Communication networks should not be corporate controlled walled gardens.
[1]: https://matrix.org
[2]: https://matrix.org/docs/projects/client/riot-ios.html https://matrix.org/docs/projects/client/riot-android.html
Communication networks should not be corporate controlled walled gardens.
[1]: https://matrix.org
[2]: https://matrix.org/docs/projects/client/riot-ios.html https://matrix.org/docs/projects/client/riot-android.html
What did email do right that chat did wrong? Afaik, there was once a good standard for chat.
With the vast majority of email users using a tiny number of email providers, it effectively is. If Gmail decides that your personal mail server is sending spam, it becomes effectively useless.
The one big problem is group conversations, however.
Friends on iPhones use the iOS IM app but when they add someone to a group chat who isn't on iOS, they get the messages as attachments to MMS. Android's Messenger app handles this OK at the moment but it didn't always. And when SMS/MMS was integrated into Hangouts, it never handled it well.
Hangouts works fine for groups messages and at least you can install that on iOS but not everyone will want to use something else when they already use iMessage. Allo works but unless you use Allo, you get SMS/MMS from some random new number.
Basically, from what I gather, good group conversation support is a function of IP-based IM services and SMS/MMS was never really meant to handle that without workarounds and hacks. But SMS/MMS are a standard that any mobile device can handle whereas all the third party services and apps require either a particular OS or at least another messaging app to be installed.
Really would've loved to see a standard protocol with proper encryption and adoption by all major platforms but that doesn't seem to work for the Apples/Googles/Facebooks/etc of the world who would rather use Messaging as a way to keep you on their platform. Why would you leave when it would mean you can't keep those group chats you have with friends, family, and coworkers you've been having for all this time?
The number is not tiny if you include work email addresses: @apple.com, @ibm.com, @whitehouse.gov, @mit.edu, @bmw.com, ...
How do you know it's not a Google hosted email address if they use a custom domain (with a Google Apps plan one can use a custom domain on top of Gmail)?
I know from experience that I have to smarthost, as Gmail and Hotmail/Outlook.com reject silently any mail that come direct from my mail server. It's a pain, but there are solutions out there.
I mean, you can get a hosted VM and install Exim and Dovecot and drop in your domain name in the right places, but there are a lot of little things you have to deal with in order to get your mail to other people and others' mail to you without getting drowned in spam. Then you're fundamentally responsible for it and it can break while you're on holiday.
I used to do this, and still hang on to my own mail domain that I've had for 16 years, but I pay for it to be hosted now on a small provider.
I've seen this mentioned a few times on HN: https://mailinabox.email/
No experience with it myself.
It started out decentralised and became ubiquitous and mission-critical before any giants could gain enough market share to control it.
The only thing chat had close to it was IRC, but that has never been ubiquitous nor mission critical, and it's less attractive to newcomers than the proprietary alternatives.
Any federated/distributed chat system would have to deal with spam and abuse upfront in its design, or people would just migrate back towards controlled systems without those problems.
There's a second issue of where the chat continuity should be kept across devices, which favours hosted systems.
(There's no fundamental reason why you couldn't build an SMTP-based chat system by imposing convention on top of the protocol, favouring short messages with no Subject header. It's just that, as implemented, email is a little too slow.)
One solution is to only check for spam if the sender is not in your address book. And with chat, you usually have to allow somebody in your address book.
In a modern system, you'd probably want encryption and authentication of who you're talking to. Which is great except now you have to do key management, god help you.
Email is going to be interoperable forever, chats will become a 90-10 business like search. The only difference is that there will be different vertical chats monopolists for different markets (personal, business, developers, etc)
Email is old enough to be grandfathered in. SMTP dates back to 1982, long before the commercialization of the Internet.
Chat doesn't have that history. It was first popularized by proprietary apps like AIM/ICQ, Yahoo!, and MSN. When Jabber/XMPP came along and created an open standard, it was a reaction to the proprietary apps, which all had significant mindshare. Facebook and Google were the only major players to adopt XMPP, and that was just temporary until they managed to get their proprietary solutions off the ground.
Also, remember that there was a time when casual users would go through the set up of local/desktop clients in order to simply read/write email - e.g. outlook express, thunderbird, apple's mail app, etc. Whether we call it setting up a server, or installing/setting up an app., eventually it won't make much difference.
This is why we've made a real effort with Olm & Megolm (http://matrix.org/git/olm/about/) in Matrix to ensure that E2E is there in the baseline, and will be turned on by default for all private rooms. (disclaimer: I work on Matrix)
That being said, I have this nagging idea in my head that the more chat apps that get thrown at people, the easier it may be for an open service to prevail.
Hipchat, Slack, Matrix
Windows, OSX, Ubuntu Linux
Democrats, Republicans, Libertarians
.NET, Java, Open Source Dynamic options -> Ruby/Python/Javascript/PHP/Elixir/Perl