My question was simultaneously answered, put on hold, and finally deleted. Why, and how can I improve my question?

Question

Regarding this question: https://softwareengineering.stackexchange.com/questions/393863/how-to-intelligently-discover-and-track-meaningful-changes-in-html-with-a-c-qt

I laughed, because the answers/opinion provided in the comments, are very useful to answering the question, probably good enough to accept as an answer, hence this meta post:

Seriously, those four comments almost completely suffice as a good answer, even if it merely attempts to dispel the practicality of what I wish to achieve, although I'd say the XKCD comic vastly over-exaggerates the complexity of my question.

In case the question was closed because it was too broad:

According to the principles here: Why questions about "the correct way" are too broad

Namely:

- State your criteria.

• The answer ideally needs to take into account, that the backend is Qt/C++ oriented, so I can interface easily with the other libraries I depend on. I can do things like inject JQuery into the webpage, or track printfs and the like. If it matters, the current engine I am using is based upon Chromium, https://wiki.qt.io/QtWebEngine

• In short, I need to handle HTML with a C++ backend instead of a Javascript Backend, which is a significant difference given one is Synchronous and one is Asynchronous.

- Be specific.

• I chose html specifically. Not xml or anything like that.
• I laid out a specific example, with opacity
• I gave my language and framework, Qt/C++ along with its WebEngine based on Chromium

- Focus on requirements.

• Like for example, lets say that only one value changes in a 100,000 line html document changes; Do I reparse the entire document, compare, and find the change, or is there a way to avoid all of that and fish out the difference?

• In short, I require the intelligent way to parse in the worst case scenario.

- Tell us what you want to achieve, and ask us how you can achieve it.

• My original strategy of just detecting every change in the html string and parsing it on an interval, seems to me, to be a very bad approach.

• What API design philosophies, or existing API's, should I employ when interfacing with HTML, that will give me the correct signals to do code injection, or when to scrape data?

• Maybe this could be worded better, but I kept it semi-broad in case strategies or APIs for doing this sort of thing already exist, or have already been established
• If not, I was going to suggest a black-list or white-list approach. Ask whether it is possible to save processing power by parsing with diffs, and how that can be achieved.

If my question is really that poor despite all of this, how can I improve upon it?

Would perhaps asking three different questions be appropriate:

• How to reparse HTML using diffs to save processing power?
• Advantages and disadvantages to doing a whitelist approach to parsing and tracking HTML?
• Advantages and disadvantages to doing a blacklist approach to parsing and tracking HTML?

Answer 1

First, let me say Robert did not really answer your question (and I don't mean the fact he only used the comment sections):

• The comment to your first bullet point gives just a starter for an idea of one possible solution, but I would not call this an answer.

• The comment to your second bullet point tells you very clearly why this part of the question is unanswerable, and why Robert thinks it is too broad.

So I don't see those comments as a valid "proof" your question is really answerable, quite the opposite.

Note further, Robert's former meta post "Why questions about “the correct way” are too broad" is not an exhaustive definition of the "Too Broad" close reason. Questions fulfilling all the mentioned criteria can still be "too broad" if

• they have many possible answers

• would require answers that are too long for the Q&A format of this site.

See also this older meta post, which contains a more basic explanation of the "Too Broad" close reason . When Robert wrote his meta post, I suggested him to add a link to that one - seems we forgot this during our discussion, but we can still fix this.

So if I were a mod, would I have done the same?

Though I share Robert's opinion in this case that especially the second part of your question is too broad for this site, I think closing and deleting your question by a single person so quickly is quite undemocratic. The very least thing I would expected from a mod is giving a askers a polite and clear explanation for the deletion.

My preferred way of action here is to leave such a decision for a question which is not obviously crap to the crowd - just because I think something counts as "too broad", others may have a different point of view, and I would like to hear them first.

Sometimes someone manages to give a very general answer to a very general question in a concise manner. By deleting your question so quickly, one removes the chance of finding such an answer from the community. It also makes the time frame for the asker very small to improve the question.

However, this is not a guarantee - I think it is not unlikely that the community would have closen your question as "Too Broad" either, just a little bit later, in case you would not have narrowed down the focus of it.

Would perhaps asking three different questions be appropriate

Maybe, but be aware asking about "advantages and disadvantages" of a certain approach is just a poll, and polls are usually closed as "too broad", as described in the meta post I already mentioned.