is there any sample code of using om.next with google-app-engine ?
not from me, alas. but i do have a major project in the works that clojurizes https://www.polymer-project.org/1.0/, if that helps. ;)
based on 30 seconds of skimming, I don't find any relation between polymer and om.next
i assume you're talking about https://github.com/omcljs/om?
Om is really overcomplicated. I stay away from it and use reagent
I recently did a project with reagent on AWS Lambda.
Worked really well.
@agreed.
Not thrilled with Lambda (nothing wrong with Lambda, it’s just no where near as cohesive as app engine)
agree there too, lambda got nuthin on gae, once you get all the administria out of the way.
om.next is more than just om
it's also a dsl for fetching dta from remote sources
like netflix/falcor
@qqq: om, reagent, etc. all really great but in the end they're hacks. good hacks, but hacks. polymer, well i suppose it's a hack too, in a way, but it's based on the stds.
iirc, reagent/re-frame is an alternative to om, but they have nothing for om.next
@mobileink curious why reagent/om are considered hacks?
yeah! defend your statements! 🙂
It all boils down to a virtual DOM. In reality it’s not that hard to write your own virtual DOM in Clojure because the language is really good at that sort of thing. The reason why React is so important in JS-land is because it’s actually not easy to write that yourself in that language.
(and have it be correct)
there's also making things compatible across multiple browsers, handgling bubbling semantics,
the core algorithm is not difficult, but the engineerkng effort is non-trivial
:badminton_racquet_and_shuttlecock: still kinda fetishizing complexity.
Clojurescript is already cross-browser so that’s a non issue.
i guess i don’t see why its a hack then; just because it’s not written in clojure itself?
Well I didn’t say it was a hack
also I personally would shy away from writing a virtual dom, i suspect there’s a lot more incidental complexity than at first glance
but out of curiosity, anyone know of any clojurescript implementations of a virtual dom?
Peter wrote this one to prove that it wasn’t that hard when you have a language like Lisp.
Now Peter is a fucking genius. Don’t get me wrong. But he also just sat down and wrote this thing for fun to see how hard it was.
And it wasn’t.
was this in a talk? would be interested in seeing it
A couple of months ago
I missed it so I bought him dinner and he did the presentation for me at a restaurant.
I also bought him drinks.
He needs to charge more for that.
Peter did a lot of the work on cljgae-template
too.
So even though all the commits were mine he laid a lot of the groundwork.
@nickbauman pays for talks in the best ways 🙂
lol
☝️ PETER SIGHTING
@mobileink with that context your comment makes more sense
@jsselman what city are you in?
I’m in Seattle
Oh. We’re in Minneapolis.
You probably need a visa to come this far into the interior these days anyway...
Or a blue card or something
Hah - I should engage more with the community; I’m pretty much just using clojure as a hobby language though. Haven’t had any opportunity to use it professionally
Most stuff I write is pretty much about keeping data structures in sync 🙂
I guess my first-experience with reagent/react was that it was a very straightforward way to develop very dynamic UIs at a reasonable layer (e.g. not too abstracted so you can still use html, but the virtual dom protects you from having to worry too much about specific browser differences)
but I’m still new to frontend development; most of my experience is backend jvm stuff so this could all be wrong
reagent coupled with re-frame is a pretty cool experience. Every thing is just a function. the atom
part of things is nicely abstracted away.
yeah i’m using re-frame at home, added by own coeffect hander for the websocket protocol i was using. took me a long time to get things hooked up properly though