Still going after all these years.
Looks like I’m about to be published in another international newspaper. This time it’s the Guardian, who are doing a piece about the phoneblog exhibition I pointed to a while back. I’ll put the pointer up as soon as I see the article.
Just a side note: this is definitely another case where my participation in the blogosphere—by which I mean blogging, photoblogging, and reading other sites—has led me in some unexpected directions.
Two quickies and then I must sleep:
feed: URL proposal. While I understand the technical arguments that a lot of folks are making—that this should be a MIME type, not a URL type, since the feeds are still fetched by HTTP, I can’t argue with the end result. One click subscriptions are the one feature I really missed after I migrated from reading RSS in Radio to other aggregators, and now it looks like there’s an emerging cross client consensus on how to make the feature work everywhere—for instance, in NetNewsWire. Right on.
Scoble says that the Radio trackback feature (which is implemented identically to the Manila implementation on this site) makes his site fail validation. Frustrating when a useful feature like that has to be turned off, but I understand the pain; I’ve been working on validation myself.
At this point there are still a bunch of errors reported by the W3C’s validator, none of which I can do anything about. They all have to do with ampersands in URLs in my articles. Ampersands are commonly used in URLs when there are multiple arguments. Manila thinks that those ampersands should be represented by actual & characters (and enforces this in its managed content), while the W3C validator insists that, at least for HTML 4.01 Transitional, the ampersands must be represented as &, even in URLs. So there’s nothing I can do; the site is as valid as it’s going to get.
(Incidentally, if anyone has Manila spitting out valid XHTML, I’d love to know. It would be nice to get off HTML 4.01.)
For more info about CMSes and HTML validation, there’s a great interview with What Do I Know’s Todd Dominey over at WebStandards.org.
I think I fixed the CSS bug that was causing my site to crash IE 5 on Mac OS X. The offending rule was for list items inside unordered lists inside the .navContainerSide class, which are specified as floating—apparently having floated list items breaks IE 5. I used the Comment hack to hide the offending CSS from IE 5 and it now appears to work. Please advise if you’re still having problems reading the site with a particular browser.
Dave is collecting RSS subscription lists in a new application on Scripting News. With an n of 51 subscribers giving their information, the top five feeds (out of 100) are Scripting News, Wired News, Scobleizer, Boing Boing, and the Doc Searls Weblog. This is what is meant by non-representative sample; still it gives an interesting insight into the people who are participating on the RSS-user list.
I’m less interested at this point in who is being read (we have plenty of other tools to tell us that) and more in information about peoples’ reading habits. How many feeds, on average, do people subscribe to? What is the blog-feed to non-blog-feed ratio?
My new site design appears to crash IE 5 on the Mac, and I think I know why—it looks like a problem with the custom list CSS that I use to show the category buttons in the sidebar. If I suppress that section of the sidebar, the page loads, but the same code doesn’t render correctly in the header. I will work on this later; in the meantime, if you are having problems reading the site, try the print friendly version.
We bought a new Whirlpool washer yesterday. I thought I had taken down the dimensions at the store, but can’t find them now, so I went on line to try to find them. And tried. No reference to the model number (LSB6400LW—well, now there will be at least one) is found in Google. And the Product Literature page at Whirlpool.com doesn’t work. At all. There is a pop up from each of the links for the Use and Care Guide and Installation Guide, which goes to a page with links that don’t work on Safari, and that lead to a broken page on other browsers.
It’s the 21st century, folks. No excuse. Fix your damn pages. There’s no reason that your content management system, or whatever, should generate such client-hostile links.
Looks like Moxie is redesigning too. Her woes make mine look small by comparison. And Dave Sifry (of Technorati fame) did some redesign work as well, mostly back end.
I think that most of the work on the redesign is done. As you can see, the site now sports a new logo, new fonts, a different design, streamlined navigation, some new pages, and a bunch of other goodies.
So what’s left to do? There are always a few things. I think there may be some weirdnesses on IE Win that I need to fix. I need to fix a few graphics here and there.
But the biggest thing, ironically, is that I’ll have to repeat the rendering exercises that I did over the last two weeks all over again, so that all the static pages pick up the new design. Sigh.
Quick shout out to everyone’s favorite Tin Man, who turned 30 yesterday. In his countdown of the ten most memorable events of his 20s, he included the Virginia Glee Club’s Northeastern Tour, which I helped organize; thanks for the reminder of good times, TM, and hope that there are plenty more to come.
We got in last night around 10 pm; Joy and Jefferson are ecstatic to be home. The trip was mostly uneventful, with two exceptions. Note: Dogs can be made to go to the bathroom in airport stalls, but not on their pee pads (no matter how hard you try), so put rubber gloves in your list of doggie carry-on supplies. Now dogs and Lisa are asleep on the sofa in our living room and I’m catching up.
I’ve been reading news feeds over the holiday, but didn’t get much of a chance to post anything about what was going on outside our own vacation. So here goes:
Things definitely to come in the redesign:
Things I’m thinking about:
Here are some of the extremely helpful resources I’ve been consulting:
Other things going on in the news around the reports of Saddam Hussein’s capture:
Oops. Guess the last one slipped in…
Another quick link check round:
Weblogs.com hit a new high water mark yesterday; at the peak during the day, 4,851 weblogs updated in a three hour period.
Note the high R-squared value (0.953) for this log normal plot, based on fitting the data with an exponential curve. Compare against the linear R-squared value from the last high water mark in October, which had fallen to 0.784. Based on the data and the steadily decreasing R-squared value for the standard linear regression, I think it’s now safe to say that the data confirms that the high water mark grows at an exponential rate. Source data, as always, is available here under a Creative Commons license.