfHosting #Paperless

Well well. It's very unlikely Monolog95 or Monolog97 will ever get a driver or wrapper. Thing is, voice font data and other speech DLLs were scattered across the system, including in \windows\speech, making integrating a simple portable wrapper solution quite impossible as it looks for them in these hardcoded areas, including registry keys. Both do this, after extensive debugging, Monolog95 is somewhat better but still Voicefont data stays in program files. In short, not an easy one, unlike Flexvoice where I could package it up nicely into one synth for everyone to quickly have. Sigh.
It doesn't matter anyway. The voice I'd want was only ever part of Monolog16, not Monolog 32-bit variants. With Monolog 95 and 97, you already get a different voice. So I wouldn't be working on one I'd actually want or use, either.
in reply to Tamas G

on the bright side, I'm less sad about my Braille display! Used some sandpaper to smoothen out the scratch (they were just 2 little dots where it got slightly nicked), and now, the scratch is virtually undetectable. So yes. Sandpaper does work on plastic to at least smoothen down scratches and get them to be good again. I was surprised to find this out.
This entry was edited (1 day ago)
in reply to Maikel 🇪🇺 🇪🇸

The F-Droid version has some features that Play store policy prevents us from putting in (address book integration, channel search). The Play version has features that F-Droid policy prevents us from putting in (FCM push notifications).

That’s orthogonal to it being paid or not though.

If it were up to me i'd probably make the F-Droid version paid as well but F-Droid doesn’t have a billing system.

Wrt my last boost, it is saying something when the NFB and the ACB agree on something and are on the same side!

But, seriously, hopefully there is less feuding between the two blindness organizations than there had been in the past. Hardly anyone alive would have been around for the original split that created the ACB, and some of the reasons for it are not relevant anymore. I would like it if we don't have hostility between the organizations and if they can work together where they agree.

This entry was edited (1 day ago)
in reply to Mike Gorse

For my part, I can say that my issues with the #NFB don’t have anything to do with what happened back in the late 50s. Rather, it’s something more recent. I don’t necessarily have animosity toward the organization in general, I was actually a member for a time about seven years ago, it has to do with the organization’s response, or lack thereof, dealing with the sexual misconduct allegations in the training centers.
#NFB

ACB is speaking out in response to recent federal action that limits how the Randolph-Sheppard Act’s blind vendor priority is applied to Army dining facility contracts. This change threatens economic opportunity and independence for people who are blind or have low vision.

Read our press release: acb.org/american-council-blind…

in reply to Marco Zehe

I agree Skeebs has its issues. Tapestry has evolved to be a decent alternative. It has some quirks too, but all in all, a better VO experience than Skeets. Not for nothing, but Mona as well as Spring when it still worked with Twitter offered the best accessible user experience. It would be great if Bluesky support came to Mona or even a repurpose Spring app. I’d have no issues paying for it either as the developer has shown time and time again that they make accessibility a priority. One can continue to hope…

I took a closer look at this new Beyerdynamic headset. The only question I still have is: is Bluetooth 6 comparable to old 2.4 GHz dongles in terms of latency? Has anyone had any experience with this? I'd be grateful for any insights.
europe.beyerdynamic.com/p/mmx-…
#bluetooth #wireless #headset #latency #tech
in reply to Felix Steindorff

Well, there isn't something like the old 2.4 GHZ dongles, since that just describes the frequency range the dongles communicate at, but not the protocols used. Usually those dongles try to minimize protocol overhead as much as possible, which Bluetooth traditionally doesn't really do. Also, those don't cap the bitrates as drastically and can use whichever codecs they like. So no, even BT6 doesn't do the one-digit latency values some headsets do with 2.4 GHZ dongles, yet.
in reply to Toni Barth

The real question however should be: do we need to care? Only a few people will be able to notice the difference between 5 ms and 15 ms of latency anyway, so BT6 is already a huge step forward and will most likely be just enough for most people. The few people left will most likely be high-class e-sports people and high precision musicians. Maybe blind people who are used to low-latency TTS too, time will tell.
in reply to Toni Barth

@ToniBarth Ok, good to know, but what is the difference in the protocols and codecs used between the two wireless standards? I don't imagine I can hear the difference between 5 and 15 ms, but I can tell the difference between 50 and 20. Hence the question of whether we are in the ultra-low range with both variants, or whether BT6 is still closer to 100 ms than 0.