Celejar wrote: > On Thu, 24 Jun 2021 01:25:37 +0300 > Andrei POPESCU <andreimpope...@gmail.com> wrote: > > > On Mi, 23 iun 21, 17:12:07, Michael Grant wrote: > > > > Apparently the lines are blurry enough for you to include Signal in > > > > that > > > > list. > > > > > > Why? Not blurry at all. Signal is just as closed a system as > > > WhatsApp. Maybe more private, but unless you know something I don't, > > > Signal doesn't talk to anything other than other Signal. Puppeted > > > bridges are not interoperability, as far as I am aware, all users > > > still need to be on Signal. > > > > You seem to be using a completely different meaning of 'proprietary' (no > > federation) than I do (closed source software, proprietary protocol that > > must be reversed engineered, patents, etc.). > > Well, Michael's original post that you challenged contrasted: > > > a standards based system such as mail or the web and a proprietary > > system such as facebook, WhatsApp, Signal, Telegram, etc etc. > > Would you call Signal "a standards based system?" I understand that the > software itself is open source, and the project does publish various > "Signal Protocal" libraries, but I'm not sure that's quite enough to > call it "standards based."
Rule of thumb: it doesn't matter whether a system is based on a standard or not; it matters how easily it plays with others. For word processing systems, you want the input and output formats to be open standards, but you don't care about the internals. For communications systems, the question is whether an outside party can easily use their own software to interact with others. At one extreme is email, where humans can telnet to port 25 on a reasonable system and send a message. At the other end is anything where you can't use a client or a server that isn't produced/managed by the central authority. Despite Signal making some of their source available, you can't write your own Signal client and have it talk to their official servers. -dsr-
