Great question. Speaking from my experience, it is a trade-off you have to consider. Imagine LiveView will have something out in October, having to wait until the next April to be able to talk about it in public, it is definitely too long.
For example, I announced StreamData for Elixir before it was part of master, and it turns out that it won’t be part of Elixir core anyway. But the discussions that happened in the months after the announcement were very productive.
So I don’t think doing the announcement before having something out is bad. We already have another thread about LiveView with @tmbb, @grych and I discussing possible implementations. I would just not jump to conclusions yet.
I’ve started to test Raxx due to its purity and simplicity. Why not adapt Drab for Raxx?
That way you wouldn’t need to throw away anything of the amazing library you have, you wouldn’t collide with/get in the way of LiveView and even better, I’m sure you would find ways to expand Drab using a different approach.
I would love to see your genius around a new approach to code web apps.
The reason I built Drab over Phoenix is because with Phoenix I have a lot of stuff already done, for free: channels, base eex engine, safe html, views, rendering. Especially channels made my life easier. Drab started as a proof-of-concept page, and I would probably not be able to do it if I need to implement channels myself.
Also, the popularity of Phoenix is important. You may take your existing project, and very easy add Drab to it.
By the other hand, if it would be completely independent, or based on the lighter framework, it might be better, because some stuff may be done differently. There is always a tradeoff.
Just like Raxx and Phoenix/Plug can co-exist in the “webserver” space Drab and a potential LiveView library can do so for server-side managed client side updates in Phoenix.
In my opinion it would’ve been nice if @chrismccord would’ve acknowledged Drab in his keynote (it might have prevented a bunch of discussion here) but I don’t get why people feel so competitive about Drab vs. LiveView right now.
Hindsight is a wonderful thing… I remembered some time after making our 2018 update (and then felt terribly bad bout it!) that I had forgotten to mention that Chris had deprecated the Phoenix Talk Mailing list in favour of the forum some months earlier (where the unity and consolidated traffic obviously helped). Sorry Chris!
(Hmmm I wonder if I can edit it in now somehow… ) (Edit: Done!)
I can understand both positions. But I make a distinction between EXPLORING new approaches and EXPLOITING what really emerges as the proven approach. While using Plug/Phoenix in production apps I started to Explore Raxx because I can see its merits too. The same with Drab/LiveView.
I can feel the pain of someone who devoted so much time and passion in developing (exploring) something that proved to work and now the Exploitation may be done under a different library.
Why not keep exploring that concept together with another exploration project that is Raxx? That is in line with what José Valim was proposing.
We need to keep exploring. I would be really sad when I see the author of Drab giving up especially because he paved the way in a brilliant manner. We can keep doing both, I think. This is what makes Elixir community so great.
Let’s say they did know: No acknowledgement of its existence whatsoever, just railroading through announcing something that has yet to have technical merit with no care of the consequences.
If they didn’t know: key people navigating blindly in their own community certainly fits the “out of touch” description. There’s a long thread about Drab on this very forum, it’s exceedingly likely that people who actually spend some time even just passively with the community will have seen it and read it. Unless they don’t care, of course.
Because you wanted to was a terrific reason to start the language because you didn’t have the position you have today in a community. It’s not a legitimate way to think when you have the capacity to change things just by announcing something at a keynote. Even if you do start a project, have the tact to plug what’s out there, maybe.
You are not a regular member of this community. If you did take some kind of public stand against something I hope you know that would require the project to somehow be morally reprehensible or the like.
The reason they have a voice to begin with is because they’ve contributed so much. If you think things should be handled differently, all you would have to do is contribute as much or more than them. I don’t think anyone writing open source has any obligation to anyone. If someone wants to contribute in a more optimal way, do it. If they want something else that’s fine too.
While your post raises some defensible points, I don’t think this part of your post has any basis on reality. You are ignoring the enormous influence that @josevalim had in the ruby language and community prior to developping Elixir. I’m a bit ignorant of Elixir’s initial history, but Inthink many people here have stated that they only decided to check ou Elixir because of José’s position in the Ruby community.
As an issue of degree, sure, but I’m well aware he was known in the Ruby community. It really has nothing on the position he is in today, though, especially with the way this community has evolved thus far. There is a special kind of responsibility that comes with being this kind of key figure.
)