Great initiative. I often wonder why Racket isn't used more in the industry, especially when you see much more niche languages being used...
One note about your site: You are subscribing people to a mailing list without sending a confirmation email?
One note about your site: You are subscribing people to a mailing list without sending a confirmation email?
Regarding the subscription: I'm using Mailchimp. A confirmation mail should've been sent out, if you subscribed. I will look into it. Thanks for pointing that out.
Languages like Java make it much easier to treat your programmers as interchangeable parts.
Ah yes, The Slashdot Cynicism.
Guaranteed to get a +5 Insightful.
What is your experience with Racket? Or failing that, Scheme? Or, failing that, Lisps in general?
Early 2010's, I remember my company had a fairly fast, stable and small(LOC wise) data feed processing application in Perl. It just worked, barely any bugs, maintained by 2 - 3 programmers.
Time came around, and your Director level people realized the teams were too small, programmers were having too much fun and more importantly it made programmers felt valuable and hard to be treated as kleenex.
They decided to give some Architect Astronaut love to the application. Then came in Java programmers. Code base size exploded by 500x. Design pattern floods followed. DI to manage complexity which shouldn't have even existed. Hundreds of classes to do things that could be done in hundreds of lines in Perl.
Managers routinely boasted how they managed to grow their teams, about their culture, and how big and great their teams were. They could never make it work as well as it did in Perl. I too moved on. But eventually I saw they just shut down the product.
No manager likes to see their work less valuable or better than that of a programmer.
Languages like Perl, Scala, Clojure, Racket, Common Lisp are never going to see decent adoption in the industry.
I know from what I've read outside of that class that Racket is powerful enough to implement any other programming language as a library. A single programmer leveraging all these features can write an application that uses bespoke domain specific languages for every component. This is wonderful for a solo project but good luck passing that on to the new recruit.
P.S. guilty as charged. I am an old Slashdot reader.
* Poor editor support. Apart from emacs and vi, many widely used editors support syntax coloring, but few do any proper indentation, and none (AFAIK) will do anything like intellisense. Only DrRacket really goes the distance on all those points and it hasn't kept up with the likes of Visual Studio Code or Sublime etc. You may say, just use Emacs! But if another dev doesn't use it they'll just ask for the language plugin for their current editor and then be mystified and frustrated at why they can't be productive.
* Poor performance. Out-of-the-box new Racket code will be slow. You can "precompile" it using Racket's built-in tools, which is great. But the process is complex and error-prone, and deploying your "precompiled" Racket code across a team in heterogeneous environment is not straightforward. My teammates are regularly stumped by how to keep our code running quickly. Racket also lacks effective profiling tools to address this.
That said, as Bjarne Strousup opined: "There are languages people complain about, and there are languages no one uses." I have used Racket in production environment for many years.