Oh no not this again. I am absolutely bloody sick of futzing around and faking it to make typewriter technology work in a half-assed way to simulate typesetting. Again. And I am even more frustrated by the enthusiasm I run into among some HTML authors. Isn't it wonderful, you don't have to plug in all the numbers in a list because the computer will do it for you? No, it's not wonderful, it's a huge leap backwards. The whole mess is a total rerun of what happened in the mid-1980s. I was working for a typesetter then, as a proofreader and customer rep with some limited knowledge of digital typesetting codes; we all loved the Mac on sight but we wouldn't dream of using it for proper work. I remember being seriously baffled by these supposedly wonderful new things called style sheets, until I figured out that they were limited versions of what we used all the time, only we called them formats. And here they are again, only now in HTML they are called Cascading Style Sheets, like this is something new and wonderful. I don't expect these people to have done proper typesetting, but haven't they at least used Microsoft Word? I remember too how thrilled the desktop pioneers were because they were able, for example, to expand and condense fonts – that is, to ruin them – and how we felt we were carrying the torch of typographic history as the barbarians rampaged through the streets. But it all settled down. The good folks at Adobe, especially, and Apple and Quark and even that gated campus up in Redmond, and Canon and Sony and HP Sauce, they all went to work and produced soft- and firm- and hardware that really does produce type, for all but the most specialized applications (don't get me talking about mathematics). As a moderately proficient Mac user, I don't have to think about this stuff much anymore, and thank goodness for that, except when I run into odd Czech characters and the like, and I sleep soundly knowing that people much more expert (and interested) than I are well on the way to solving that problem too, with Unicode; and I don't want to know about that, I assure you, I just want to be able to use it as needed. And now along comes HTML, and it's d- ampersand- eacute- semicolon- ja bloody vu all over again and by the way we're not quite sure that all of your readers are going to be able to see it the way you intend. Here's a neat trick: You can use a single-pixel transparent GIF and adjust its width &/or height to simulate indents and space and so on. Well, whoop-de-do. What I want is to be able to specify a paragraph indent and forget about it. The single-pixel GIF is a clever workaround to an indefensible level of crudeness. I don't really mind that HTML is crude; there will be decent authoring software before long (PageMill 4.0, perhaps). What I mind is people promoting it as if it was a usable tool for writing and presenting web pages; it's much more like a second-tier operating system. But then, some UNIX freaks have been putting over this same shit for years. They acted, and for all I know still act, as if Nroff and Troff and Pic were adequate tools, when they simply are not. As late as 1990, I was stuck with using them on a contract job; my boss actually showed me, with great pride, a picture he had created with Pic, all x's and o's and vertical bars and underline characters. Well, gosh, a tap-dancing elephant, how interesting. As an operating system, UNIX shows no sign of going away, nor should it; but as user software, its built-in tools have been obsolete for years. It takes a peculiar kind of blinkered arrogance to assert otherwise. Actually, not peculiar. Bloody typical is more like it. Every bloody élite despises their so-called inferiors. And this stupid hierarchical food chain goes right on down: Doctors despise nurses, and nurses despise janitors. Try keeping a hospital sanitary without the janitor. Do you really think that a doctor knows more about the effectiveness of specific cleaning materials than a good janitor? Ah, says the average doctor, but the janitors are stupid. No, you idiot, you just treat them like shit so no one with any sense is going to put an ounce of gumption into janitorial work. Publishing's no different: Commissioning editors look down on production editors, who take it out on typesetters. I've watched this one from all three vantage points, and been as much amused as annoyed by noticing how I personally have confounded some of the classist assumptions. As an educated Englishman (Oxford, no less), I get class privilege handed to me; when I was in the role of typesetter's customer rep, it confused the hell out of some of the dumber editors – was I a peer, or a peon? Of course, they wouldn't admit to any such thing; but it was obvious. For the record, some of the smartest, best-informed and most competent individuals I have worked with have been typesetters, wine bottlers, short-order cooks and practitioners of various other working-class jobs; not all, certainly, but then I have worked with some dumb clucks in publishing offices and such-like bastions of the eddicated class too. Back to the computer scientists. Why oh why do these silicon heads think they can get by without incorporating five hundred years of practical research on legibility into the beginning of their specifications? Oh, they're not alone. Garrett Hardin, polymath, essayist (author of The Tragedy of the Commons) and general all-round egghead once wrote a book which he insisted be printed with brown rather than black ink; he was convinced that brown on white was more legible, and of course he knew more than the mere printers. Well, I saw the book (I proofread much of it at the typesetter, and we got a file copy), and if you are sitting with a bright white light shining over your shoulder directly onto the pages, he may be right; if you are in ordinary reading conditions, he's wrong. I haven't noticed a mass switchover to brown ink. (Shock! I just read on the web that someone has done a new study on the subject and proved that black ink on white pages is the most legible. So now it's official. Science triumphs again! Omigawd. How many more times am I going to be forced to read headlines like "Old Wives' Tale May Have Basis In Fact, Say Scientists"? Let us say it one more time, very very clearly so even scientists can understand: Trial and Error leads to valid conclusions. Just because the experimenters don't understand the theory does not invalidate their results; and the promulgation of an abstract rationale is not the same as a discovery.) So here we are again, reinventing the typographic wheel. It'll all work out in the end. It's too important not too. But wouldn't it have been easier if there had been experts involved in the first place? Dammit, it is important that type look right! And the history of the computer industry is littered with siliconhead knowitalls saying type doesn't matter and then a few years later having other people change the standards and/or invent workarounds precisely because type does matter. We've been through this movie before and that's what gets my goat.