5 ms·
The point of Lisp is to use it to create good tooling. Whether it's MOP or Racket's macro system or ELisp... the point has always been maximal power and customi
by srcreigh 5y ago
The point of Lisp is to use it to create good tooling. Whether it's MOP or Racket's macro system or ELisp... the point has always been maximal power and customizability.
Lisp is the system to create systems. It's not the endgame. It's not a finished product. That is the point.
What is better tooling? What improvements can be made? How can it be done?
These questions are wonderful questions to consider and experiment with. Every lisper does - it's the entire point of using lisp. Everybody has answers.
There are, however, some lispers who, never thinking they could provide all the answers, spent countless days to create tools to make it easier for other people to express their answers. Macro systems... ELisp... MOP...
We shouldn't expect all the answers to be given to us. It's not wrong, but it's so limiting.
(P.S. JavaScript is what happened when Brendan Eich was tasked with adding Lisp to the browser, but was asked by the suits to make it look like Java. The entire area of thought comparing JavaScript tooling to Lisp tooling runs really deep. Are they even different?)
- overgard 5y agoThat sounds kind of nice in a weekend tinkering project way, but what about getting actual product made? If I'm using something like typescript, I download a free IDE, optionally install a couple plugins, and in ten minutes I'm productively actually making something and I can easily share it with the world for them to modify. The Lisp thing sounds like I spend the rest of my adult life trying to build the perfect setup instead of getting on with actually making something?
- diskzero 5y agoIt sounds like you have found a solution that works well for you. I don't like to tinker around with my coding environment, which is why I pay someone to do that for me. The primary purpose of my software development efforts is to ship products and generate revenue. Anything that I can do to optimize this process, I will do. If I thought using VSCode and Typescript would reduce costs and increase efficiency, I would use it. This simply isn't the case for my use case. If your primary use case is developing websites, then VSCode and Typescript are perfectly good solutions.
- overgard 5y agoI guess my main point though is that most other languages don't require complex/expensive tooling though. Like, I also use VSCode for writing C++ microcontroller code and writing python scripts or writing C# game code for Unity. And not just VSCode so I don't sound like a shill here, I'm also able to use the same set of unix utilities, other generic tools, etc. without having to adopt an entirely new tool stack. I just look at the Lisp situation and it's either something like Emacs which just does not appeal to me, or proprietary and expensive tools where I worry if the vendor will even be around in a few years
- brabel 5y agoI am moving away from VS Code and IntelliJ for this exact reason actually: on emacs, there's ALWAYS a package that supports anything you want. And it's so lightweight compared to either VSCode and IntelliJ, it's really refreshing... Have you tried emacs/SLIME? It really isn't complicated, in fact it's the most straightforward stack I've ever tried except for maybe JS right on the browser.
- srcreigh 5y agoIn March 2020, somebody asked PG what language/tech he would recommend for competitive advantage, if they were willing spent 1000 hours learning it ("Is it still LISP?"). He said Clojure. https://twitter.com/ndehouche/status/1240218141742878720 https://twitter.com/ndehouche/status/1240218141742878720 Personally, I like to imagine PG scoffing at the idea of spending only 1000 hours (6 months, 40 hours per week) learning Lisp, then sarcastically recommending Clojure aka macro-less, continuation-less, Java-lisp. You want to spend 10 minutes? Use Typescript! ;) (I love Typescript too btw.)