Now that we've reached that limit, what should be done? Should the limit be increased? Or is it, "Sorry, game over. We can no longer improve this answer."
Seems to be an edge case not worth bothering with.
When was the last time someone went through and checked that the books and PDFs on that list both:
- Are still online and available
- Actually free for distribution or whatever the beer was?
That's how you can improve that answer if you really want to keep trimming its hedges. Or use the inline version of Markdown hyperlinking. Saves you both characters and sanity tracking the titles to their corresponding links.
That list post is a mutant of epic proportions. One where people will slap on a new book here and there, but never quite go through the list as needed every now and then to prune its ratty pubic hairs.
Each of those sections - with the possible exception of "Language Agnostic" - could just as well be its own question, with its own wiki answer. And with its own tags.
And I wouldn't have to scroll through a nightmarishly-long list to get there...
Surely the folks who actually spend time hanging around a given tag are better equipped to answer questions like this anyway... Especially for the more niche topics.
Of course, if you want to still link all the questions together in a big master directory, that should stay well under 30K for a good long while. Or... Just link to a tag search and call it a day.
I wouldn't even be worried about it. Something like this is going to happen in the near future:
- People are going to raise hell about it being there
- (Optional) Prolonged debate about the worthiness of it's existence at all on meta
- Moderator deletes question
- More hell-raising on meta
- (Optional) Someone undeletes the question, but leaves it closed
While being somewhat facetious, there is a serious problem with that post. If the post is that large, just how much use can it actually have?
Who is going to go through and read all of those books? At this point, you might as well just search for "programming" on Amazon and you'll come up with pretty much the same thing.
How does one even decide which book on the most highly upvoted answer should be read first? It's impossible to tell and your feature request will just make that monstrosity even more unwieldy.
I can't see the reason for a mechanism that takes something that is unmanageable and exacerbates the very thing that makes it unmanageable.
- Curate the list. Review all the books in each category and present only the top three.
- Split the list up into more answers. Perhaps by popularity, by application, by interpreted vs virtual machine vs compiled, etc.
- Use a URL shortener for each link - should save a bunch-o-bytes
- Encourage someone to move the list off-site to a blog or elsewhere, and have them maintain it