So I'm afraid the whole "Is that Matrix?", "Yes, I'm using FluffyChat/Riot/Whatever to connect to Matrix" conversation is still going to be there. Just like someone might say "ooh, what Web Browser is that?". This much is a feature, not a bug.
If I could wave a magic wand, formatting matrix usernames as “[email protected]” instead of “@alice:alice.com” feels like a similar opportunity to make things more friendly to newcomers.
It’s these little things that make a huge difference to success. TL=com/ORG=ycombinator/HOST=news addressing vs news.ycombinator.com. Or calling Outlook an “MUA” instead of just saying it’s a “mail client”.
It’s technically correct. Some people say that technically correct is the best kind of correct, but often it feels like the most convincing strategy is the one where you don’t need to add explanation or bon mots to win over opinion — people can just see it’s right.
I’m surprised that Matrix didn’t think they would benefit from trying to become well known as Matrix first, then diversify into FluffyChat etc second.
In terms of Matrix IDs, we very very deliberately made up the @user:example.com format to avoid folks ever confusing them with email addresses like [email protected] (and to hint at a federated form of @mention notation, a bit like ActivityPub subsequently did). We find it super confusing that SIP URIs and JIDs look like [email protected] if you drop the scheme prefix (as often happens), given there's zero reason why [email protected] on email should be configured to correctly route to the same entity as sip:[email protected] or xmpp:[email protected] - and in fact the chance of it working in the general case are incredibly low (thanks to aliases, filters, routing rules, etc).
That said, we do not expect Matrix IDs to be used in the general case - we've always wanted people to discover each other based on their existing identifiers (email, phone number, etc). In practice this hasn't gone smoothly as nobody has properly solved decentralised identity (or if they have, they haven't hooked it up to Matrix yet), so we're in a bit of a limbo. But the solution is definitely not to make Matrix IDs look like email addresses.
So, my point here is that this is not "technically correct is the best kind of correct". If we renamed Riot to be "Matrix Client" we'd be quite rightly crucified by every other Matrix client developer out there. Just like Mozilla would if they renamed Firefox to be "Web browser" (even though they were effectively responsible for the modern web browser as Netscape).
However, we absolutely have made things even more unnecessarily complicated by maintaining separate brands for Riot, Modular and New Vector, hence consolidating them together. But we simply can't get away from the fact that the end result will be a Matrix client, not the Matrix client.
I mean I don't think the browser analogy is that great, it seems more the exception than the rule. Eg most IRC clients have the word IRC in them, or something very close (eg mIRC, irssi, etc). Most email clients have the word "mail" in them. Torrent clients, same. Why shouldn't a Matrix client have the word Matrix in it?
Just call it iMatrix or uMatrix or something like that, and you're removing a lot of confusion. If mIRC and uTorrent and Gmail can get away with that, so can you.
(and I also think PicoTorrent will get more downloads because it's easier to Google for)
Such as thunderbird, icedove, mutt, gnus, mew, outlook, outlook express, etc?
Gmail, Hotmail, Yahoo Mail are email services first and foremost. Most native clients do not have mail in the name.
Yes, but mail isn't the same as SMTP or IMAP or POP.
In public communication, it almost never helps to start a sentence with "well, technically".
The name told you what you could expect the technology to do. That's why it made sense to include these words in the name of email clients.
The same is not true for "Matrix".
Chat seems to be closest?
So MatrixChat might work?
Genuinely asking.
Well, that's only sensible—you're not sending an SMTP message. SMTP isn't a message format. It's a wire protocol for passing along pre-formatted messages. What you're sending is a MIME message, over SMTP. (Just like you send HTML over HTTP.)
The distinction is even clearer in the email case, actually, because those same MIME messages also show up in other transfer protocols (POP, IMAP.)
There are actually a few email clients that use "MIME" in their names. But it's pretty rare.
Conveniently, though, "MIME" is short for "Multipurpose Internet Mail Extensions", so "Mail" is actually in the name. ;)
Those get named after the overall framework (if they're not just named "fooVPN" or simply "foo") even though there's no single IPSec protocol, and many many different underlying protocols.
In English, the R makes for a good pun.
There are decent terms in the screenplay of the Matrix movie to draw from.
Despite nobody being able to spell most of them, the ship names would be good. Nebuchadnezzar, Logos, Icarus etc.[0]
There’s also other terms such as “construct” (the local Matrix instance /site before you enter the real matrix)
Only 2 problems in computer science I guess.
And the first Matrix client was called Matrix Console. Much like all of the above, it sucked - it was a first generation thing I wrote in Angular on the Caltrain the day before we launched Matrix at TC Disrupt 2014.
So I'd say the analogy is pretty good. Riot (originally Vector) is a second gen client designed to make Matrix mass market - just like Netscape did for the Web. You can get away with the first crappy experiment being named after the protocol, but beyond that point it's just gauche.
If they had called FluffyMail they probably wouldn't be the juggernaut they are today.
Branding matters.
That’s because swathes of people simply conflate the two. To a lot of people the icon for their web browser may as well just be called “The Internet”. I’ve heard that some single digit percentage of Google searches are people searching for Facebook.
I think you give too much credit to the level of technical knowledge of the average person.
You’re a lot less likely to try to browse to an incorrect spelling of the domain name.
Bookmarks, the solution you are looking for is bookmarks.
Work any sort of tech support and you'll find that many people who are otherwise intelligent, capable, and at the top of their field look at a computer as a box that they do certain specific things to, and it does those few things that they understand. They've never needed to delve, so they don't, and things even as simple as Windows shortcut keys outside of Ctrl-X/C/V, or any browser use beyond open it, type in the search box can be utterly mystifying.
Then take a person at that level of understanding who is paranoid that if they do something wrong they won't know how to get back, or that they'll lose their very carefully memorized workflow, and try to teach them something new. Even one small change that would save them surprising amounts of effort.
Most people _aren't_ imbeciles, but when it comes to things they don't know on "that box", the edge of the world is very sharp and bottomless.
They understand mouse, keyboard, screen,... and Internet, which is their browser.
This is not to make fun of them. I drive a car and only know where to pour gas. I have zero interest in anything else in the car, I just want it to work. Carburator? Spare tire? Not interested.
If you like car analogies so much, the car equivalent of bookmarks is programming the radio's bookmark buttons to tune your favorite stations. I am not a car mechanic, but I sure as shit can do that much. For that matter, I also know that my car doesn't have a carburetor but has a spare 'donut' tire in the trunk. You too probably know these things of your own car. Again, I am not a car mechanic and neither am I some sort of polymath genius with advanced general knowledge across a broad number of domains. Cars interest me about as much as American football, which is to say not even remotely. I'm just a regular Joe who's not a complete imbecile.
I work for 25+ years in IT in huge technological companies. I can tell you that in these technological companies the level of knowledge about things above the very basics (and how to use a bookmark is way above that) is poor to say the least.
This is something I know very well, not some kind of vague deliberation. If you have managed IT operations in a large company you would know. I therefore assume that you do not.
Now - expand this to the whole population.
And as for the car - I had to check if I had a spare a few weeks ago, when the car mechanic who was coming to fix my punctured tire asked me. He had to direct me as to where to look. I am probably an idiot, of the kind of idiots who are not interested in cars. But I do not call the ones who are not the same.
If you were stranded with a punctured tire in an area without cell service, would you have died from exposure? Or would you have figured out how to change your tire? I don't think you're an imbecile, so I think you would have figured it out. It's an easy sort of problem for an adult human to solve using little more than common sense, even if you've never been taught how to do it and never cared to learn anything about it before. The general population is capable of much more than you give them credit for.
You introduced the shoelaces level unnecessarily at the start of this whole debacle. Computers are, to the people we're holding under discussion here, a BAZILLION times more complicated. You can't make your shoes vanish by holding a lace slightly wrong. You can make your work 'disappear' in any of myriad ways, by accidentally mistaking a single keystroke, or leaving out a step, or doing something 'extra' as one person put it to me.
And the tire analogy is also skewed. Tires are another thing that's phenomenally less complicated than computer use, to someone who hasn't studied it yet.
You're holding up much simpler concepts and saying that if people can figure out a 10^3 complicated situation, they can figure out a 10^50 complicated situation.
So yeah, you indeed never worked in IT then. Never mind, this discussion is already beyond ridiculous.
Though "the web" and "the Internet" are harder to keep apart for many.
50% of the time when going to Facebook I mess up the typing and end up searching for it instead. I suspect that means I’m in those stats.
For people like my mom, I've explicitly told them to search things rather than type in URLs directly.
For downloading software for which I don't know the publisher's name, I usually look for the Wikipedia article and check the "official website" external link. Sometimes I even double-check the article's history, just to be sure.
I get that the environment is heterogeneous. It just felt a bit too lacking in clear branding compared to something like Signal.
“Is that Matrix?”
“Yes! Although technically matrix is the protocol. This is FluffyChat connected to my local HackerSpace’s Matrix instance.”