Are questions about language design on-topic?

This is a good question. It depends on how much this site is about getting language X to do Y versus the higher level concepts of software design and computer science that should largely transcend specific languages.

We seem to mostly get the first kind of question here. As a result, I think that's what this site has evolved into, regardless of what it may have been first conceived to be. That's also what the largest volume of questions are about all across the internet. Compare SE's Stack Overflow (coding question) to its Computer Science site. SO has way more volume. This reflects the industry. There are way more "grunt" programmers than high level architects.

That would say that your language design questions don't belong here. However, I'd hate to turn them away. It would be nice to have a place on Codidact where such things can be discussed, and not get drowned in a flood of posts about how to make the green widget turn purple when someone hovers a mouse over it on an odd Tuesday.

Ideally, we'd have a separate site for software architecture and computer science. But, I'm sure that would be just another ghost town if we created it. Instead, I think the solution is to have a separate category here for Software Architecture. The five questions/year wouldn't seem so bad on a site that otherwise has decent (for Codidact) traffic. The low volume of the separate category wouldn't really hurt much.

If it becomes successful, then we consider splitting it off as a separate site.

posted almost 2 years ago

CC BY-SA 4.0

Olin Lathrop‭

1069 reputation 1 24 90 11

The ambition of this site was always to give more room for subjective and big picture questions compared with Someplace Else. However, I believe programming language design falls under the topic of general computer science. It doesn't match anything in the current on-topic list.

General computing/programming-related big picture questions posted here in the past had mixed reception. Some got a lot of up-votes, some got a lot of down-votes. So it seems to depend a lot on how good and well-researched the question is.

For example, I don't think a question such as "What are the advantages of compiling versus interpreting?" will be well-received. It's a very broad question and an answer would have to be quite lengthy in order to be fulfilling. Also there's the in-between scenario of byte code, which is used by languages like Java and has evolved over the years. An answer would have to dip into that as well, because simply staying at "compiled vs interpreted" sounds a bit like various long-since outdated discussions from the 1990s comparing C++ with Visual Basic (and the end goal was always a rant over how bad VB was :) ).

Generally, I'd hesitantly say that questions about program language design would be welcome, but they must be carefully phrased. Particularly:

• Does this question list all the relevant requirements of the programming language getting created? Or is it yet another "what is best" question without a "best" criteria?

• Is this a question someone can answer without writing a whole essay?

• Is this a question I can somewhat easily research myself online or by reading a book on the topic.

  In my past experience of similar questions, a lot were met with the standard comment of "go read the Dragon Book". Except few who recommend it has actually read it... I haven't read it myself, so I have no idea how good it is.

  I think we would welcome questions from people who have done that level of research and answers from people who knows even more. It would sort under "hard to answer niche questions" but that alone doesn't make something off-topic.

posted almost 2 years ago

CC BY-SA 4.0

Lundin‭

8917 reputation 36 121 1025 165