elle VS biff

Speaking very loosely, primitives on the JVM are values which are represented directly in memory, instead of as pointers to objects on the heap. Clojure generally treats everything as a pointer to a heap object. There is no specialized equivalent for, say, a vector of shorts, or a map where values are floats. The compiler can emit specialized function signatures for... IIRC longs and doubles, but other types (e.g. byte, float) aren't directly accessible--they go through widening conversion. It's also easy for the compiler to quietly fail to recognize it can preserve primitives in some kinds of loops, so you wind up with what Java calls "autoboxing": wrapping a primitive in a corresponding Object type.

Here's a recent example of some code in a hot path inside Elle, one of Jepsen's safety checkers. It does a lot in primitive, packed structs and bitmasks to avoid pointer chasing.

https://github.com/jepsen-io/elle/blob/main/src/elle/BFSPath...

There was actually a Clojure version of this earlier that got pretty close perf-wise, but I wound up dropping to Java for it instead:

https://github.com/jepsen-io/elle/blob/913cbff5ebb19ba850c0a...

I definitely believe Clojure needs a rails. Not only will it help beginners get started, if it can help people get started faster and build fast like Django and rails do, I think it'll help more with adoption.

Biff and fulcro seems like they have a shot at this

Biff- https://github.com/jacobobryant/biff

Fulcro - https://github.com/fulcrologic/fulcro

Jacob is doing a fantastic job with https://biffweb.com/ If the Clojure community would focus more of its manpower on such projects, then I think we can make Clojure the obvious choice to start a software business, by saving an insane amount of time. And time is by far the scarcest resource in a startup.

If you can get away with not using React, I highly recommend Biff. It uses XTDB and Rum by default but they can be swapped out pretty easily for Postgres and Reagent. I'm planning to publish some docs on how to do that when I have a chance.

The best one of these imo is https://biffweb.com

I'm also really liking the strategy of the old-school is new again with sever side rendering serving actual HTML instead of JSON for certain things, using HTMX, an example can be found here: https://biffweb.com/

Take a look at Biff project https://biffweb.com/

+1 for Malli and XT! For the relevant parts of Biff, see the example app's schema and the transaction reference docs. Biff has its own transaction format which includes schema checks via malli and various other conveniences, and it gets translated into XT's lower-level transaction format. Might provide some inspiration at least.

Rum is used throughout, though mostly via middleware[1], so you (almost) never see any calls to `rum.core/render-static-markup`. But all of the hiccup-style data structures (`[:div "foo"]`, etc) do get rendered by Rum.

htmx doesn't render anything on the backend; rather it gives the frontend more ways to interact with the backend. e.g. say you make an inline form--htmx gives you the ability to display/submit that form without refreshing the entire page, but all the html that's sent to the frontend is still getting rendered first by Rum.

[1] See https://github.com/jacobobryant/biff/blob/6353c406adef034448... and https://github.com/jacobobryant/biff/blob/6353c406adef034448...