Thoughts From Eric Archive

Actually, I shamelessly used that title simply because it’s a little play on words.  By and large, my impressions of Mix 06 and what I’ve seen here are positive.  This isn’t my last word on everything going on here, but I wanted to share.  Enjoy!

• You can drag-rearrange tabs in Firefox just by click-and-dragging the tabs.  Seriously, I had no idea.  Thanks to Dan Short for setting me straight on that score.

• In his keynote, Bill Gates said “we need microformats”, which I didn’t even know was on his radar.  For more about that, head on over to microformats.org.

• Microsoft is coming out with a new Windows-only Web design tool called Expression.  It’s pretty slick, with features like visually illustrating margins and padding in the design view and what seemed like smart management of styles.  Unfortunately, I had a little trouble following what it was doing, mostly because I saw it presented in a talk and didn’t have hands-on time.

  Basically, Expression seems to be FrontPage done right, with a relentless focus on standards-oriented design principles. It has its own rendering engine for the design view, and the whole thing was built from the ground up, which means it isn’t trapped by legacy rendering concerns, but it made several of us wonder why that isn’t what they use in IE7.

  I also had trouble mentally distinguishing it from other visual Web design environments like Dreamweaver, but that’s probably because I don’t use a visual design environment.  BBEdit 4-evah, baby!

  Speaking of which, there are no plans to port Expression to the Mac.  Whether that’s good or bad probably  depends on your worldview.  Look for public betas of Expression somewhere in the June 2006 time frame.

• It was publicly stated that the current build of the IE7 beta available from Microsoft is rendering-behavior complete.  In other words, the only changes to IE7 from now until it goes final will be fixes to security holes, crash bugs, and browser chrome/UI stuff.  Whatever its CSS support does or doesn’t do, that’s how the final version is expected to behave.

  Ladies and gentlemen, start your engines.

I’ll take a few minutes on that last point.  A little while ago, I said that designers should remain calm and not hack their sites to fix them in the IE7 beta because it was a moving target.  That is no longer the case.  It’s now time to start testing sites in the IE7 beta and identifying any layout problems that may occur.  (And there will be problems.  No browser is perfect.)

I’ll be doing this as soon as I can, and I encourage everyone who can to do the same.  Here’s the other key point: IE7 is scheduled to go final in the second half of 2006 (I couldn’t get anything more specific), so we have a calm period of at least three months in which to find out how things stand before IE7 goes final.  This isn’t an accidental circumstance, either.  The IE team has deliberately done this in order to give Web developers time to figure out what’s coming and how to deal with it.

This is entirely in keeping with the new spirit of the IE team, which has impressed me again and again at this conference.  Once upon a time, upgrades to standards support were blocked by the cry “We have customers!”, which was maddening both because it impeded progress and because it was true, as I wrote back in 1998.  The usual counter-argument was that Web designers and developers are customers, too.  We just weren’t (often) treated that way.

Now we are considered customers of the IE team—not the only ones, but important ones.  Not every decision will go our way (even if we had a single “way”, which of course we don’t) but our needs and concerns will be considered.  As further proof besides the “grace period” built into the IE7 timeline, the IE team is creating tools and resources meant to make it easier to update sites for IE7.

I’ll have a good deal more to say about all this in the near future, but those are the big points in my head right now.  I expect to hear Dave‘s, Andy‘s, and Molly‘s takes on all this, and hopefully others will add their thoughts as well.

• Mixed Impressions was published on Tuesday, March 21st, 2006.
• It was assigned to the Browsers, Microformats, and Standards categories.
• There have been fifty-three replies.

There’s been much talk of this year’s SXSW and how overwhelmingly huge it was.  I don’t have a whole lot to add to that, really.  I thought last year was out of control.  This year left it standing.

Without question, the best panel I saw was “How to Convince Your Company to Embrace Standards“.  This was not due to the topic, though that was good too, but because the panel was tightly assembled and packed with good information.  Most panels are a collection of folks who sit onstage and leisurely toss out assorted thoughts for an hour; I should know, having been on many such panels in the past.  For this one, everybody had specific points to make and made them concisely.  There was a lot of preparation, and it showed.  It very much raised the bar, as far as I’m concerned, especially since I’m thinking of proposing a panel or two for 2007.  Kudos to all involved.

Also:

• It was interesting to sit on the “How to Roll Your Own Web Conference” panel with Jason Fried and hear his experiences.  You may recall that when I wrote about event pricing, I said one way to find an event’s optimum price was to run it over and over and keep raising the price until you stopped selling all the seats.  That’s exactly what’s happening with the “Getting Real” workshop.  It will be interesting to see where they level off.  Assuming they do.

• John Allsopp‘s presentation during the WaSP Annual Meeting was an interesting experience for me.  It also covered a bit of the same ground I plan to cover in my keynote for @media.

• In two different lunches, I told people from computer book publishers that their whole business model is in danger of collapsing.  Interestingly, both (more or less) agreed.  Sadly, it seems that only one is working for a company that’s aware of this fact, and it isn’t the company you’d probably assume.

• I’ve decided I much prefer El Sol y La Luna to Las Manitas when it comes to the food.  Las Manitas, of course, wins on the basis of proximity.  Also for not deafening its patio patrons.

• No joke: I got into our rental car in Austin and the Avis Preferred hangtag said “ETA: CSS“.

Good times.

• SXSW Summary was published on Saturday, March 18th, 2006.
• It was assigned to the Speaking category.
• There have been six replies.

After a short evening walk, Tantek and Kat and Carolyn and I arrived at 219 West in Austin, Texas for the WestCiv gathering.  The crowd inside was quite loud and densely packed, saturated with so many Web geeks that it was threatening to precipitate a site right onto the carpet.

Just as we got close to some people we knew, a staff member appeared at our elbows.  “I’m very sorry, folks, but this is a 21-and-over establishment,” he said with a distinct lack of sorrow.

So if you wondered where we were, or happened to spot us in the 90 second window before we left, or I walked past you without interacting and never got back to you… that’s why.  We weren’t blowing you off; we just ran afoul of an odd local custom.

As a result, we’ll be absent from the vast majority of the evening gatherings at SXSW.  Hope all you kids have fun.  We’ll be hanging out with the other grown-ups and getting some sleep.

• No Kidding was published on Saturday, March 11th, 2006.
• It was assigned to the Speaking category.
• There have been eleven replies.

So tomorrow I head out to SXSW along with most of the rest of the industry, just like everybody else.  There are, as usual, about two dozen sessions I want to see, all of which conflict with each other.  I’ll be on three panels, two Sunday and one Monday (as listed over at Complex Spiral), and will be doing book signings on Sunday around lunch time.  There will be a bookstore there, but if you already have a book of mine, bring that too.

Only a few days after I return from SXSW, I’ll be shipping out again for MIX06 in Las Vegas.  They’ve been running a “Remix MIX” design competition in the spirit of the CSS Zen Garden, and I’ve consented to be one of the judges.  I’m actually looking forward to MIX for a whole bunch of reasons, but at the top of the heap has to be a chance to try out IE7 and talk to the team members in person.  That’s three-quarters of the reason I’m going.  Also, I’m curious about Microsoft’s “Atlas” framework for AJAX development, but that’s more of a bonus reason.

Besides, if you look deeply enough, you discover there are really only two “scenarios” (a.k.a. tracks) at the conference:

• Next Generation Browsing Experience
• Beyond the Browser

Yeah, I think I’d like to know what they’re thinking.  So off to Vegas I go, once I’m back from Texas.

Yee haw.

• Southwest Twice was published on Thursday, March 9th, 2006.
• It was assigned to the Speaking and Travel categories.
• There have been no replies.

I don’t think I can say this without sounding smug, so I’ll just say it: this is what I was talking about.  If you went ahead and tried to hack your site so it worked in the previous beta, I’m sorry, but I tried to warn you.

It’s also why I said CSS hacks weren’t necessarily dead yet, or even likely to cause real problems, because there’s every chance that IE7 will be close to being another Firefox (in the standards-support and layout-behavior senses).  We can’t be sure of that yet, of course, but the results described by Molly are pointing in that direction.

Sure, we’d like to see a hack-free Web, but that point will not come until a few years after IE7 finally ships.  No, that’s wrong: we’ll never have a hack-free Web.  But we might reach a time where cross-browser presentation has become not only commonplace, but subconsciously assumed, like our current expectation that a browser will know how to handle hyperlinks.

• IE7 Revs Up was published on Saturday, March 4th, 2006.
• It was assigned to the Browsers category.
• There have been fifteen replies.

Once was lost, now is found: “Images, Tables, and Mysterious Gaps” has been resurrected from the Great Bit Bucket Beyond and given new life on Mozilla.org.  In fact, it looks like just about all the technical articles written by me and the other members of TEDS are available.  Look through the full list of CSS articles, for example.  You can dig into any number of topic areas from the main page of the Documentation section.  (Scroll down to the “Mozilla Developer Center Contents” headline.)

Some other popular articles from my Netscape days gone by:

• Consistent List Indentation
• Defining Cross-Browser Tooltips
• Issues Arising From Arbitrary-Element Hover
• Understanding Underlines
• Liberty! Equality! Validity!

So far as I’ve been able to determine, some of the less technical pieces, like the interviews with Doug Bowman and Mike Davidson, are not available.  Not now, anyway.  Perhaps one day that too will change.

• DevEdge Content Returns was published on Thursday, March 2nd, 2006.
• It was assigned to the History, Standards, and Writing categories.
• There have been six replies.

It only took most of a year for this to happen, but WP-Gatekeeper 1.5 RC5 is now available.  The only change is that it will now auto-add the challenge to any standard WordPress 1.5 install from the moment you activate the plugin.  Before now, this auto-insertion wasn’t working on any WordPress install that had gzipping turned on, as many do.  A heap of thanks to Jeremy Dunck, who first identified the problem; and Andy Skelton, who showed me how to solve it.

For those who joined the party in the long silence since RC4, Gatekeeper is a WordPress plugin that lets you manage a series of challenge/response pairs.  The default challenge is “What color is an orange?” (correct response: “orange”), though you should definitely disable that one and add your own.  This helps stymie spambots, though of course it is easily defeated by a manual spammer—and they do exist—and it can do nothing to stop trackback spam.  I actually stopped using Gatekeeper on meyerweb when I installed Akismet, which may be good enough for most people.  For those who can’t or won’t run Akismet, though, Gatekeeper is a decent alternative.

Gatekeeper is technically a CAPTCHA, but it is a fully accessible CAPTCHA, as it uses no images.  It’s also highly configurable, allowing you to add as many challenges as you like and then rotating between them randomly.  I know of a few sites that are quite happy with Gatekeeper, and recently caught wind of a Django implementation of the same concept.

So it’s there and ready for use by those who are interested.  If I haven’t heard about any bugs within the next month or so, I’ll strip off the RC designation and go with 1.5 final.  And about time, too.

Note to WordPress 2.0.x users: I have no idea if WP-Gatekeeper 1.5 will work in WP2.  It may.  Then again, it may not.  I’d be interested to know either way.

• Gatekeeper 1.5 rc5 was published on Wednesday, March 1st, 2006.
• It was assigned to the Tools and WordPress categories.
• There have been five replies.

The alpha 2 release of S5 1.2 is now available (177KB ZIP file; also available for previewing in the testbed).  There isn’t any major change here, but I did add some notable enhancements to the notes window.  These are:

• On any slide with incrementals, an indicator of incremental progress will appear in square brackets next to the overall slide show progress on the title line.  It’s a little crude, I admit, but it serves the purpose well enough.
• Clicking on the title of either the Elapsed Time or Remaining Time counters will “minimize” them.  Click a minimized title to maximize the box.  The actual minimum and maximum effects are purely CSS-driven, hooked onto a collapsed class name.  I’m still pondering the best way to handle this feature, so the class name may change, as for that matter may the mechanism by which one can min/max the boxes.  Suggestions are welcome.
• Keypresses and clicks are passed from the note window back to the slide show.  In other words, the slide show fully is fully operable from either the slide show window or the notes window.  The only difference is that the notes window doesn’t have the navigation links and popup navigation menu (said difference to disappear in a future release).

That’s it.  In the process, though, I uncovered a bug that shows up in Safari 1.3.1 and 2.0, where it’s ignoring the show-first feature for incrementals.  I’m going to assume that the problem lies in the getIncrementals() function, though of course I could be wrong.  If anyone can spot the error and provide a fix, I’d be grateful.

Update 2 Mar 06: in addition to the Safari problem, I’ve discovered that IE/Win doesn’t seem to share event information between windows.  Thus, if you try to run the slide show from the notes window, errors get thrown.  I managed to fix this in clicker() by adding a test for notes-window events, but trap() has a very different structure and I’m not sure how to fix it.  Thus the testbed in IE/Win currently lets you advance the slide show from the notes window by clicking the mouse button, but keyboard navigation throws an error.  If anyone can tell me how to get around this, even with a pointer to a good article on passing events from one window to another, I’d be very grateful.

• S5 1.2a2 was published on Monday, February 27th, 2006.
• It was assigned to the S5 and Tools categories.
• There have been fifteen replies.