Adrian Cochrane @alcinnz@floss.social

@docskrzyk Yeah, the telemetry stuff in Ubuntu has been a problem for awhile. Is is a dependency of ubuntu-desktop, too? Because that's a super backwards idea if so.

I gave up on mainline Ubuntu awhile ago (Unity!), switching to Xubuntu for a smaller footprint, and now to SparkyLinux. I use Chromium for browsing here and there (not very often), and when they made its .deb into basically a container for a snap, I was done with them (not a snap fan).

Wow, there's already lots of #plan9 content on gemini gus. #gemini

Electron and npm and all that bullshit is starting to feel like a distant memory..

The code for making a GUI application with keyboard controls in #plan9 basically fits in a toot.

@grainloom @tindall Yeah, which is why "newbies" are most likely to be older people rather than kids.

But still, the people doing the teaching ought to know better... but compatibility issues and marketing make it hard to stray from the common path.

The hardest problem in computer science is apparently writing a chat application that can connect to different servers at once and also not fuck up notifications. So far discord and slack fail at this.

It's interesting to think that jurrasic park was made on a computer that probably isn't powerful enough to watch jurrasic park on youtube

@alcinnz
My reading is that Harfbuzz determines which symbols to use (glyphs) and calculates where they will be placed. This doesn't sound impressive, but we're talking about natural language and centuries of typographic practices

I think turning vectors into dots and populating a buffer with that data is a separate process? I can't be sure from what I've read so far

(Thought you might like to know that ppl are attempting to engage with the material)

I guess the other question raised by this discussion is: what kind of software do people want to see, delivered either on mobile or over the web, that doesn't currently exist/isn't free?

@Nora, Tech Witch there's little better for simple recording and waveform editing than audacity, period.

also, ffmpeg is used in a lot of professional video projects and gets rolled out in custom broadcast setups.

making anything good better takes paid support – people ned to have the time to make real contributions (as is the case with webkit, audacity, krita, ffmpeg, apache, nginx, the linux kernel, openbsd and so on). it's open and anyone can have a copy for free, but making and building it and doing the work of reaching out to human beings to determine its future and to promote it, all costs money that's totally possible to have. gnu imp had a fork that was heavily used in vfx for a few years, cinepaint – professionally supported and promoted, until something a bit tighter and better came along.

also, ggpo is running in commercial games and that's open source – what's stopping it from being the messenger for any office suite? aside from people with the time to mash 'em together.

@tindall see also: Autodesk going around to various animation studios and telling them that if they use blender they'll lose the rights to their own work, conflating GPL with copyright

Popcorn Computer's Pocket P.C. has been opensourced!

https://blog.popcorncomputer.com/2020/08/05/pocket-p-c-open-sourced/

#opensource #openhardware

New Blog Post: Gentoo Isn't for Me

https://www.paritybit.ca/blog/gentoo-isnt-for-me

"I've been using Gentoo on and off since I tried it out, but I've come to the conclusion that it's just not for me. These are the reasons why."

#100DaysToOffload

camera | filter | strip metadata | compress | post on social media and each of those components can be swapped out by the user

branching pipelines that upload to multiple social media at once

i promise i'm not going to write another fucking novel but just. please imagine graphical unix philosophy programs for mobile phones

Continuing on from yesterday, how does Hafbuzz position glyphs onscreen? And how does this impact the final substitutions?

To compute glyph positions for OpenType fonts, Harfbuzz first clears out the old data.

An initial approximations (which is enough for English) involves iterating over each character summing the widths as computed by the font. Or heights for vertical text. Afterwhich it computes the width (or height) of space characters specially.

1/?