XMFUCKYOU

[brentdax]

I don't like XML.

No, wait, let me restate that. I hate XML.

No, wait, let me restate that. I despise the over-engineered, over-verbose, over-committeed (that isn't a typo), over-hyped, and under-powered pile of steaming horse shit that is XML with every cell in my body.

And I dearly wish there was a YAML parser for C#.

Comments

[izuko]

I think I started a protest against it about seven or eight years ago. Called it KISS HTML (Keep it Simple Stupid). Never got anywhere.

Let me also say that I hate the W3C's vision for HTML 4.0 and beyond. Yes, I know style sheets are very powerful, and very flexible. But those depriciated tags were easy to use, and allowed those of us who are all program languaged out to write clean HTML instead of resorting to the bastard code of BluntPage.

By trying to force the ideological goal of separating style from content, they've made HTML less accessable and less useable. Can you imagine what would happen if Toyota made cars -more- difficult to drive?

[countalpicola]

> But those depriciated tags were easy to use, and allowed those of us who are all program languaged out to write clean HTML instead of resorting to the bastard code of BluntPage.

That, in part, is what the Transitional DTDs are all about. HTML 4.0 Trans and XHTML 1.0 Trans have a bunch of deprecated tags, but they've only removed the most useless of tags from the earlier specs. Now, sure, Strict is a different animal, but nothing says you have to write in Strict. Indeed, nothing says you have to write in HTML 4.0 or XHTML. HTML 1.0, 2.0, 3.2 all have valid DTDs and you can, if you really want to, code to them.

Personally, all I use anymore is XHTML 1.0 Strict, but that's unusual even for people who do web design professionally. There's nothing wrong with using older specs. Just be sure to set the DOCTYPE tag at the top accordingly.

[chiave_trust]

Then there's me, who wrote in HTML 4.0 Strict before I converted it to XHTML Strict (which basically consisted of me changing the DOCTYPE tag and one other tag in the entire site).

It's clean, but you don't have to use it. I just happened to be using CSS as well, so seperating style from content worked alright in that case. (And I'm a bit of a masochist, apparently, in writing Strict when I didn't have to at all, on the first webpage I've written in a couple years.)

[izuko]

I'll stick with Transitional until browses no longer support it (which, given the crappy code written by page writers, will be forever). Still, I do have to be just a little incensed that they're moving a useful tool further away from usability.

[codepoetica]

XHTML1.0 Trans for me.

For pushing data around, however, which I think is the point of Brent's post, I abhor importing XML. Parsing the soup is just an absolute pain in the ass, driving me to use

• url=encoded&data=strings
• variable=assignment\n
• http: headers\n
• ( {"json": "strings"} )

Although json is a pain in the ass to debug sometimes. I have absolutely no problems with exporting my data in an XML document of some form, but in general avoid accepting data in that form. Especially if it's being passed between two of my own programs. If I need to accept XML data in a short one-off script, I'll regex it out of the XML document without care of the structure, simply because it takes much less time.