Hacker News Reader

Comment by pmontra | Hacker News Reader

My emacs instance is currently editing projects in Elixir, Ruby, Python and Node. None of them remotely resembles what it would be in a Java like IDE. There is no need to "understand" a project in those languages, edit save and reload has been enough for me and my customers since 2006. For Java, I used Eclipse and Netbeans. I consider the need of an IDE as a bad smell in the design of a language.

View on HN

I think you are used to your way of doing things and your opinion is limited by your perspective. I would suggest you be open to new things, try a trial version of an Intellij IDE(with an open mind)

It makes logical sense that a program that can understand the entire project provide you more help then a more limited program. For example you refactor some code and make a typo in a function name, now because you changed the function name this function is no longer used from other code(because that code uses an old name), because is not used anymore it is highlighted in the IDE as unused and you just caught a typo error (I am expecting now that TDD would have caught this, or a typo in a var name but not everybody is testing everything and the problem is caught in the IDE without any effort from you)

I mean, that's a pretty basic feature that Emacs supports with flycheck+LSP.

I can't judge that, i tried VS Code for web dev and with it's defaults it is a big downgrade , for me it seems that LSP is not enough, you need to have something that parses the entire project libraries from vendors/mode_modules, your own JS/PHP , the html/template files and after all is parsed create a big dictionary with all symbols and what we know about them.

Can your plugin go that?

I'm not sure how language servers handle multi-language support (JS and PHP), if at all, but language servers usually do parse all the project's dependencies as well as project code and give you jump to definition, autocompletion, and popup documentation on all of it.

And I consider programming languages not designed with IDE tooling in mind a design smell for the said programming language.

Any language that can be compiled is by design suited for IDE tooling. Think about it, if a compiler can analyze the code then so can an IDE.

An example of a language that can't be analyzed by IDE tooling is the English language. Case in point the English language can't be compiled either.

The parent poster is basically saying that there are languages that are so complex that to even program in these languages you (usually) need IDE tooling to assist you. C++ and Java are good examples of this.

Naturally it can, but it will always be an half baked experience, which is why languages like Python and Ruby are still miles away to replicate Common Lisp and Smalltalk development experience.

I rather prefer C++ and Java than any of those "I don't need IDE" languages.

I have been using IDE based languages since I became a Borland customer around 1991, and while XEmacs allowed me to survive without IDEs to the point that I still remember most of my way around ELisp, I don't miss those days.

If you were designing a programming language with IDE tooling in mind, what are the things that Common Lisp and Smalltalk do to better assist IDEs than Python/Ruby?

Anders explains it better than me,

"Anders Hejlsberg on Modern Compiler Construction"

https://www.youtube.com/watch?v=wSdV1M7n4gQ

Basically ability to see the language AST at all time, do time checking with broken code, incremental compilation, REPL and IDE editing experience is fused together.

Here is another short examples,

"7 minutes of Pharo Smalltalk for Rubyists"

https://www.youtube.com/watch?v=HOuZyOKa91o

"MountainWest RubyConf 2014 - But Really, You Should Learn Smalltalk"

https://www.youtube.com/watch?v=eGaKZBr0ga4

IDEs do a good job with dynamic languages too, if you document your types and function definitions it will provide you same features as a static language. You need to avoid abusing eval or similar constructs like p[kind+"_name"].

Also worth pointing out that even static languages often have escape hatches into metaprogramming that defeat tooling.

I've seen Java codebases use reflection heavily enough to utterly confuse the IDE.

Yes, just to add the information in case someone is wondering, for example PHP has "magic" functions , I do not use this feature but is used by ORM libraries to make it easy to access table columns. But using PHPDoc I can specify the existence of this columns and then all the warnings are gone and I can get auto-completion help(this helps a lot to improve coding speed and can catch errors if you put a type in a table column name.

This will not work for stuff that is random object from a JSON but so far I never worked with random data.

Reply on news.ycombinator.com