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

help-circle


  • dan@lemm.eetoAsklemmy@lemmy.ml*Permanently Deleted*
    link
    fedilink
    English
    arrow-up
    53
    arrow-down
    1
    ·
    9 months ago

    On mobile: multiple top and bottom tool/nav bars that automatically show/hide themselves when you scroll. They’re invariably more irritating than if they were just pinned at the top of the page (or perhaps viewport, but ideally page - I can scroll to the top of I want it back)

    On desktop: animations tied to scrolling.

    Anywhere: any kind of popup, modal, etc that I didn’t click on something to get. Please fuck alllllllll the way off.


  • dan@lemm.eetoAsklemmy@lemmy.ml*Permanently Deleted*
    link
    fedilink
    English
    arrow-up
    17
    arrow-down
    1
    ·
    9 months ago

    The browser implements the text selection behaviour, but how infuriating it is depends on how convoluted your page construction is.

    On a simple page with no floats, overlaid elements, negative margins, absolute positioning, hidden stuff, and other css layout tomfoolery, it’s perfectly predictable. It’s only when designers do designer things does it start to break down.





  • dan@lemm.eetoProgrammer Humor@programming.devOrder
    link
    fedilink
    English
    arrow-up
    34
    ·
    edit-2
    11 months ago

    Lossless compression algorithms aren’t magical, they can’t make everything smaller (otherwise it would be possible to have two different bits of input data that compress to the same output). So they all make some data bigger and some data smaller, the trick is that the stuff they make smaller happens to match common patterns. Given truly random data, basically every lossless compression algorithm will make the data larger.

    A good encryption algorithm will output data that’s effectively indistinguishable from randomness. It’s not the only consideration, but often the more random the output looks, the better the algorithm.

    Put those two facts together and it’s pretty easy to see why you should compress first then encrypt.







  • Technically speaking, things are far more secure today than they were back in the dawn of the internet. Protocols are now almost exclusively encrypted where they almost exclusively weren’t. Private communication is (in theory) easier to achieve.

    Practically speaking, however, now there’s always somebody there attempting to monetise your interactions. To mine useful information from what you say, or to sell you something while you say it. Or both.

    That’s only going to get worse with the rise of AI, as companies realise the vast databases of past interactions might actually be worth something.

    Best you can really hope for these days is to retain some anonymity and some separation between your public personas.





  • I mean yes I agree with all your points. But I stand by the assertion that it’s too expensive. I could handle $5/month, perhaps, but 300 searches is waaaay too few. That’s 10 per day. I did 10 searches this morning before I got out of bed.

    For unlimited searches it’s twice the cost of a streaming service. Yet it has negligible bandwidth costs, and significantly less storage cost, probably less development cost. Sure a small user base too, but at that price they’re really going to struggle to grow it!

    It’s really just too expensive.