Welcome to the future friends when we must pay to battle the keymasters and gatekeepers on their terms to do what we used to do ourselves, for free.
First they took my RSS, then they took my XMPP... Who can guess whats next?
Luckily, we're in the middle of a decentralization renaissance. Instead of managing your identity on a server controlled by Facebook or your XMPP host, you can manage your identity on a blockchain controlled by no one. We can build identities that any app can use. When it comes to network effects, federation loses to centralization, but centralization loses to decentralization.
Pick up the tools—Ethereum and IPFS—and start building systems without gatekeepers. We will win.
UX is hard. Getting great user experience takes IMHO a lot more pain than getting algorithms and systems to work right (in most cases). UX is about endless revisions, a lot of boring and often ugly code, obsessing about pixels and quirky details of your UI's "story," and other things that are very time consuming, not fun, and that the vast majority of programmers hate doing.
As a result, people generally have to be paid to work on UX.
This is doubly true because most nerds don't need great UX. They know how things work and can use "raw" hacker projects from the command line. So there's really no motive in most cases. "Works for me."
Until and unless there is an economic model for a less centralized less winner-take-all Internet economy, vertically integrated closed silos will continue to dominate if for no other reason than the fact that they can pay people to do the boring tedious work that separates a nerd project from an app non-nerds want to use.
Edit: It's really always been about UX to some extent. Personal computing won back in the 70s and 80s because it offered far superior user experience to closed silo mainframes. But the Internet and especially mobile have changed the game pretty dramatically. Now closed silos are winning because they offer superior user experience.
It's not entirely a technical problem. At some point I think a few sacred ideas will need to be questioned, such as the idea that open source must also be "free as in beer."
If there is no economic model, nothing is going to happen. Closed silos will continue to dominate in all the areas that cost money, and among those UX is probably the most significant differentiator that will guarantee their dominance. I love open source gift culture but there are some pretty severe limits to what can practically be accomplished without some source of capital or subsidy.
It might not work, but if it does, no one can stop those crowds. If I find a new shiny object to chase, the software will keep running as long as people want to use it. We can reshape our world with software that can't be stopped.
We've strayed too far off topic, but you and any readers should feel free to email me anytime. It's in the profile.
Source please. Absolutely 0 non-tech person uses blockchain technology besides paying for ransomware maybe.
I don't understand why open seems to immediately mean better and proprietary is bad. I'm all for an open system if the benefit is clear, but it often makes no difference and ultimately comes down to how the project/software is managed. There are a lot of "open" software products that are controlled by a few and thus, not very open.
Well, we ended up with tens of platforms with X, so obviously companies went with one of the latter two solutions. We know a few of the companies did continue with XMPP and modified ejabberd, so they definitely did XMPP implementation of video for example. Yet, they decided not to publish anything, so globally we're stuck in the same place again.
Similar thing happened to HTML with ages of stagnation. Now we finally standardise what's implemented instead of implementing what's standardised. I don't see how XMPP itself is an issue. Don't like that they're slow? Use your own extension and publish it. Either everyone will use the same solution, or you'll learn why it sucks.
I do feel excluded from various online chats but for anything important, they text, call, or email me. So perhaps it is not too bad. I end up saving time by avoid gossip and funny videos.
That is why I left whatsapp, it has been a remarkable experience, https://www.goodreads.com/user_challenges/4336854 I read around 12books this year itself, https://medium.com/@surajp/i-crippled-my-phone-too-1300761aa...
It is a relief not looking at the phone every minute of the day despite the fact that nothing important is going to be communicated via whatsapp, or nothing of that significance to me, everything that I really really need to know comes to me via call
(Not singling Facebook out -- there have been lots of open standards and every chat platform has moved away from them.)
When you're small, you have no users, and need to start gathering network effects. So you champion and implement protocols that allow you to interoperate with other networks. You start to eat other companies lunches as you provide a better product.
As you get larger, that same interoperability starts to become a liability as other networks start eating in to your userbase. So you cut out interoperability and start building a walled garden.
A bottoms up approach where we put the protocol first instead of experience is the "Architect Astronaut" way of looking at solving this problem and should be avoided - http://www.joelonsoftware.com/articles/fog0000000018.html
Guess I'll just tell everyone that matters my phone number and mail address, if they do not already know, and call it a day. I've wasted so much of my time and attention on facebook.
Of the current crop of messaging platforms, I'm cautiously leaning on Signal (it works with sms, I might get some of the people I actually communicate to use it -- even if I would prefer it was more federation/multi-server/self-host friendly) and http://www.mattermost.org/ for an xmmp/irc/etc work-a-like.
I've given up the idea of using IM with facebook users, iMessage users, or Google hangout users. I might occasionally use slack or gitter[1] via the irc interface, or very rarely via the awful web interface (not they're any worse than other web interfaces, but it's just as gaudy and slow and insecure as other web apps, by virtue of being a web app (image libraries, font rendering have issues, they require downloadable code (js) to run -- they can be expected to have a security record comparable to MS office documents with macros enabled)).
I somehow don't see the irc interface to gitter and slack as being good ways to integrate with a single Android app though. Maybe mattermost will help with that, so I can have one app that does sane IM+SMS, and one that does email. Hopefully without the "5 different accounts that can't talk with each other"-nonsense of having separate AIM, XMPP, MS messenger accounts in a single monster-client. It was barely tolerable on the desktop, I won't abide it on my cell phone.
I am indeed sad people couldn't just get together and agree on a sane XMPP sub-set with client-server and server-server TLS, and optional server-side message storage.
But that didn't happen, so now I'll just be without a "big" messaging/chat service (other than IRC, IRC still works).