Here’s a rant on Om-Next:
https://www.reddit.com/r/Clojure/comments/3vk58p/a_rant_on_om_next/
I’m not sure anymore what problem Om Next is trying to solve. My first impression was that Om Next was trying to solve the shortcomings and complexities of Om (Previous?). But somehow along the line, the project seems to have lost its goal.
It looks like the author is more keen on cutting edge ideas, than a boring but pragmatic solution. I, and probably many here, do not have the fancy problem of data normalization and such. Well, we probably do, but our apps are not of Netflix scale so that applying things like data normalization would make a noticeable difference.
What would actually solve our problem is something that can get our project off the ground in an afternoon or two. Clojurescript language is enough of a barrier for aliens from javascript, we do not need yet another learning curve for concepts that would not contribute much to our problems. Imagine training a team for Clojurescript, and then starting training them Om Next – that’s won’t be an easy project for the rest of the team, or you.
My guess is that Om Next will end up where Om Previous ended up after the hype and coolness dust settles down. Timelessness is hidden in simplicity, and that’s something that the Ocham razor of Reagent delivers better than Om, Next or Previous.
Legit?