• 0 Posts
  • 54 Comments
Joined 1 year ago
cake
Cake day: June 21st, 2023

help-circle





  • Or seek to implement those features in a free & open way; but the features have to necessitate the effort & if the features are not clarified then no effort, especially a distributed one can even begin to replicate them. What do you suppose ought to be done first when building an app - the feature request or the code?

    In any case, I wrote this right at the top of the post before getting to the good stuff, so you could have stopped there if you wanted to:

    “We will release the design and/or reference code to the public when the initial version is stabilized.”


  • I was doing it more as a wishlist or an idea board to draw opinions from. However, as stated; as nice as it sounds, I’d rather admire the code than have the features without the code. Whoever did this though, was quite thoughtful about risks, including the - the ‘bus factor’.
    As a comparison, I really liked the “Connect For Lemmy” app when I first joined up.
    It has some great features, some of which are still not on Jeboa. The dev even said he’d open the source if there was interest. I gave up on waiting, uninstalled & now use Jeboa on mobile exclusively because Connect is still closed, as far as I know.
    I am FOSS or nothing, if I an help it - especially in communications apps. If the source fails to materialise then I will forget this too - even if I can indulge in speculating on the methods used. There is a fair amount of skill & thought required to pull this off. The best part though, I’m hoping, is that as a protocol, it would be a great chat accompaniment to & not a replacement to Lemmy or Mastodon. Again, without the source, I can only guess.
    Regardless, the fedi is young enough to be influenced by good ideas early, before it becomes too costly in time, effort & complexity, to undo or implement changes later.



  • That’s what he wanted everyone to believe - offering 420 jokes as a share price he was willing to pay. He was ready to pay anything & even said so in an interview - that X is not about money, it’s more than that. Though he complained about how much he paid for it, he got a bargain for the ability to centralise, control, monitor & distort the dissemination of knowledge, perception of culture, international wars, to prolong or incite culture wars.
    Imagine one guy deciding which governments could use it as an emergency response tool, tracking millions of accounts; many of whom were in positions of influence ie politicians, law enforcement, judiciary, reporters, authors, financial types, companies of all sizes, government agencies, local councils & everything in between.
    I remember thinking what a bargain he got that ability for. How much would any dictator pay for this sort of ability?





  • According to their FAQ: https://wiremin.org/#/FAQ

    “We will release the design and/or reference code to the public when the initial version is stabilized.”

    EDIT: After having fully read the FAQ, I must admit, their DHT (distributed hash table) protocol sounds very neat. Being something of a protocol geek, I’ll be the 1st to admit that I may be biased, so if a 2nd pair of knowledgeable eyes could try burst my bubble, just in case I’m missing something. I’d appreciate it. Thanks.
    According to them, albeit with no documentation, source plus a firstname.lastname@gmail address as a contact. The lead developer seems to be an O’Reilly publisher of MySQL books but I can’t confirm that until I hear back from the developer.
    Basically, these are some features:

    • “Wiremin is a protocol, not a service” - so no Terms Of Service
    • No registration - the app cryptographically adds users by device, so no email, phone numbers or contact reading to recommend contacts.
    • No data privacy concerns as the app is incapable of storing or collecting data,
    • No ads by default; therefore none of all the related drag that comes with that.
    • No central storage or processing servers - all done ‘on device’, E2EE,
    • Community driven, (I’ve asked for elaboration of how this is achieved)
    • No metadata tracking or leakage,
    • PoW (Proof Of Work) to stop DDoS,
    • Mnemonic backup of account info,
    • opt-out diagnostics after crashes
    • Unsend within 5 minutes…
    • “Information we share” … We don’t have user information of any sort, so nothing will be shared." They go on to disclose needing to share with law enforcement info about third-party services you interact with though, like Google, Firebase or Apple push notification ids - which you can turn off

    It really is quite impressive, technically & they seem to have coded themselves out of the equation as they can’t even see your password or recovery passphrase.

    Be that as it may - as you rightly point out, all we need now is the source, without which, I doubt I could back it.
    I’ve reached out requesting info on whether it will be full or partial source, roadmaps, transparency, funding for devs etc before downloading & trying this but it sounds fantastic - too much so, when compared to, say Discord or X.

    PS: My interest in this mostly academic intrigue, I don’t work for them or know them, I just found out about this a few hours ago. Most of the wording here is copypasta. I once wrote commercial protocols in the IoT space that were loosely based around early XMPP - until Google killed it. Which is why I’m so happy about the fedi existing despite the best efforts of Google.
    Anyway, this protocol seems to resemble the functionality of the FOSS app Syncthing, at least on the surface. Of course with the addition of other tech. I’d love to see one or two features like this in Lemmy or Mastodon, even if I had no interest in the code. The cost of infrastructure & the need to administer it, for example, would disappear overnight.