First, the duck/boo interruptions: I’d like to know if anyone else experiences them. It’s odd that shat would not have the same effect; he uses the same code. I’ll look and see what I can find.

HTML is less of a language than pdf or Microsoft .doc – those document formats can contain logic. TeX contains a full-blown language. If you wish to put languageness on a spectrum, HTML would still fall way over on the ‘not’ side.

The inconsistency between spaces and line breaks isn’t about having tags and escaped commands, it’s that two special formatting characters are treated completely differently. They fill the same role; they should work the same. (Neither should be tags, which perform a different role.)

‘Separate data and presentation’ has been a mantra in computer science at least since the invention of the GUI. TeX is not great about that, but you could go nuts and accomplish it (LaTeX makes that a lot easier). TeX, however, is so much about creating one specific sub-pixel perfect layout of a document that exists only to be laid out, that the versatility is less important.

I could have sworn Berners-Lee was a Physicist as well. He was certainly influential, but that influence wasn’t all good, and really not all that amazing. As I said before, there were other people solving the same problems; he just happened to be in the right place – next to a big, widespread network with lots of peers who were pining for something like this. (Actually, I think his real influence came later as one of the people maintaining the standards.) Influential, yes. Visionary, no.

The flaws in HTML were painfully obvious to anyone who wanted to do more than slap text on the screen and add a few pictures (that the text did not flow around). I took a look at HTML back then and thanked my lucky stars that I was not going to have to use such a clunker. Oh, how naïve I was.

HTML 2.0 made the format somewhat more useful, and things have progressed. One-third of the tags in the original HTML have since been removed, though sadly not line break or horizontal rule. Somewhere early on tables and forms were added, making the format suck a little less. And that’s what its been for two decades, a gradual easing of the suckage, at the expense of the one shining attribute HTML had to offer: universal accessibility.

As far as already knowing better, look no farther than SGML, which was a generalized, conceptual description of document markup that HTML sorta-kinda followed, except when it didn’t, which was a lot of the time. The intervening years have been a gradual drift towards SGML. We’re still not there. Berners-Lee, I think, was somewhat in denial about the gulf between HTML and SGML. Near the start would have been a great time for the geeks making Mosaic to step forward and offer guidance.

To be fair to Berners-Lee and his cohorts, they made something that was adequate for a specific, very limited purpose – sharing static text information among peers. Aesthetics were not considered to be very important, nor flexibility. It’s when the world jumped on HTML and started wanting to use it for everything else that the flaws became relevant.

Since then DOM has been added, providing a way to describe all the pieces of a document (what they should have started with), which made CSS possible, and now with JavaScript ubiquitous and consistent it is finally possible to request content for and update only one part of the page, though you also need custom stuff on the server to handle those requests. The Google search over on the sidebar uses AJAX, for instance.

The cache is a complex mechanism that works differently in different browsers, which was invented as yet another stretch of duct tape to fix HTML’s flaws. Granted, caching would have been useful anyway, but even more efficient would be to not even ask for the same content again. And while images go into the cache, all the text on the sidebar has to be reloaded (since it’s in the code for the main page), and re-rendered, and the scripts re-executed. Cache was only a partial solution to something that the HTML guys should have addressed long ago.

Your hacker example made me think of this specific story that illustrates your point. The ‘classic’ mac operating system was amazingly secure, in part because way, way, down inside the original engineers decided on a weird way to invoke operating system functions. It was a pain in the patoot to work with, but for reasons I’m not totally clear on, made cracking the mac nigh-on impossible. Now, of course, Mac uses an operating system which at its core has stuff to make it easier for one computer to control another. Usually it works the other way, however. Look at Microsoft Windows a few years ago. They kept adding stuff that was useful for large businesses, but also turned out to be useful to hackers.

That notwithstanding, it’s more the underlying stuff beneath HTML (HTTP and so forth), built in an environment of happy collaboration, that has attracted the malicious no-goodniks with their hacking and their denial-of-service attacks and whatnot, so that point can probably be set aside.

I suppose I should tack a conclusion on to all of this. HTML changed the Internet, put the power to publish into everyone’s hands, and was a poorly-designed piece of crap that could have been done much better, to the point where fixing all its problems has undermined its original strength.

First of all, isn’t Berners-Lee a high energy Physicist? (I couuuuuld look up on wikipedia). And wasn’t he one of the few scientists named as most influential people of 20th century in Time magazine or something? In a sea of celebs and politicians. Scientists never git nothin, and now you want to take this away? Your dad is a high energy physicist. I think a couple of kittens expired when you wrote your words, and a russian grandmother tore out her own eyes. You, SIR! are a father-hatin, kitten-killin, russian-grammy-blindin provocateur! And iconoclast. Probably a sociopath as well. Since ad-hominem attacks are the pinnacle of rhetoric and debate I could rest my case here, while you gasp for breath like a mewling, scabies-ridden puppy. But I will press on.

First of all (you don’t merit any second, third or more of alls you iconoclast!), I should clarify that my only reason to invoke TeX…
(Gaw! %##@&!
Both the duck and mister boo are completely shutting down my erudition. Are you doing this on purpose because you are a mewling puppy afraid of real debate?!? I’m running firefox 3.0.10 on linux. What seems to be happening is they are taking my cursor out of the editor window. Thus I have to re-click at end of last typing to insert myself back into comment window. Oh and I see no special formatting of TeX.)
…is to say that they wanted to have process and results do exactly what they want and not to invoke it as the model for formatting web documents. Well, maybe a little.

Okay, wiki says Berners-lee is a computer scientist, not a physicist. I was thrown because he worked at CERN. So you’re not a father-hater, but those other things, …MAN!

Stepping thru your points, in order, the first thing we come to is that HTML is not a language. And you imply they are stupid because they didn’t know the difference. Knight of the British Empire – stupid! Well, what do you mean by a language? A programming language? Are you saying that you want data put into a program, the crank turned, and results output? This strikes me as semantics. Nobody really takes piss home in their pocket when they go to take a piss, but we all know what is meant. HTML is a set of specific, technical, jargonations that allow us to speak of the pompatus of web. It is a language.

Your next point is the inconsistency of having both tags and ampersand escaped commands. This is both a trivial and major point. OToneH it is hardly the straw to break the camel’s back, but on the other hand – if it is worth doing, it is worth doing right, and it is esthetically not very good. It strikes me as the different approaches that science and engineering take. The stereotypical scientist is organic and wild-haired and pushes the boundaries forward. The stereotypical engineer is buttondown, precise and makes it all elegant. Since wiki says Berner-Lee is an engineer, it is a surprising lapse.

Your next point is having the content and the instructions on how to display the content all mixed together. Now this is an interesting point, and worth lots of friendly discussion over a fine brew. A smoky porter, perhaps. Or a pumpkin stout. Now tex does exactly this kind of mixing, and it harkens back to my point of “I’m gonna display something, and I don’t want Bill Gates telling me how he thinks the results should look.” But your point is stimulating, and I can’t really address it without knowing more. What would you propose? If you wanted to serve people something to read and/or hear how would you separate content from instructions? A fascinating concept, I sincerely look forward to your thoughts.

Your next point I also can’t address so much, because I would like/need more detail. You say that when the web and attendant HTML exploded, the flaws became apparent immediately. How so? Were there technoGeek conferences on the flaws? Even at the early stage?
Some flaws you mention: Different browsers rendered pages differently. Well, why did they? Why aren’t they (instead of HTML) the asshats?; formatting options extremely limited. How so? What did you miss? What did you wish to format that you couldn’t?; Unscrupulous people…what vulnerabilities were there? Finally what are the lessons from other branches of computing that would have helped these flaws? I dunno the answers, that’s why I’m asking.

Your next argument is that the HTTP is reloading all of the static stuff over and over again. Such as the headers, and sidebar. Really? I didn’t realize that. I thought that was what cache was all about. Nonetheless this is a really good point. In a world where AT&T and others want to start chargin for bandwidth, cutting the fat and inefficiency is really important. And I’m with you – if I had been an early HTTP pioneer, I would have certainly programmed the protocol not to waste effort on the static stuff. What were they thinking. I can agree on this point. A serious lapse. If – as you suggest – it currently takes complex coding to sit on top of the CSS that sits on top of the HTML, then it is a true tragedy.

First of all, I’d also add, that somewhere you complain about the organicness of the development. Your point is well taken, but there can also be a beauty in organicism and I wouldn’t dismiss it lightly. It is entirely possible that once upon a time an unscrupulous hacker pulled his hair out because he couldn’t do evil thing X since the organic web was doing a certain convoluted, inefficient method Y.

Finally, first of all, this comment was sooo painful. The duck/boo really was agony. If the above was hyperbolic ranting, there is no hyperbole here: It took me an extra fifteen minutes to rant because of the interruption. Oddly, Shatty Baby didn’t cause problems.

Haven’t seen Mr. Boo yet, and the duck and Shatner seem to be showing up less often, although that may be a matter of perception.

Opera and Safari both treat svg like any other image.

Microsoft proposed a different format for the same purposes, and has never joined the svg parade.