• 0 Posts
  • 453 Comments
Joined 2 years ago
cake
Cake day: July 1st, 2023

help-circle






  • Look, I get it, but people are allowed to have complex relationships with their language and history. I don’t know the specifics of this guy and I’m not exactly defending him (or his complaints, which seem petty), but nobody owes anything to a language because if their blood.

    I’m Irish, and the oppression and near-loss of our language is a real pity, but I can’t deny either that we have have 150 years of real, actual Irish people speaking, writing, creating, singing, dreaming in English. Yeats, Joyce, Beckett, O’Brien, Heaney, Lynott, Sinead O’Connor, Samantha Mumba… they’re not less Irish for having created or performed in English.

    It can be hard for Irish, native-English speaking people (i.e. to a statistical approxination, literally everyone) to understand the amount of effort and resources that are poured into funding Irish-language art that the majority of our modern nation cannot read or understand, in a language that they associate mostly with an abusive and failing school system and patriotic guilt, while English-language art in general struggles with the odd and cold assumption from our society that the ready market for English-language Irish art and culture abroad will pay for anything with actual reach. If you have a weird niche idea, you’d better make it even more niche by sticking a cúpla focail in it - now a tiny fraction can enjoy it, but at least you get paid.

    The same applies every time I am frustrated with the state of our healthcare system in a waiting room and pass a tall stack of pointless support resources and documentation in the Irish language, the cost of which could have paid for at least one dose of medicine.

    We have very little going for us resource-wise in Ireland, but by christ we can write in the global language at least, whether it be for job opportunities or art. If someone sticks a knife in me I shouldn’t be called a traitor for keeping the knife.


  • The Octonaut@mander.xyztoComic Strips@lemmy.world"Politics"
    link
    fedilink
    arrow-up
    4
    arrow-down
    9
    ·
    9 days ago

    It is absolutely inane. You know what the poorly drawn, oddly-sized, barely present, not-even-bleeding knife-wound adds? Nothing. Who reading this requires a stabbing to understand the context? Nobody. How many months and from how many thousands of MAGA morons has she profited in advertising her titty photos with bland, humourless, safe comics? She could have had no knife and made it clear that yes even just supporting MAGA is stillbdetrimental to your “friend”. Or she could have leaned into her idea and had a knife-wound that actually looks like a knife-wound. Nope. Make it silly, take the edge off, keep it safe, please don’t unsubscribe if you’re not actually knifing transpeople then you’re probably fine.



  • There’s also the element of wasting the blind person’s time. I work in enterprise software and our application meets WCAG guidelines but… it’s a busy, text-heavy, actions-heavy application. It can take 5 minutes for a screen reader to read the entire page. Other websites are worse - images as buttons, flavour images and hero banners and icons everywhere. Again, a much more accessible version is just presenting them with what they actually want - a cleaner, leaner, more contextual page like the ones we built in the 90s before images loaded instantly.

    So part of me wonders if blind people actually enjoy “listening to memes” or if they’d rather skip it and hear a text-based joke or an audio/video joke. I did specifically say I wanted a blind person’s opinion on it.

    I think you’ve underestimated the wordcount and I think you don’t get how memes are shared if you think adding 4 minutes to their re-transmission wouldn’t matter. I cars that blind people enjoy the internet but I absolutely do not think “searchability” is a good reason to transcribe “Drake meme but it’s an animal girl. Top panel. Animal girl looking repulsed. The item she is repulsed by is the logo of a Linux package manager called Flatpak…”


  • Are there any blind people on Lemmy, screenreading this? I get why alt-text is useful functionally on things like application interfaces, and instructive or educational text, but do you actually enjoy hearing a screen reader say "A meme of four oanels. First panel. An image of a young man in a field. He is Anakin Skywalker as played by that guy who played Anakin Skywalker in the Star Wars prequels. He says ‘bla bla bla’. Next frame. An image of a young woman. She is Padme as played by Natalie Portman. She is smiling. She says “bla bla bla, right?”









  • the accepted terminology

    No, it isn’t. The OSI specifically requires the training data be available or at very least that the source and fee for the data be given so that a user could get the same copy themselves. Because that’s the purpose of something being “open source”. Open source doesn’t just mean free to download and use.

    https://opensource.org/ai/open-source-ai-definition

    Data Information: Sufficiently detailed information about the data used to train the system so that a skilled person can build a substantially equivalent system. Data Information shall be made available under OSI-approved terms.

    In particular, this must include: (1) the complete description of all data used for training, including (if used) of unshareable data, disclosing the provenance of the data, its scope and characteristics, how the data was obtained and selected, the labeling procedures, and data processing and filtering methodologies; (2) a listing of all publicly available training data and where to obtain it; and (3) a listing of all training data obtainable from third parties and where to obtain it, including for fee.

    As per their paper, DeepSeek R1 required a very specific training data set because when they tried the same technique with less curated data, they got R"zero’ which basically ran fast and spat out a gibberish salad of English, Chinese and Python.

    People are calling DeepSeek open source purely because they called themselves open source, but they seem to just be another free to download, black-box model. The best comparison is to Meta’s LlaMa, which weirdly nobody has decided is going to up-end the tech industry.

    In reality “open source” is a terrible terminology for what is a very loose fit when basically trying to say that anyone could recreate or modify the model because they have the exact ‘recipe’.