Show newer

Debuted as a portrait model in a 3h live drawing session today - what an experience!! :) Like meditation, but the whole room stares at you and your head produces 10 thoughts per second. At the same time you get the weirdest bodily sensations and visual phenomena from the stillness and focus. Utterly enjoyed it. (⌒‿⌒)

Simon Repp boosted

I wrote an article about the need for low-carbon and #sustainable #computing and the path towards zero-carbon computing.

In short, we need to dramatically limit the growth in emissions from computing, or by 2040 emissions from computing alone will be close to half the emissions level acceptable to keep global warming below 2°C.

And it is possible to do this.

https://wimvanderbauwhede.github.io/articles/frugal-computing/

Anybody use one of those minimal card-sized phones? I'm intrigued by the idea of a ~30g phone that only does calls and texting but both the bulk of products on offer and the available means to order them seem like a surreal tech lottery ...

If you're into generative vector graphics I might have something cool for you: I finally found a workflow for script-based generation of vector primitives and text, including - and here it gets interesting - the ability to perform arbitrary scripted boolean operations between all objects (!), and the subsequent possibility to export all results as svg.

This might not look or sound like much, but to my surprise it was incredibly hard to find a library/approach for this. paperjs.org/ is what saved the day, they took on the complex and somewhat unthankful job of implementing fast and accurate boolean operations within an API and framework that is accessible and productive to work with. I've also used github.com/opentypejs/opentype here to get the text as a vector path, but it seems this might be do-able in Paper.js too, possibly.

While researching this I looked into scripting inkscape with python, directly using inkscape's libgeom library, using OpenSCAD, etc. ... so as you see I was already quite desperate. :D I'd be delighted to learn if you know of any alternative approaches to get the above-mentioned workflow done. This is powerful stuff and the more ways to do it, the better. :)

IPv6: I have a vague todo item on my personal agenda to "be less ignorant about IPv6" in my life. Any reading recommendations that teach basic practices that everyone can/should follow to help the cause? Or is it more productive to just "wait it out" until either someone yells at me why my $foo does not speak IPv6 or I finally crash against the wall in my own projects when my cozy IPv4 carpet is pulled out from under me at some point?

People who have broken into my toilet (2021 update) 

1) The painter who, without prior notice, painted the toilet ceiling after some water damage occured the week before (I came home and he was just in there happily painting as I passed by)

2) Random plumber that plumbed somewhere else on the same floor (I unwittingly opened the door to my toilet when I came home and that man was standing there, happily peeing)

3) A conglomerate of police, firefighters and the emergency medical service (one or all of them, don't know – breaking in while I was in my kitchen, unsuspectingly cleaning dishes and they were in parallel also breaking into my elderly neighbour's flat next door)

What the fuck. xD

TIL: I was totally unaware that recent-ish LCD screens still suffer from (albeit temporary) image burn-in. Somehow I always thought my OS/WM/graphics card/driver has some seriously weird pixel buffer/compositing issues but today (when a boot selection screen burned itself onto the screen lol) it finally dawned on me that the problem might be a little more down to earth. :D

Simon Repp boosted

Finally got around to writing down some audio programming tips for the musically inclined trying to learn, based on my own personal experiences and struggles learning how to do it myself: pbat.ch/wiki/audio_programming

Are there interesting examples on the web for deliberate extreme video processing/compression techniques that address the ever-growing ecological impact of video streaming?

I'm specifically looking for general purpose approaches that use ubiquituous tech (e.g. gif, h264 coding, ascii video), not patent-pending AI facial reconstruction tech or handmade ASCII movies ;) - although if you have some completely out-of-the-box thoughts on the topic I'd be very curious to hear about them too!

Ethical & libre voice recognition tech question, very vague: Are there applications/libraries out there that fit the description "beginner-friendly, approachable, hackable"? Quality can be subpar! :D

I spent some 3 hours on research and my findings so far are that there are capable FOSS projects out there which address ML scientists (i.e. not me), and then there is mycroft.ai/ which looks like a naive dystopic fantasy fulfillment with a pseudo-ethical facade (but I would love to be proven wrong about that).

TIL: "(Database) sharding" as a term was coined, or at least influenced, by Ultima Online's game lore.

"(...) Garriott's production team attempted to mitigate this [game balance] issue by separating the global player base into separate sessions, and rewriting part of Ultima Online's fictional connection to the end of Ultima I: The First Age of Darkness, where the defeat of its antagonist Mondain also led to the creation of multiverse "shards"." (Wikipedia)

And a 7min video article where Garriott recounts it in detail: arstechnica.com/video/watch/wa

Also there's now a --theming-widget cli option which injects a small overlay widget into all pages and with which one can interactively explore different color configurations for the visual theme - the configuration options are automatically generated as text and can be copied into the config file. (sorry for the weird video compression, wasn't meant to be an encoding glitch art video :D)

Show thread

Random faircamp progress excerpt: Added decoding (and thus waveform generation) for FLAC and WAV, experimented a bit with "waveform thumbnails" for the tracks of releases (might remain an experiment only), today added font customization (supplying custom files, default and custom system font stack options, etc.). I might make the system font stack the default, because environment, page load speed, non-latin scripts, etc., but let's see, in general I'm rather fond of bespoke typo choices and would rather eliminate other optional bytes on the page :)

Show thread

TIL: Take uncompressed audio with a sample peak not below -1.0 and not above 1.0, send it through an mp3 encoding/decoding round-trip, and it might come out with a sample peak below -1.0 and above 1.0 (!). This is (depending on encoding parameters) by design, and encoders expect decoders (or its consumers) to clamp/clip these peak overshoots themselves. Reality never ceases to surprise me with new intricacies where I didn't expect them. :)

github.com/notviri/rmp3/issues

Threw some other albums at my waveform computation logic, and I must say I'm beginning to like this idea a lot more °˖✧◝(⁰▿⁰)◜✧˖° The juxtaposed waveforms really add quite a unique graphical signature to the albums (even, or especially, with everything else removed), and somehow just seeing some of the curious waveforms totally triggers my curiosity to hear what's going on in those tracks. :D

Show thread

Been looking a bit into the topic of waveform representation in faircamp. I like the idea of integrating something that adds a varying graphical quality to all releases and tracks without forcing the artist to provide artwork, i.e. having the system generate something graphical from the audio itself, which also has functional value ... Delicate matter anyhow, not yet sure if there will be waveforms in the end. :)

Show thread

Skin cancer prevention now commodified as an app with subscription model, development and research naturally supported and co-financed by the state, hospitals, universities & patients' data - market you can take over for the profits now. Also tough luck if you're a little too un-white, sorry, no training data! Subscribe now and "be one step ahead in the race to beat skin cancer!" This is fine!

TIL: Unsplash is rotten to the core

In a bizarre blog post [1] from December 2019, the Unsplash executives/founders present their new ad strategy "Unsplash for Brands": Devoid of any sense of ethics, they neatly lay out how advertising on unsplash itself would be ineffective and annoying - after all these ads would be identifiable as such - so old school! Instead they present their "win-win-win" idea: Instead of providing unsplash users with a mostly unbiased photo selection from independent creators, they'll give them a biased selection (paid by megacorps) of photos from non-independent creators (made by megacorps) - all the same beautiful bland hipster photography unsplash is famous for, but now filled to the brim with BRAND PRODUCTS and BRAND VALUES.

Apparently without any sense of irony, without any self-awareness at all, they present a GIF that displays a full reversal of Apple's famous 1984 ad [2] to bluntly explain their vision: From here on out, all major news outlets on this planet will stop using diverse imagery to depict (e.g.) a laptop - for now they will, in complete uniformity, show that same brand image, portray those same brand values, and best of all, no one will be annoyed, no one will even realize, that the world has been eaten by ads.

[1] unsplash.com/blog/introducing-
[2] youtu.be/VtvjbmoDx-I

Update on the announced self-hostable alternative: Got tons of work done on the foundation over the last days (architecture, reading, transcoding, caching, syncing, configuration, cli arguments, ...). Today was the first time I did some design exporation (first glimpses attached - these are already in-browser screenshots of the functional interface, not mockups :)). I'm developing out in the open at codeberg.org/simonrepp/faircam °˖✧◝(⁰▿⁰)◜✧˖°

Show thread
Show older
post.lurk.org

Welcome to post.lurk.org, an instance for discussions around cultural freedom, experimental, new media art, net and computational culture, and things like that.