Racket doesn't have much frameworks.  You will probably do from-scratch a lot more in Racket.

Any time you think "framework" from some other language, probably either it is easy to do what you want with the Racket base language features or a small existing package, or you will have to put substantial work into writing a new package that does what you want (or that interfaces with a non-Racket tool or service that does that).

One of the advantages to writing it yourself is that you can then understand it, and make it do what you want, and not be at the mercy of the framework and its developers.  (Example: One mostly-Racket Web&service system, with necessarily complicated data models and some kinds of big data, was able to evolve over many years, with a very small team, who had to do most things from scratch, and, despite this and other constraints, it ended up being a noteworthy first among AWS deployments.)

Of course, some things (e.g., R, a GPU ML engine, PostgreSQL) you usually don't want to reinvent from scratch.  But even then, sometimes it turns out you don't really need, say, an RDBMS, and maybe, say, a from-scratch replicated RAM-based object graph works much better for your purposes.

Other times, the biggest practical drawback from doing from scratch is not the cost of writing it, but developer career moves after.  If you do your work well, you will probably understand some underlying technologies and system behavior a lot better than most framework users will.  But the majority of contemporary developer recruiting, after the first post-college job, is based around commoditized specific framework and language keywords.  I'm not sure, but I suspect many employers might also want you to say "agile", "scrum", and "bro" a lot, preferably with an affable Californian surfer accent. :)

--
You received this message because you are subscribed to the Google Groups "Racket 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to