[roel_v]

"I suppose I'm just confused at your anger"

Oh I'm not angry, that's just my way of talking. You should meet me in person some time, it's fun, really ;)

"over the suggestion that you use p for paragraphs, actually add labels to forms, use em for emphasis, section for sections and h1-h6 for headings. Is it really that hard?"

Oh no it's not, and everybody should do it. That's not what the 'semantic web' or 'semantic html' is though. But if he's going to talk about grandiose subjects like 'semantic html', then the discussion is about that, and 'semantic html' is dead and buried. If he just wants to say 'use p/label/em/hx appropriately', then just say so, don't bury it under pseudo-philosophical waxing (but then again, his post wouldn't have made HN, so as it is, mission accomplished, right?)

[IanCal]

> That's not what the 'semantic web' or 'semantic html' is though

Semantic web != semantic html. Maybe that's the source of the problem here. The article says absolutely nothing about the semantic web.

Semantic html is absolutely about using tags correctly.

[roel_v]

"Semantic web != semantic html"

True, in the sense that "sustainability != solar power". They're different concepts that cannot be compared "by equality", to speak in those terms; but semantic html certainly is part of the semantic web.

Of course we can go around and redefine "semantic html" to mean "use <h1> instead of <div class="header1">, a heavily diluted interpretation of "semantic html" as it was conceived in the context of "the semantic web". And I cannot argue with the point that "semantic html" is being used that way sometimes, maybe even increasingly so. Now I have no interest in letting this discussion degenerate into linguistics, semiotics and descriptivism vs prescriptivism, but if we are to use that watered down definition, the whole article loses what could potentially have made it interesting and descends into 'duh' territory.

[IanCal]

> Of course we can go around and redefine "semantic html" to mean "use <h1> instead of <div class="header1">,

Redefine? I'm pretty sure it hasn't meant anything different. It's about semantically marking up a document.

> but if we are to use that watered down definition, the whole article loses what could potentially have made it interesting and descends into 'duh' territory.

But you assumed it was around that, then said that was dead. The other definition (the one wikipedia and everyone else seems to use) is something that lots of people still don't follow. Did you see the "new way of writing" last week? That had an article structure but didn't mark it up properly, screwing with the order when you removed CSS and made a big block of json invisible on the page.

In fact you yourself said:

> Just toss in a few divs, use simple tech like html and css to make it look good, send the invoice, go home

Can you see why that sounds exactly like the people who don't use things like <section>, <nav>, etc? The same people that won't bother putting in alt tags because 'nobody reads them'? In fact, XKCD is an example of buggering this simple thing up. HN is built out of tables, where replies are indented with a spacing gif. Basic semantic HTML is not commonplace.