Pretty straightforward, really
Great experiences come from true utility and functionality. Not putting lipstick on a pig and calling it The Next Best Thing That Will Make You Happy And Productive.
I haven't had any real issues with desktop or mobile apps either.
Custom emojis are fun, inline markdown is useful, but it takes a lot of memory for something that, to me, seems like it should be lightweight. We've had IM since the 80's, after all. There's a part of me that has a visceral reaction to seeing an IM client taking more than 100mb of memory (though to be fair they seem to be getting improving that a lot.
Overall Slack is a net UX improvement over everything that came before.
It's a similar "lowest common denominator" issue with IRC, XMPP, mail and many others. Yes, XMPP can - in theory - do almost everything Slack does, but there's dozens of clients supporting a different subset of the protocol. Of course nobody wants to use it.
IRC is an early protocol and wasn't designed for extensibility purposes. A new chat protocol just needs to replace it.
- Jami
- Matrix
- Signal Protocol
- XMPP
The better question: why aren't new and existing technology companies embracing these protocols and making them ubiquitous?
Personally I think it's because companies have a vested interest in creating walled gardens. The more users that are forced onto a platform, the more money they make. It's against their business interest to create a product that works with their competitors.
[1]: https://en.wikipedia.org/wiki/Comparison_of_instant_messagin...
There's more to it than a simplistic "business interest" to explain it.
For example, the Signal protocol is free to use. No payments for patent royalties or software licenses required. Moxie Marlinspike's company Signal LLC (Open WhisperSystems) is a non-profit entity.
But notice that an organization that isn't pressured into making a profit and that's using an underlying free protocol -- does not want to federate with LibreSignal[0].
This an example of where the "openness of a protocol" runs into a hard reality of "closed wall of implementation". The protocols themselves can cost $0 but implementations (e.g. cpu+disk+bandwidth+staff) cost more than $0 which leads to complaints of it being "not really open".
If people do not examine the economic forces surrounding any protocol, they will always be perplexed why the internet is not as free & open as they think it should be. Even non-profit entities are not immune from economic forces.
[0] https://github.com/LibreSignal/LibreSignal/issues/37#issueco...
Because they don't want to be at the behest of protocol maintainers to improve their product? Their goals and the goals of maintainers would never align.
Open source is great, but open source moves at slow pace compared to what a revenue-driven company can. There's a reason that most of the big open source projects have a company backing them.
Complaints about corporations capturing HTML are not new; Google is merely the latest in the line. (That's not to diminish the fact that it's still bad every time. It's just this isn't the first time.)
People ascribe too much credit to a protocol. The real issue is the funding model to support spending money on that protocol.
By focusing the lens on the funding, the web wasn't as "open" as people think. The so-called "open" web was a government project which wasn't fully open to the public until about ~1995. Before then, you had to work on government-funded projects (ARPANET & NSFNET). The "open" web was actually quite closed off to people that didn't belong to an institution to give them an ".edu" or ".mil" address.
>I wonder sometimes what we're missing because the conditions that let the web grow out of the internet have disappeared.
The "conditions" were government sponsorship to pay for protocols.
- Vint Cerf developed the TCPIP protocol -- but he was funded by DARPA (USA government). Mr Cerf didn't have to go to venture capital firms on Sand Hill Rd and get funding to pay the salaries of his team to build a proof-of-concept or "MVP". He didn't have to worry about not making payroll if his TCPIP protocol project failed.
- Tim Berners-Lee developed the http protocol and also HTML language -- but he was funded by CERN...which is one step removed from being funded by a consortium of governments[0]. Therefore, TBL didn't have to wear a "salesman" hat and pound the pavement begging enterprise customers to buy his "http protocol stack" so he could buy his expensive NeXT workstation[1] to help build and test the protocol.
Those are prominent examples but it doesn't mean people can't create new protocols on their without government help. The SSL protocol was developed by Netscape -- a commercial business. I believe the the git protocol was developed by Linus without a government grant. (Also note that http v1 was funded by CERN (government sponsored) but http/2 & http/3 appears to be mostly funded by Google (non-government sponsored).)
But notice what happens with "open protocols" like SSL and git: they eventually run into a wall where somebody has to spend money on it. The SSL protocol is free of royalties but the Certificate Authorities issuing certificates are not free of charge. The git protocol is free but enterprise accounts for Github are not. The ActivityPub protocol is also free but Mastodon servers are run by altruistic admins spending their own money or requesting Patreon donations. Somewhere, money has to be spent and that ultimately constrains what the "free web" will be. Releasing a "free to use protocol" is not enough.
Therefore, it doesn't matter if the old IRC protocol is free of charge. People still have to spend money on auxiliary features outside the scope of the protocol. Slack spent money on those extra features (e.g. searchable history, etc) that nobody spent on IRC.
[0] https://en.wikipedia.org/wiki/CERN#Member_states_and_budget
[1] https://en.wikipedia.org/wiki/Tim_Berners-Lee#/media/File:Fi...
Reminds me of the "free as in puppies" angle.
AP doesn't have to run on HTTP, but any alternative has the same problem as anything else that needs its own client. Even browser support for FTP is on the way out, and it used to be standard.
> "Historically many key protocols, such as TCP/IP and HTTP, have come from researchers. Subsequent iterations of these protocols were often handled by nonprofit organization that tried to wrangle with more or less success the various commercial interests that sprang up around this protocols (the companies that were making and selling software and hardware based on them). The more money was involved the harder this became.
> "Now, however, we have a new way of providing incentives for the creation of protocols and for governing their evolution. I am talking about cryptographic tokens . . .
> "I can’t emphasize enough how radical a change this is to the past. Historically the only way to make money from a protocol was to create software that implemented it and then try to sell this software (or more recently to host it). Since the creation of this software (e.g. web server/browser) is a separate act many of the researchers who have created some of the most successful protocols in use today have had little direct financial gain. With tokens, however, the creators of a protocol can “monetize” it directly and will in fact benefit more as others build businesses on top of that protocol. Given this new incentive I expect a lot of resources to be devoted to protocol innovation."
~ https://continuations.com/post/148098927445/crypto-tokens-an...
People no longer release protocols, they release products. The open net, the open web has truly died.
I wonder sometimes what we're missing because the conditions that let the web grow out of the internet have disappeared. What next leap are we missing?