Another traveler of the wireways.

  • 10 Posts
  • 43 Comments
Joined 1 year ago
cake
Cake day: June 12th, 2023

help-circle
  • Huh. I had to take a quick glance at the timestamp of this article to make sure it wasn’t old, as this was the same issue that inspired a short-lived fork a few years ago.

    I get where this article’s coming from, as I got where the people trying to fork it under a different name were too, however for better or worse I think what matters even more to people is that the software works well and has a good UX. If I remember right, I think Glimpse intended to focus on improving the UX but it didn’t last long enough to do much in that regard.

    Better than my memory though, here’s an article that gives some insight into what went into the cessation of its development.


  • At a glance this sounds even more intrusive than it’s been with Win10 (and maybe 11?), and sadly it’s no surprise as even without AI junk, I think the defaults with Win10 (and maybe 11) are to track your PC use to try to provide some “convenience” features, e.g. display of recently used programs/accessed files when you go to open a new desktop (Win key + Tab).

    If they would be more transparent about this and indicate whether and how much of that info, “anonymized/depersonalized” or not, is being taken by them, I think people would still be understandably annoyed but more understanding; at least with an easy opt out or better still, the default being that you must opt in for any of it.



  • When other’s, be it corporations or people start to decide which information a person can and cannot access, is a damn slippery slope we better level out before AI starts to roll out en masse.

    You highlight the bigger issue here than AI alone tbh. This is why another critical element is becoming literate and teaching each other methods of independent research, using multiple sources to develop an understanding, and not relying on any singular source, especially without careful review.

    All the technology in the world can’t help a person learn and understand, who hasn’t yet learned how to learn, much less understand.




  • How do you stay in the know about this kind of stuff? I’m curious about all the cool stuff out there I wouldn’t even know I’m curious to find.

    I was going to mention YaCy as well if nobody else was, so I can chip in to this somewhat. My method is to keep wondering and researching. In this case it was a matter of being interested in alternative search engines and different applications of peer to peer/decentralized technologies that led me to finding this.

    So from this you might go: take something you’re even passingly interested in, try to find more information about it, and follow whatever tangential trails it leads to. With rare exceptions, there are good chances someone out there on the internet will also have had some interest in whatever it is, asked about it, and written about it.

    Also be willing to make throwaway accounts to get into the walled gardens for whatever info might be buried away there and, if you think others may be interested, share it outside of those spaces.



  • […] but the profitable parts (algorithmic indexer and the app view with advertisements) are very much still in the hands of BlueSky the company.

    Yeah, we’ll have to see if/when they fully enable third parties to run their own indexers and app views to see how committed they are to all this, and even then as the thread you linked indicates, there would remain many questionable architectural problems to AuthTransfer.


  • I came away from reading over the AuthTransfer protocol and its handling of moderation/enabling users with a very major sense of, “We outsource almost everything!”

    As L. Rhodes writes:

    […] One effect of ATproto’s structure is to multiply the number of administrative relationships for which each user must decide for themselves—often on little to no information—who deserves their trust. The complexity of its infrastructures seems like it would sometimes make it difficult to assess when that trust has been betrayed, and by whom.

    So Bluesky may redistribute some technical power from host admins to users, but it also gives them much more to navigate. It makes their need for power more desperate, and I’m not at all sure that the power trickling down to them through those other layers of infrastructure will be sufficient to the need. No doubt many will compensate by sticking to the parts of the network operated by Bluesky itself—apparent choice, de facto lock-in.





  • Ideally these communities would be prevented from appearing in the “Trending Communities” list or local/global feeds unless someone other than the owner was subscribed to them, but wouldn’t be private in the sense that no-one could see them. Just they wouldn’t get wide distribution.

    This raises a distinct but interesting additional feature request that might complement “private” or exclusive communities, as well as others that might like to prepare a community before promoting it: a hidden or unlisted setting for communities.

    That would enable what you mention here, preventing their appearance from trending, and perhaps also user profile/data areas (i.e. if one can indirectly view others’ subscriptions, this might offer a way to obfuscate/hide that from others besides admins).




  • Yeah, I can see where you’re coming from on this. Personally I’m not a fan of Active as the default, yet I also don’t know what might be preferable to others. With that being the case, I thought it might help to highlight some ways to work with it in the meantime, especially given the outside perspective.

    For those here, I think it’s probably good to advise them to consider trying different sort methods till they find one that suits their preferences if they find themselves annoyed by the defaults, which has been happening for awhile already anyway.


  • I should elaborate a little, while I think commenting is important under Active sort for helping surface posts, I also think it is being balanced out by the vote scores. For a post to be surfaced you’ll want it to be both valued (upvoted) and commented on, then if you’re hoping for it to remain visible for awhile, you’ll want to see fairly steady commenting (that’s part of why you can see posts from days ago lingering around under Active, I think).

    I’ve not really noticed too many situations of flamebait style posts doing this, as I suspect they’re generally downvoted…Aside from the beating-a-dead-horse sort of posts that, despite receiving the usual, “Ugh this again” sort of comments, seem to otherwise be valued by some lurking voters.




  • I’d argue, oddly, that it’s easier to hold a single corporate entity accountable for data breaches than mastodon instance owners.

    It’s likely the case that both of are bad from a data security point of view, but at least with the corporations you know who to shout at.

    I’m inclined to agree, albeit I’m of two minds about it. On one hand, singular entity is technically easier, but being corporate means it’s likely to have more wealth/resources to make it untenable for people to hold accountable. Whereas on the other hand, if you put in the effort to pin down a Mastodon instance admin or even a few admins, chances are they won’t have those kinds of resources to really defend themselves, so you may be more likely to hold them accountable.

    That is, compared to a corporate entity which may drag things out for a slap on the wrist settlement/fine or the like. I can see the different angles to where you’re coming from though.