18 ms·
The Group That Rules the Web
- jasonm23 12y agoTo be honest I could barely make it three paragraphs without wincing at this claptrap. It's overly broad and blatantly inaccurate.
- nnethercote 12y agoI read the whole thing and thought it was excellent. I didn't wince once, and I didn't see anything inaccurate.
- jasonm23 12y agoI wince when non-French persons render words such as "èlite" with an accented e. It's typical pretentious New Yorker twaddle. Especially when the elite they are referring to, held no such position of hatred, or features desired. The main issue with HTML was looseness of definition, which has plagued us ever since. Avoiding correctness to benefit those who cannot be bothered to simply follow a prescribed document structure was the greatest error. Editing in a browser page? Pfft, nonsense. Nobody of technical note, gave a fig about such things.
- ftrain 12y agoTim Berners-Lee advocated for in-browser editing in the early days of the web and saw it as a critical part of the platform. This is documented in his book "Weaving the Web."
- tagawa 12y agoI agree it's aimed at a broad audience (not a bad thing in this case) but could you give an example of what's inaccurate?
- jasonm23 12y agoI was being pretty harsh, I'll admit, a bad habit. I was just irked at the opening assertion that "élite hypertext thinkers" and "computer savvy" people despised it. Far from it, they were the only people on it.
- farnsworth 12y agoThat's actually true, at least according to Tim Berners-Lee's book. Hypertext was nothing new when the WWW came around and other systems were more powerful.
- jasonm23 12y agoThe drama attached to "despised" is not true however, and the features lacking, ie. Editing in a browser or "who has linked to my page" were not of overwhelming concern. The author is simply attributing features which seem personally important to them/journalists, as the most pressing concerns of the people making/using the web at the time (early 90s) I was on Usenet virtually every day keeping an eye on discussions on SGML vs HTML, when the web was in its infancy. Certainly consensus was not always smooth and non-controversial. But to say anyone (who wasn't simply trolling) despised HTML, let alone the "èlite", is utter nonsense.
- tagawa 12y agoThanks for taking the time to clarify.
- dgreensp 12y agoPaul Ford nails it, for a lay audience no less.
- qq66 12y agoThis is exceptionally clear writing that can give someone who doesn't know what a Web browser is an interesting glimpse into Web standards and why they matter. I hope that a business model for quality journalism can endure.
- walterbell 12y agoJournalism needs to be repositioned as "selling time" or "life extension". Good writing takes much longer to write than to read, not counting investigation, research or subject matter expertise, which in some cases may be a lifetime's work. When you pay for quality writing, you are saving (buying) time and avoiding paths that would waste your limited time. Money can't create new time, but it can fund journalism that saves your life-time.
- Vladipoteur 12y agoYou are right, moreover, time is not money, you cannot have it back!
- icebraining 12y agoYeap, hats off to author (Paul Ford).
- meowface 12y agoThe New Yorker has always been pretty good on this front. As other publications decline, their content quality has mostly remained as good as it's ever been.
- emodendroket 12y agoLike so many other parts of the economy journalism has been "hollowed out" with middlebrow publications like Time suffering the worst. "High-end" publications like the New Yorker and the Economist are actually doing well.
- Animats 12y agoYes, that's a good article. I really hoped that XHTML would win out. Have you ever looked at the HTML5 parsing rules? They codify acceptance of malformed comments and specify standard handling of bad HTML. There are pages of special cases needed just to build a tree from the tags. Check out Python's "html5parser" for the amount of effort required per input character to make this work. At least in XHTML, the tags balance.
- olau 12y agoFor XHTML to win, we'd have to lose our history. Yes, it would have been simpler in terms of parsing, but I'm glad practicality beat purity. Everytime I see a CDATA section in some old code, something inside me dies.
- cwmma 12y agoWhile xhtml was nicer to parse, xhtml only described how to render a valid page . How often are pages like that in the wild, even if the page itself is fine some user add on or script can put other stuff in, possibly invalid. HTML5 described is so complex because it it unlike XML and it's strict validity, HTML5 works with even the most mangled tag soup
- RyanMcGreal 12y agoForgiving HTML parsing is one of the great strengths of the web. http://quandyfactory.com/blog/39/the_virtue_of_forgiving_html_parsers http://quandyfactory.com/blog/39/the_virtue_of_forgiving_htm...
- innguest 12y agoThe main point of that article is: > They're missing the point. The virtue of forgiving parsers is that they vastly increase the pool of people able and willing to create web content. I find it very hard to believe more people are coding because the threshold of what is an error has been relaxed. Although I do admit 'On Error Resume Next' is what kept me motivated to learn programming in VB6. So I'd say it might be an advantage for getting more people hooked in programming webpages, but professionally the price we pay to have this easy ramp up for beginners is very high.
- joshAg 12y agodoes anyone have a nonpaywalled link? going via google didn't work like it does for nyt.
- emodendroket 12y agoI can see it fine despite not being logged in; maybe clear your New Yorker cookies.
- innguest 12y ago> With that complexity [featurism] came Balkanization. [...] Anarchy would ensue and photographers would complain and complain. As this Balkanization was beginning to happen [...] Can anyone remind me of how bad this actually was? One paragraph was not enough to jog my memory.