Salamander

  • 2 Posts
  • 26 Comments
Joined 3 years ago
cake
Cake day: December 19th, 2021

help-circle
  • I did not know of the term “open washing” before reading this article. Unfortunately it does seem like the pending EU legislation on AI has created a strong incentive for companies to do their best to dilute the term and benefit from the regulations.

    There are some paragraphs in the article that illustrate the point nicely:

    In 2024, the AI landscape will be shaken up by the EU’s AI Act, the world’s first comprehensive AI law, with a projected impact on science and society comparable to GDPR. Fostering open source driven innovation is one of the aims of this legislation. This means it will be putting legal weight on the term “open source”, creating only stronger incentives for lobbying operations driven by corporate interests to water down its definition.

    […] Under the latest version of the Act, providers of AI models “under a free and open licence” are exempted from the requirement to “draw up and keep up-to-date the technical documentation of the model, including its training and testing process and the results of its evaluation, which shall contain, at a minimum, the elements set out in Annex IXa” (Article 52c:1a). Instead, they would face a much vaguer requirement to “draw up and make publicly available a sufficiently detailed summary about the content used for training of the general-purpose AI model according to a template provided by the AI Office” (Article 52c:1d).

    If this exemption or one like it stays in place, it will have two important effects: (i) attaining open source status becomes highly attractive to any generative AI provider, as it provides a way to escape some of the most onerous requirements of technical documentation and the attendant scientific and legal scrutiny; (ii) an as-yet unspecified template (and the AI Office managing it) will become the focus of intense lobbying efforts from multiple stakeholders (e.g., [12]). Figuring out what constitutes a “sufficiently detailed summary” will literally become a million dollar question.

    Thank you for pointing out Grayjay, I had not heard of it. I will look into it.









  • I ordered four of the simpler devices this weekend (LilyGO T3-S3 LoRa 868MHz - SX1262) and I have been reading about antennas.

    Since I live in a city I am not super optimistic about the range. But I am still very curious about the concept, and I would love to be surprised.

    After doing some search about antennas, I have decided to test the following combination:

    I also have a vector network analyzer (LiteVNA) that can be used for checking antennas, so I will also try to build some antennas myself. I doubt that my custom antennas will approach the performance of the professional ones… But I just find it such a cool concept.

    Have you already gotten to play with it? What is your experience so far?





  • I have been running an instance without a slur filter for about a year and a half. It is not a big instance, but big enough to have some experience in the field.

    In case you are curious, 100% of the many times that I have encountered the n-word in my instance it has been in the context of a very banable offense, and it often requires spending some effort investigating and purging images from the database. The slur filter would block many these federated posts and comments from reaching my instance without the troll/spammer getting any feedback about this.

    The filter can be a useful practical tool. The reason I keep it off is because I’m stubborn about not policing the words that people can and can’t say. But when I consider what I have experienced and reflect about this, I become more and more skeptical about my choice. The problem is still manageable for my small instance, so I can keep the slur filter off. But I can see that when dealing with this problem at a much larger scale one would want to use any tool at their disposal to make the job easier.






  • I have just tested by uploading/re-downloading an image, and the EXIF data is removed.

    I then looked through the Lemmy issues and found this issue related to the image-uploading back-end (pict-rs) removing the EXIF data. In response to this issue, the developer of pict-res (asonix) comments that striping the EXIF data was one of the original motivations for building the uploader.

    I am not sure about how to search through the source code of pict-rs, and it seems like this step is not properly documented in the readme file, so I have not been able to find exactly where the metadata removal operation takes place. I think that this is done by invoking ‘exiftool’.



  • Salamander@mander.xyztoAnnouncements@lemmy.mlLemmy v0.18.0 Release
    link
    fedilink
    English
    arrow-up
    1
    ·
    edit-2
    1 year ago

    In the admin settings there is a tab that allows admins to add an icon. So far I added :lemmy_hearts: and :mander: to test. I am posting from Jerboa now and the custom icons did not get invoked. I don’t know yet if you can also invoke them from a different instance.

    What I did notice is that the icon gets fixed to a size that is bigger than a normal icon, and the upload button doesn’t seem to be working - i had to feed the form the direct url to an image.