log_e.hixie.ch

2004-03-28 20:33 UTC Ramblings from the North

I'm at the back of a dark room sitting next to an air vent listening to Let Me Entertain You being played by a big band on a stage 20m in front of me.

Look me up in the yellow pages
I will be your rock of ages
Your see-through fads and your crazy phases, yeah
Little Bo Peep has lost his sheep...

This dark room is far underground for a shopping mall which has been rented away for the night for the purposes of entertaining 4000 people. Actually the mall is only half the party, they are also using the concert hall across the street.

The 4000 people are all musicians, members of Norwegian concert bands who are here for the annual concert band competition. I'm here because I'm a member of one of those bands, the Romerike Konsertorkester.

"Here" is in Trondheim, the furthest north I've ever been as far as I know. This town has the prettiness of Bergen but the weather of Oslo. In fact the weather is probably even worse than Oslo. It was sunny yesterday (if ridiculously cold), but today someone aimed an incessant stream of crystallized water pellets at the town for no apparent reason, and it became all slippery, white, and wet.

The band I'm in came fourth out of about twelve in our division. We're in division 1. (It's a zero-based index. There is one division above us, the Elite division, and three divisions below us.) I've never been here before so I really have no point of reference, but based on the looks of extreme joy on my fellow band members, I'm guessing this is extremely good. (We were apparently hoping to be lucky and place eighth or ninth.)

I'm currently singing along quite loudly to the music being played on stage. It's drawing looks. It has to be said, I'm the only person who has a laptop here. They are all enjoying themselves with alcohol and tobacco, much to my disgust, but hey, to each his own.

Before sitting down here I tried to get myself a drink.

Me: Har du eplemost?

Bartender: ...?

Me: Har du eplemost?

Bartender: ...!

Me: EPLEMOST?!

Bartender: ...!!

They didn't have any apple juice.

Me: Vann?

Bartender: ...?

Me: Vann!

Bartender: ...?!

Me: VANN!!

Bartender: ...!!

They didn't have any water either.

Eventually I found a bar that did have water, about 100m from where I'm sitting now. I came back to enjoy the Jazz with my glass; by the time I'd gotten here I'd drunk half of it already.

I'm not going for more water for a while. Trying to push myself through the increasingly drunk people isn't too fun.

The age range here is quite dramatic. There are people from early teens all the way to retirement age. Sadly the smoker population spans the same age range. There are also quite a few cute girls, although many are smokers too, and none of them are the right cute girl.

Much to my amusement somebody just asked me if what I was doing (using a computer) had anything to do with what was going on around me. I went for "no". How anyone can communicate with the level of noise in here I have no idea. There's no wireless network, so the obvious answer (an IRC network) is right out. It certainly isn't vocally. I can't hear a word of what people say even when they shout in my ear. I've often wondered this about events of this nature.

As is probably obvious, I don't go to these kinds of these parties very often. My kind of party is usually very quiet, with very little sound except the hum of several dozen fans and the occasional expletive from an indignant party-goer. (And at that kind of party, IRC usually is indeed the preferred communication method.)

Young man, there's no need to feel down.
I said, young man, pick yourself off the ground.
I said, young man, 'cause you're in a new town,
there's no need to be unhappy.

Everybody just stuck their arms in the air simultaneously, then folded them, then threw them to the side, then brought them together again above their head. How odd.

The same guy just asked me what I was doing again. I gave him the same answer (DIARY!!) and he nodded and went away again. He'll probably be back.

In other news, I recently bought 30 DVDs from the UK. (Not coincidentally, last weekend my parents came to visit, which was nice, since I hadn't seen them in ages.) I would have bought the DVDs in Norway, but the prices here are astronomical, and despite this being highly appropriate given the content of the DVDs, it was not very attractive to me.

This purchase increased my total DVD ownership by a good base-10 order of magnitude. And it is quite impressive on the shelf.

The DVDs were seasons one through six of Stargate SG-1. My flatmates are eating it up like addicts, it's quite funny. They watched season one in about a week. They'll probably have finished season two by the time I get back tomorrow. I'm having a great time rewatching the early episodes, too... I might even resume my detailed analysis of key episodes which I started a year or so ago.

Jack O'Neill: ...what now?

Master Bra'tac: Now we die.

Jack O'Neill: Well that's a bad plan.

My ulterior motive for purchasing these DVDs was actually to attempt to get someone hooked on SG-1. I'm going to get an extended period of quality time with this someone and her three cute ferrets at Easter to attempt this.

I didn't used to think ferrets were cute, by the way. Now I consider them almost honorary cats. That's quite a statement, coming from me.

During the rehearsals for the band I mentioned above we sometimes spend many minutes playing sections where I have long 24 bar rests, so I've been taking the opportunity to work on a proposal for a Web Applications specification. This would be something along the same lines as Web Forms 2, but specifically for client-side application development. (I asked for ideas along these lines a few weeks ago — thanks again to those of you who responded.)

Some of the things I'm thinking of are:

• A per-domain, per-window "file system". This would allow you to be running multiple instances of the same application (from the same site) without the instances writing over each other's cookies. Anyone who has tried to compare prices of multiple Travelocity tickets knows what I mean by that. (I once nearly bought the wrong plane ticket because of this, so I feel quite strongly that there is a problem to solve here!)
• Native popup menus and context menus. As fun as it is to have to guess how the menus on every site work, I'm thinking it's about time we got consistent UI on this feature.
• For people who want it, command updating. This is where you have several access points for the same feature, for instance a menu item and a toolbar button, and you want them to all share the same "disabled" state, so you don't have to disable every one in turn when the option becomes unavailable.
• Sortable tree views and list views with rich formatting. Do a search on eBay, and you'll find you have to hit the server whenever you want to change the sort order. That shouldn't be necessary.
• Exclusive sections: At the moment, an implicit semantic of HTML documents is that all the content is always visible, you need but scroll to it. But in some applications this makes no sense: you may not want the login screen to be on the screen at the same time as the query page, for instance. At the moment you can work around this using CSS and display:none, but that's not semantically correct, and it breaks when you disable stylesheets, which it shouldn't. Having markup that actually means "hide this until it is activated" would solve this.
• Content-area popup windows. They're currently possible using positioned divs, but that breaks down when the stylesheet is disabled, and is semantically dubious at best. (Any time a div is the answer, there's a hole in HTML.) The idea would be to introduce the idea of a native "subwindow" element which would work as a movable window within the content area, which can be dragged, resized, etc, as required.
• Remote eventing. I was recently reminded that HTTP is a client-pull protocol, but that in many cases there is a need for server-push. Now, you can fake this using an iframe to a document that the server doesn't close and into which it keeps dripping script blocks, but that's a hack to say the least. What would be nice is a way of letting a remote server trigger DOM events that standard DOM3 Event technologies could hook into.

The house lights just went up for a split second (someone probably leant on the wrong switch) and after they went back down, a girl came over and asked me what I was doing. I'm writing about the party, I said. She found this fascinating. I have a feeling "I'm brainstorming ideas for extensions to HTML to enable richer user experiences with Web applications" would have gotten a rather less warm response, but maybe I should try it if someone asks again. Sadly they would probably reply "but you should come enjoy yourself! Have a beer! Horrah! Skål!" or some such, oblivious to the fact that I live for this spec stuff and don't like the taste of beer... Did I mention "to each his own" already?

Speaking of girls, I've noticed I get a lot more looks from girls now that I've shaved and cut my hair. I don't recall ever noticing this happen when I cut my hair in the past. I wonder what the difference is this time.

Incidentally, XBL is progressing nicely, despite seemingly insurmountable differences between "us" (the CSS working group members who are interested in XBL) and "them" (the SVG working group members who were behind RCC). I have a bunch of action items to deal with tomorrow, and then we'll be down to pretty much only one real blocking issue before publication.

Pingbacks: 1

2004-03-12 14:58 UTC New arch nemesis (still a kind of script, though)

The reason I nominally joined the SVG working group is so that I can remain one of the editor of the XBL spec (draft basic version, draft complete version) while we publish it through the SVG working group. As part of this, I've been spending the week so far working mostly on editing that spec.

Doing this, I ran into some edge cases that I need to define. If you create an html:script element, then insert it into a document, should it be evaluated? If you then take this element out of the document, and reinsert it, should it be evaluated? If you take this element and clone it, then insert the clone, should it be evaluated? If you take this element's parent node, and rename it using renameNode(), causing the children (including the script element) to be removed from the document then readded, should the script get reevaluated? What about if an ancestor is adopted into another document, causing ownerDocument to change? And what script context should be used in that case? What if the node doesn't have the script included inline, but uses a link to an external file instead?

I have a headache.

2004-03-05 10:40 UTC W3C Technical Plenary 2004

All this week I was at the W3C Plenary in Mandelieu La Napoule, in France. (Saying it's in Mandelieu La Napoule is a neat trick for those people who don't want to tell their companies that they are going to Cannes lest their management think they are using their travel budget for fun. The hotel is about 4 meters from the official edge of Cannes.)

Some choice quotes that I've heard so far (left anonymous to protect the guilty...):

So in my right hand I have a content declaration...

So if we dropped it. Would they lynch us? Yes.

The other solution is to give it an RDF property— (laughter erupts)

So the government normalised your name?

Hard day tomorrow? Hard? No. It's Semantic Web IG.

*bzzz, bzzzz* *bzzz, bzzzz* Our bathroom is ringing!

Wednesday was the middle day of the Technical Plenary, where instead of working group meetings everyone gets together in one big room and watches presentations and panels and asks questions. In the past this has sometimes been a lot of fun, but in this particular instance I found most of the sessions somewhat tame. It's much more fun when they pick very controversial topics.

One of the sessions was on Writing Test Suites For Your Working Group (or some such, I forget the exact name) and I was one of the panelists. For my slides I used a document Tantek originally published back in April 2002: the test suite documentation for the CSS Working Group. This document was obviously never written with slides in mind, so I applied the following stylesheet:

* { margin: 0 ! important; }
h2, h3 { font-size: 3em; }
p, dl, ol, ul, span, #abstract, #contents, #intro,
#conventions, #history, #usingtestsuites, #cover,
#getting, #next-and, #test-content, #something,
#sooner, #atomictests { display: none; }
.head p, .head dl, .head ol, .head ul { display: block; }
.head span { display: inline; }
.head dd { padding-left: 50px; }
.copyright { padding-top: 1.25em; }
hr { margin-top: 0.75em ! important; }
h1 { padding: 2em 0 0.75em; }
h2 { padding-top: 3em; page-break-before: always; }
h3 { padding: 5em 0 1em; page-break-before: always; }
#validtests + p { display: block; text-indent: -500em; }
#validtests + p a { text-indent: 0; font-size: 2em; float: left;
   padding: 0.5em; content: 'Suite #' counter(suite, decimal-leading-zero);
   counter-increment: suite; }
a[href="http://validator.w3.org/"],
a[href="http://jigsaw.w3.org/css-validator/"] { display: none; }
#testthetech { width: 80%; }
#incremental { font-variant: small-caps; text-align: center;
   letter-spacing: 0.12em; text-indent: -0.8em;
   padding-top: 9em; content: 'Fin'; }
[lang=tr] { content: 'Tantek \C7 elik'; }

However, since I couldn't change the document to add this stylesheet (the document is on the W3C Web site) I instead wrote a small little javascript: URI that dynamically added this stylesheet, using the DOM and an embedded data: URI, to the document, when I followed it (which I did my pasting the following into Opera):

javascript:void(function(){var e = document.createElement('link');e.rel='style
sheet';e.href='data:text/css,*{margin:0!important}h2,h3{font-size:3em}p,dl,ol,
ul,span,%23abstract,%23contents,%23intro,%23conventions,%23history,%23usingtes
tsuites,%23cover,%23getting,%23next-and,%23test-content,%23something,%23sooner
,%23atomictests{display:none}.head p,.head dl,.head ol,.head ul{display:block}
.head span{display:inline}.head dd{padding-left: 50px}.copyright{padding-top:1
.25em;}hr{margin-top:0.75em!important}h1{padding:2em 0 0.75em;}h2{padding-top:
3em;page-break-before:always}h3{padding:5em 0 1em;page-break-before:always}%23
validtests+p{display:block;text-indent:-500em}%23validtests+p a{text-indent:0;
font-size:2em;float:left;padding:0.5em;content:\'Suite %23\' counter(suite, de
cimal-leading-zero);counter-increment:suite}a[href="http://validator.w3.org/"]
,a[href="http://jigsaw.w3.org/css-validator/"]{display:none}%23testthetech{wid
th:80%25}%23incremental{font-variant:small-caps;text-align:center;letter-spaci
ng:0.12em;text-indent:-0.8em;padding-top:9em;content:\'Fin\'}[lang=tr]{content
:\'Tantek %5CC7 elik\'}';document.getElementsByTagName('head')[0].appendChild(
e);}())

Throw in Opera's projection mode and *tada*, instant presentation!

Also on Wednesday, we have the Birds Of a Feather lunch tables. This is where people sign up to sit at tables with assigned topics, and have a friendly chat to see what other people think about something or other.

During this lunch I accidentally joined the SVG working group. I don't mean I accidentally sat at the wrong table and found myself sitting with lots of SVG people. I mean I accidentally became an SVG working group member. I had no idea that sort of thing could happen during meal times. I'll endeavor to be more careful in future.

This brings the total number of official W3C-related activities I'm involved with to four: the CSS Working Group, the SVG Working Group, the Linking task force, and the Binding task force.

I've also come away from this week with a bunch of new and renewed action items. Expect new releases of the CSS3 Generated and Replaced Content module and the CSS3 Lists module to come out "soon".

All in all, a very productive week.

2004-02-10 19:28 UTC OperaShow Generator

For your next presentation, don't use proprietary formats! Use the OperaShow Generator to generate HTML and CSS which you can then render in any browser that supports the 'projection' media, such as Opera.

One of the great things about working in the Technology and Research department at Opera is that I get to play with all kinds of new gadgets and technologies. For example, last week one of our embedded device customers said that one of our competitors had tested Opera against their browser and found Opera was slower, and Opera was asked to prove that this was not the case, so Herman and I were yanked out of our normal routine and told to test the two products all day until we had an exhaustive analysis of the two browsers' pros and cons. (We quite happily obliged, showing that over GPRS, on the test pages that the competitor in question had used, we out-performed them in page load performance by seconds, sometimes dozens of seconds, and sometimes sheer orders of magnitude — not to mention that said competitor had no DOM support, ridiculously limited CSS support reminiscent of desktop browsers from the late 1990s, and so forth.)

One of the technologies Opera is quite proud of is our CSS media handling. There are internal versions of Opera that handle handheld, tv, screen, projection, print, and even speech. And some of these modes have fallbacks, for example if a page is viewed in handheld mode but has no handheld style, it will fall back on the SSR (Small Screen Rendering) view.

We also have lovely "content adaption" mechanisms — for example in some builds you can enable MSR (Medium Screen Rendering) mode, and then Opera will adapt the formatting of the page to fit your window. I should point out that this is not at all standards compliant, but it isn't supposed to be: its purpose is to handle the cases where the page wasn't designed for the device in question, and would be unreadable if displayed per the specs. Authors who target these devices can rely on standards-based rendering, since MSR won't trigger unless the page doesn't fit.

One of the medias that Opera supports which no-one else yet supports is projection. (Mozilla doesn't yet support this, it's bug 120398.) When you switch to full-screen mode, Opera switches from using your screen styles to your projection styles. This would be of very limited use, if it wasn't for the fact that the projection media is a paged media type, much like print, so properties like page-break-after work!

This lets you do slide shows in HTML with no JavaScript and no :target hacks, which is pretty cool. Håkon (my boss) wrote a little script that will let you make a three-page presentation without even touching the markup. As an added bonus, it even embeds the images you specify using data: URIs, so the page is completely self-contained.

I played around with it and came up with this cute little presentation.

2004-02-01 22:19 UTC Mozbot 2.6 candidate builds

If you use mozbot, I urge you to update to the latest development version. Assuming no blockers are found, I hope to release version 2.6 in a few weeks.

Here is a summary of the changes since the release of mozbot 2.4:

• New modules:

  Spell

  Checks for spelling mistakes:

  <Hixie> Checks for speling (sp?) mistakes.
  <mozbot> Suggestions for 'speling': spelling, spalling, sapling, speeling, spieling, spiling, spilling...
  

  ServicesLogin

  Enables mozbot to log in to network services such as Nickserv, K9, Q on Quakenet, or X on Undernet.

  Insult

  For those of you who mourn the passing of the insult server at the University of Colorado, you'll be glad that the insult server code was ported to mozbot so that mozbot can now natively insult people!

  Quotes

  If you have a MySQL server you can use the Quotes module to store and retrieve quotes. This module can automatically create the appropriate tables and so forth when you tell it which server to connect to.

  Currencies

  Mozbot can now provide you with up-to-the-minute mid-market quotes for currency exchange, using the information provided by xe.com.

• The bot now supports logging into password-protected IRC servers.

• The "!seen" command now stores its data in a separate file so that the configuration file doesn't grow without bounds.

• You can now tell the bot to stop talking to you by saying shut up.

• Added some protection for loops and annoyance attacks: mozbot will refuse to say the same thing more than a few times in a row.

• The Converter module now supports time conversions and mass conversions as well as the already-supported length, temperature and integer conversions.

• The XMLLogger module now logs mozbot parting a channel.

• The Greeting module has been greatly improved so that it now is much more talkative and fun.

• The built-in help command is a little more helpful. The help also now has two styles: compact (the default, as seen in previous versions), and tidy (the new longer alternate style). You can select the style you want by changing the "helpStyle" variable:

  /msg mozbot vars General helpStyle 'tidy'
  /msg mozbot vars General helpStyle 'compact'
  

• Mozbot will now handle nickname changes much more gracefully. It will also handle nickname clashes and erroneous nicknames in a much more solid way than the previous version.

• Similarly, mozbot is now much better at handling being kicked, parting a channel, being invited to or told to join a channel, and quiting IRC. You should no longer see the "I think I'm already *in* channel #foo!" message. (In fact I guarentee that you will no longer see it, since I changed the text of the message while I was making it less likely to appear.)

• If you use a bot that has strange restrictions such as not accepting long IRC names, or requiring correct usernames, mozbot will now automatically detect the problem and fix it on the fly with no user interaction.

• Mozbot is now able to ignore other bots entirely, and ignore people talking to other bots. You can enable this using the ignoredUsers and ignoredTargets variables, which are described in detail in the INSTALL file.

• Improvements to the mozbot module API: it now supports notices, modules can now get to the current message queue, and some other minor changes. The documentation has also had a number of errors corrected.

• The Infobot mozbot module is now slightly more helpful, and for those of you who are using mozbot primarily for its Infobot abilities, you can enable the eagerToHelp mode which will make the Infobot module respond even if asked a question without a question mark (this mode is on by default for compatibility with the Infobot bot). For those of you who just want mozbot for giggles, you can even enable ridiculouslyEagerToHelp mode which is just that. (Ridiculous, I mean.)

• The Bugzilla module has had significant updates and is now compatible with the latest version of the Bugzilla software.

• The Sheriff module has been updated to handle some common things people put in their sheriff lines.

• Over thirty other bug fixes.

You can get the full change log from bonsai.

Please let me know if you find any bugs. (If you can't catch me online, file a bug.) Thanks!