question 3

question 2 (answer 2)

If the web page is Mandarin, the charset will not be UTF-8. I've never changed either meta tag. Again, it's been a boilerplate "blank.html" that I copy and paste every time.

question 1 - 2023/08/17 21:30 - Thursday night question 1

You sent me an abbreviated version of the charset meta tag. I would not assume that version will validate. Note that my version is longer and does validate.

As I said in the initial post, those first few lines are "boilerplate." They don't have to change. I've never changed them. It's also boilerplate in the sense that it's more aribitrary than stuff in a programming language. One doesn't have to declare such things in programming languages. This is all to say that among all the things to understand, all of the gory details of the boilerplate are lower on the list. You asked, though, and I have to appreciate all questions. So...

I'll let you look up metadata to whatever extent you need to. The short version is that meta or metadata is data about the data. I'm not sure how valuable the meta tags really are to a browser, but they are in the HTML standard as necessary, so they need to be there.

UTF-8 is closely related to ASCII. Remember that everything in computing comes down to numbers and conceptually speaking, 1s and 0s. A computer doesn't directly know what a letter is. A letter is represented by a number in the ASCII chart. You can run various "od" commands to get an idea of what files really look like.

More specifically, ASCII is 7 bits or 128 characters, and UTF-8 is 8 bits or 256 characters. The "Extended ASCII Codes" are on that page, too. UTF-8 is more or less "Western language character sets," or English plus. Or what data looked like before the computing world got around to Mandarin, Arabic, Russian, etc. I am nearly certain that UTF-8 is precisely the original 7 bit ASCII and then the extra bit for 128 more characters. ASCII and extended ASCII. UTF-8 is as opposed to UNICODE. UNICODE probably allows a few dozen human languages by now plus emojis plus who knows what else. I don't even know how many bits / bytes UNICODE is up to. At least 4, I think, but probably more.

So the meta tag says roughly that page is a Western language. There is also a popular character set that SatanSoft defaults to, and I see often. "Latin" something. That might make Western European languages easier to render.

As for "my" version, "text/html" is a MIME type. I'll let you look that up. In short, though, it repeats that this page is HTML. Even though that seems redundant. I am not totally sure it's necessary for validate, but I saved my boilerplate years ago and haven't questioned it.

initial post

This is roughly the simplest valid HTML page. The following validation stuff is not needed for validation, but it proves validation. Press Control - U in most browsers for "view HTML source code."

For initial lessons, most of the stuff in the header / head / HTML "head" is boilerplate. You can consider it as decreed by the HTML gods. The "title" is what you see in a page's browser tab. That is not boilerplate, of course.

Please do click the validator image. I'm running my own validator these days. If you look at my older pages, you'll see the W3C's validator, which sometimes breaks due to recent privacy changes. But that's another story.

Very, very few pages online are HTML valid. I think it's very important, though, both for learning and professionally. For one, in Python or JavaScript or any other dev language, you cannot get away with syntax errors. The program will not run at all. Such things will crash a website. So your HTML should be syntactically correct.

My latest apprentice asked if the timestamp and version below will update automatically. Right now, no, but that sort of feature is not so hard to accomplish. Right now I'm doing it manually. Also, right now one has to refresh the page to see the latest. Making it auto-refresh is doable, too, although that gets a bit trickier.