Posts in the Tech Category

Imagine the ability to spread a bunch of sub-millimeter sensors around and then collect the data.  Now go read “Companies test prototype wireless-sensor nets” at EE Times Advanced Technology.  Then, assuming you haven’t, go read A Deepness In The Sky by Vernor Vinge.  It goes on rather longer than I might have liked, but it’s still full of interesting ideas, including the use of a “smart dust” sensor network.  It would seem that privacy as we know it really will be over.  David Brin touched on this topic in Earth, and apparently addresses it directly in The Transparent Society, which I haven’t read.  Is the only defense against technological invasion of privacy the ability to detect the invasion, and the ability to counter-invade?  I don’t see too many other options, frankly.  It might be possible to jam some forms of invasion privacy, but who could afford the gear to detect and defeat all possible forms of invasion?

Along similar lines, the more I hear about the things that can happen to IE/Win users, the happier I am about being a Macintosh user who works for Netscape.  The very idea that a Web browser can be taken over, and seriously mess up the operating system in the process, makes my eyes cross.  I’m starting to wonder how any company with the slightest shred of concern over security could possibly justify running IE/Win.  Here’s a scenario:  a high-level manager wanders past a site that does a drive-by download of a toolbar which then does its own download of a small program that quietly transfers all of the hard drive contents to another system.  Hey, were those your corporate secrets leaving the building just now?  Yeah, I think they passed a virulent data-eating virus on its way in.

Then again, if said company is also running Outlook, I suppose getting upset over virus infections became passé a long time ago.  In any case, the only defense against this sort of thing is the ability to find out that it’s happening and put a stop to it.  If there were a way to inflict similar damage on the original perpetrators, we’d all be mini-Cold War actors: don’t mess with my data and I won’t mess with yours.  In that kind of situation, how long would it take someone to decide a first strike was a good idea, and how much damage could they inflict?

• Stay in View, Please was published on Thursday, January 30th, 2003.
• It was assigned to the Browsers and Personal categories.
• There have been no replies.

I’ve watched word of the Web Standards Meetup work its way through various blogs like a slow conceptual pulse, so I may as well keep it going.  I seem to be the only one in my area signed up so far, and I’m slightly bummed that there are cool people like Nick and Molly signed up in cities far away from me.  Of course it was only after I signed up that I realized the meetup is the same night I’m presenting at a local community college, so I’d be incredibly late or in absentia anyway.  Of course, people interested in standards could come meet at the talk, and then we could all go to a nearby place afterwards.  It’s an idea.  It might even be a good one.  You be the judge.

I’d been toying with the idea of trying to get local Web folks to assemble for socialization and (human-relation) networking for a while now, but thanks to Meetup I don’t have to undertake the organizational effort.  That’s pretty cool.  It just makes sense someone named Eric would be involved in something that cool.  We’re everywhere, and everywhere we go is cool.  That’s the kind of cool we are.

Some time yesterday, someone asked how my new book was coming along.  I asked which one they meant.  It turned out they meant Eric Meyer on CSS, and that they’d spotted a mention of it on MozillaZine.  So I went over to see what they had to say, and discovered they were saying that evolt had posted an interview with me last Saturday.  Well, shoot the horse and slap me silly!  I’m always the last to know.

Actually, said interview was already available in Italian, and I forgot to mention that fact until now.  I sense a karmic balancing here.

So, Opera 7.  It’s out, and yes, it does wonderfully well on css/edge, so you can all stop e-mailing me now.  I’ll update the demo pages one of these days to say so, I promise.  Just not today.  Opera 7 still suffers from some disappointing CSS bugs, though.  One is on this page right here, assuming you’re using the default page stylesheet or one of its variants.  The entry dates should be appearing below the horizontal line at the top of each entry, not above the borders.  Also, on my Speaking page, the :first-line underlining of li children of ul#upcoming is being applied to more content than it should.  Neither is really tragic, but they are a touch annoying.  Opera 7 also has some problems with negative text-indent values on block-level links; it seems to be flipping the sign on the value, so that it’s positive.  But I could be wrong about that, since I haven’t invested a lot of time in detailed analysis of the behavior.

There have also been some changes that make OperaShow do odd things to some of the files available on my Speaking page.  It’s probably a case of my writing my projection-media CSS to cater to Opera 6 bugs, and Opera 7 having fixed said bugs.  I probably won’t get around to fixing up old talk files, so if you really want to see them as first written, keep a copy of Opera 6 around.  Hey, at least you can have multiple versions of Opera on your computer, just like most browsers.

These are just the things I’ve noticed from a little surfing around.  To a degree, I’m just picking nits, but I also wonder how many other “combinatorial” problems I’m going to encounter.  It’s one thing to pass a test suite, which is a set of controlled circumstances that is easily predictable.  Dealing with the wonderfully wacky ways authors like to combine bits of CSS in pursuit of a given effect is something else entirely.

Is Opera 7 better than Opera 6?  Yes.  Does it have a good CSS engine?  Yes.  Is it the best CSS engine I’ve seen?  No.  Close, but not quite.

On a completely different front, the interface on this medical detector is really cool, not to mention the technology itself is pretty nifty.  Give ’em about 15 years to wedge in more advanced scanners and extra features, and we’ll have tricorders after all.

• Increasing Circulation was published on Wednesday, January 29th, 2003.
• It was assigned to the CSS and Personal categories.
• There have been no replies.

Lists seem to be my topic-of-the-moment, but this is about a different kind of list.  It’s hard to believe this, but it was a year ago today that the mailing list css-discuss was launched.  By my slightly rough count, the list carried 17,455 messages to its subscribers in calendar year 2002, and has nearly reached 19,000 messages total.  Just over one thousand of those messages were sent in the first eight days alone, which sent a lot of early subscribers running away in horror, and caused me to wonder if I was going to be crushed by the flood.  Due to overloading problems, we had to switch servers twice before finally migrating to the current home, graciously provided to us by evolt UK.  The list averages about fifty messages per day, both at present and as a lifetime average,.  Even though we moved the list to a new server and required people to actively re-join the list at that time, it has nearly 1,500 subscribed accounts.

I’ve worked hard to keep it a practical list with a high degree of signal, and I’d like to think those efforts have paid off.  We have the occasional hiccup, and every now and again a thread gets out of hand, but overall I think the list has been a lot more worthy than not.

I’d like to extend my deepest thanks to John Allsopp of Western Civilisation, which provided the list its first home; to John Handelaar of Userfrenzy and evolt UK, who manages the server where the list now resides; and to all the members of the css-discuss community, who make it useful through their contributions, discussions, and continued respect for one another and the list itself.  No community can be truly great without active support from its members, and the members of that list are very active in making it the kind of place I’d hoped it would be.  My thanks to you, one and all.

• A List, A Year, A Look Back was published on Friday, January 24th, 2003.
• It was assigned to the CSS category.
• There have been no replies.

Douglas Bowman ruminates over the use of list elements (i.e., <ul> and <ol>) as the basis for navigation links, tabbed interfaces, weblogs, and just stuff in general.  Is it okay to use an unordered list to hold the lists that drive your site?  Should a weblog just be an enormous ordered list?  If you do those things, does the semantic meaning of the list change to the point that it’s no longer really a list?

Well, kell co-ink-e-dink!  Tonight’s talk at COMMUG deals almost entirely with ways to take lists and restyle them to get panels, tabs, flowchart-like structures… pretty much everything Doug was talking about.  I’d even been planning to talk a bit about the semantic joyride such approaches can mean, at least to some people.

So here’s the short version of what I think: looked at a certain way, pretty much everything can be represented as a list.  The U.S. Census, the Solar system, my family tree, a restaurant menu, the stuff I did yesterday, all the friends I’ve ever had and lost—these can all be represented as a list, or a list of lists.  So the question isn’t really whether we should be putting all this stuff into lists.  The question (at least at this stage of the game) is whether or not the markup structure meets the job, and helps with accessibility concerns.

Now, should we have markup structures that meet the jobs more closely?  Maybe.  XHTML 2.0 has the <nl> (nestednavigation list) structure, which comes closer to making the markup name match the structure’s intended role.  I’m not convinced this is necessary; it’s already possible to just take nested lists and turn them into menu systems using CSS, assuming a sufficiently capable user agent.  It’s still a topic worth consideration and exploration.

• Listing Toward the Future was published on Thursday, January 23rd, 2003.
• It was assigned to the CSS category.
• There have been no replies.

So last night Kat and I headed down to the Cinematheque to meet up with Ferrett, Gini, and Jeff to see Jesus Christ Vampire Hunter.  Wow!  It was… well, it was… I mean to say, it… it’s not really describable.  But it was quite funny.  I might pick it up on DVD if it ever comes available.  As Ferrett said on the way out, “Oh, I can’t wait for the commentary track for that one.”

Meantime, the recent ruling on Eldred v. Ashcroft sparked a lot of debate on a computer book authors list to which I belong.  I stayed out of it for a while, because I didn’t have much to say, and then suddenly—as is often my wont—I realized I had something to say after all.  So I said it, and I figured, what the heck, I could say it here too.  So if you want to know what I think about copyright terms, feel free to read away.  If not, no sweat.  It’s automatically copyrighted either way, as it happens, and now nobody else can say the same thing for more than a century, or something like that.  How much sense does that make?

I still don’t know why I think so, but this is darned cool.  I’m probably just jealous I didn’t think of it first.

• Oh, There’ll Be Plenty was published on Sunday, January 19th, 2003.
• It was assigned to the CSS, Personal, and Politics categories.
• There have been no replies.

I’m feeling better, thanks.  About most things, anyway.

If you’re seeing layout or other rendering bugs on this site in Safari, as some people have said they are, please use the bug icon in the browser to report the problem.  I can’t run Safari or else I’d report problems myself.  Apparently there are some weirdnesses with the navigation links in the sidebar, if nothing else.  Whatever problem you see, it’s worth reporting, so please do.

Most of you probably already know that Mark Pilgrim is upset with XHTML 2.0, and many of you may be aware that Tantek and Daniel Glazman are in agreement.  I’m broadly sympathetic with their frustrations, but since I was never that thrilled with XHTML in the first place, I can’t get too worked up about the breaks between 1.x and 2.0.  I never really got why HTML had to be reformulated as XML.  Yes, I’ve read all the arguments about later ease of conversion and all that.  I suppose there was some good in easing authors into XML authoring habits using a language they mostly recognized.  That just didn’t seem like enough.  This site has been, and continues to be, HTML 4.01 Transitional for a reason.

I do broadly agree that XHTML 2.0 is way too unrealistic for its own good.  It outright drops too many things authors find useful, like the style attribute (although I admit I’m biased there) and heading elements.  For that matter, yes, Virginia, there is a difference between abbr and acronym, so dropping either one seems like a mistake.  On the other hand, if this stuff was deprecated instead of eliminated, I’d have many fewer points of concern about XHTML 2.0.  I’d be worried that the deprecated stuff would be dropped in the next version of XHTML, but XHTML 2.0 would bother me less.

Then again, given that you can take XML and CSS and create your own documents out of whatever markup language you can invent, and use XSLT to bridge the gap between old browsers and new ones, I find XHTML to be of minor import.  If it gets too ivory, then it will be ignored, and some other XML-based language will take it place.  Or, more likely, lots of markup languages.  Either way it will be interesting, and the XHTML 2.0 advocates won’t be able to blame anyone else for the explosion of non-interoperable languages.  Which, I suppose, is the point of all the sturm und drang of late.  If XHTML 2.0 were interoperable with XHTML 1.1, people wouldn’t be nearly so upset.

Wow… all this concern over making things work together.  Can it be that the Web is getting all growed up?

• Agony and Ivory was published on Tuesday, January 14th, 2003.
• It was assigned to the Browsers, Standards, and Web categories.
• There have been no replies.

Simon Jessey has confessed he wrote the Amazon review I mentioned on Thursday, and furthermore says he pictures me mostly as Li Mu Bai with a little Jen Yu thrown in.  Hmmmm… that’s definitely an interesting image.  Anyway, I’ve created a new presentation option for the site to celebrate being called “The Li Mu Bai of Cascading Style Sheets”: please enjoy wo hu cang long.  Note that this new theme has a layout bug in IE5/Mac which appears to be related to the alternate-style switch, and not the CSS itself.  There isn’t much I can do about it, as the bug doesn’t happen in static test documents.

On the other hand, Robert Kirkpatrick wrote in to advise me that I should work to be the Cheng Pei-pei (who played the Jade Fox in Crouching Tiger, Hidden Dragon) of CSS.  Of course, we’re mixing actual people and characters here—in the movie, of course, Li Mu Bai is the more skilled, but in real life it’s likely Cheng is more skilled than Chow.  Either way, I’m flattered.

I’m also feeling under the weather, which is ironic given how nice a day it is today.  Off to the couch for tea with lemon and maybe a movie.  The Killer is always a good choice.

• Green Destiny was published on Sunday, January 12th, 2003.
• It was assigned to the Books, CSS, and Writing categories.
• There have been no replies.

In response to my rantings yesterday, David Hyatt has stated unequivocally that the Safari team did not, in fact, co-opt Netscape evangelism efforts during development.  I’m really very glad to hear that’s the case, and if I hadn’t had such a bad day Tuesday, I probably wouldn’t have mentioned the rumor in the first place.  Then again, the end result of my ranting is a negative rumor laid to rest, so perhaps it was all for the best.  That’s what I’ll tell myself to feel better about the whole situation, anyway.

To make it formal: I apologize for casting any unwarranted aspersions on the Safari team, Apple, etc.  With any luck this will help stamp out the rumors that were reaching me.

On to more trivial matters!  This is quite possibly the coolest review I’ve yet received:

Last year, I watched “Crouching Tiger, Hidden Dragon” and I was amazed at the swordsmanship on display. Swords were no longer weapons, but extensions of arms – as if they were new appendages grown especially for the task. Eric Meyer can wield CSS (Cascading Style Sheets) in just the same way as those actors could wield swords.

Being a big fan of the movie, I can’t help but be deeply flattered.  I’m just wondering if said reader pictures me as Li Mu Bai, Yu Shu Lien, or Jen Yu.

• Lookin’ Up was published on Thursday, January 9th, 2003.
• It was assigned to the Books, Browsers, Web, and Writing categories.
• There have been no replies.