Masto meta thoughts
The biggest threats to the Fediverse:
1. Mastodon dot social might just decide to never close registrations
2. Facebook/Twitter might build ActivityPub support into their product (does your instance have a policy against federating with them?)
3. New government regulations ostensibly designed to curb Facebook's excesses might contain requirements for social media services that can only be met by very well-funded mega-corporations
- takes roughly 10 seconds to load and several seconds to save (on a fast SSD in a reasonably new laptop)
- crashes if I have Franz open and then keeps crashing after displaying the splash screen on start-up until next reboot
- to the best of my knowledge cannot reliably export its data to...well, anything
- has multiple weird rendering glitches in the UI
I should have gone with Ledger. :(
Just mending these completely normal jeans https://scholar.social/media/NgWRdIc7vQnK_55mGBE
This spring I'm doing my Master's Project on (and, I think, in) #rust, the language. Besides a toy kernel for an OS course at university a couple of years back, I don't have any concrete experience with it and I probably should. Does anyone have any suggestions for good first projects to learn the language?
I'm tempted to make a mini CP-inspired minimal constraint solver, but it might be a bit too much.
I'm a current PhD student at Edinburgh Uni, studying the learning dynamic of artificial neural text models. Information about me aside from that:
- I have a condition that makes it painful to type or write. I dictate all my code and interactions with dragon, talon, and voicecode.
- I play roller derby! My name is Gaussian Retribution. I'm back on skates from months of coalescing after tripping over the flower girl at a wedding.
- 🏳️🌈 ♿ 🔯
- I &
If you didn't yet know, Google is datamining GMail etc for purchase confirmation emails amongst other things and aggregating it for their own purposes in an extremely detailed way.
(Everything you purchased with order ids, detailed billing details, VAT, including subscriptions etc)
Do not believe a single word about how this information is kept private. There is zero reason to build a system to aggregate this and then not use it to make money out of.
Also, it interests me that the systems we are currently building are typically designed for a typical use case. Who needs Cassandra, Hadoop, et al? Very big companies crunching either fast or large data sets do. Those are probably not the problems of a democratic computing community not built on advertising and social manipulation for profit.
I have a feeling that the FLOSS movement in spirit was constructed in the previous centralised computing paradigm with timesharing systems, and in some sense is left there. I wonder what its corresponding movement of our time would be, now that the pendulum has swung back from the desktop computing paradigm and into dumb terminals over a document format?
It frustrates me that many machine learning topics I have encountered seems like both bad maths and bad software engineering with all the (program) variables named x-hat or q-prime where that has absolutely no relation to their function, and all functions are represented by equally opaquely named matrices.
Hey since this is blowing up let me remind all of you that I literally wrote a book on this & related subjects.
I'm looking for an IM solution to replace Facebook Messenger, does anyone have suggestions? Onboarding needs to be nice to non-techies, work on Android and iOS. App needs to not look like crap or crash. Decentralised is appreciated. Group chats required, but not in the same app as non-groups. End to end encryption mandatory.
I have a few things I do routinely to weed out bad software; I put spaces in file names, use Unicode characters, etc. If someone didn't do a minimum of QA, these things surprisingly often trigger bugs, and I can avoid discovering that the program is a piece of garbage further along, when I have committed time and effort to it.
Using a case-sensitive file system in macOS is, surprisingly, one of those things. Massive swaths of software just doesn't start.
Scholar Social is a microblogging platform for researchers, grad students, librarians, archivists, undergrads, academically inclined high schoolers, educators of all levels, journal editors, research assistants, professors, administrators—anyone involved in academia who is willing to engage with others respectfully. Read more ...