The Substantially Similar Weblog

I’m guessing this will propagate through the meme pool· pretty rapidly, but check out The Buttafly Guide to Interpreting Friendster Photos· (and An Ode to Friendster·) if you have spent any time perusing the zeitgeist indicator known as Friendster·.

Apparently there’s another Kessel who has maintains a weblog:

Kessel did great in practice yesterday. Her heeling was fine and her attention was good. I had a chance to remind/reinforce attention when some people walked by… she wanted to see where they were going!

No relation, I think.

According to CNN:

There are over 4 million blogs on the net, more than half run by teenagers. Research group Perseus says the typical blog is written by a teenage girl who updates it about twice a month.

These sorts of “statistics” always strike me as suspect. Both the 4 million figure, and the “teenage girl” figure. The article goes on to claim:

But unlike www.bigwhiteguy.com most of them will be little seen, if not abandoned. At least two thirds of the blogs out there today have not been updated in months.

Does this sort of figure even make any sense? There are a lot of “abandoned” web sites out there, dead links, etc.. I’m not sure we can conclude anything from that. If a blog isn’t read and isn’t written, does it really exist?

In my view, if CNN wants to capture the zeitgeist, they should focus on active blogs, both from the reader’s and writer’s points of view. Otherwise, it’s kind of like writing a story about how many newspapers there are out there that are no longer in print, and thus no one reads them. What use is that information?

Good discussion on linux-elitists bashing broken challenge-response systems (or, to clarify “challenge-response systems, which are by definition broken”). Including an excellent explanation of the problem by Karsten Self.

Incidentally, if you don’t know what the Joe Job is, here’s a good definition.

I’ve decided it’s time to jettison the background image (“wallpaper”) for this weblog. Time to move on to bigger and better things.

I’m actually shopping for new blog software, as I’ve decided I’d rather not be maintaining the software that generates this thing for the rest of my life. And it doesn’t do a very good job at generating good URLs, and the comments feature is broken.

I’ve been experimenting with b2, b2++, WordPress, pyblosxom, and blosxom. So far blosxom seems to be the winner, but I’m open to suggestions. Anyone want to advocate for a particular package? Requirements are free software license (DFSG approved, for example—so no MoveableType) and able to run on a standard GNU/Linux server. I’d also like it to have a good (and sustainable) URL system, permit more than one keyword or topic to be associated with each blog entry, and be an extensible framework where you don’t have to spend a year studying the code before you can write a plug-in.

New cool feature: the google calculator. Type all sorts of mathematical expressions in as your google search, and you get the result as your answer! It must all be part of google’s scheme to replace the URL with the google search. Actually, to replace your whole computer with a google search.

Aside from things like simple arithmetic (e.g., (24*7/3)^2 mod 5, you can also ask it things like three cups in tablespoons (and it tells you how many tablespoons!).

I also read somewhere that one of the top searches in google is cnn. That is, many people go to google and search for cnn rather than going to the fairly easy to remember cnn.com. Mozilla Firebird now seems to take you to the “I’m Feeling Lucky” google search result if you just type some random string into the URL bar.

Steve points out that you can also do searches for things such as radius of earth, and get a nice numerical result back from the google calculator.

I’ve been working to get the Lexis-Nexis online legal research system to generate valid HTML so the service works with browsers other than Microsoft’s. Back in April, Lexis promised the fix was imminent. About a month ago, I wrote them back explaining that they had fixed it for users who spoofed their browser agent as Netscape, but not Mozilla (or for Netscape users on GNU/Linux). I got no response, so wrote them back yet again more recently.

Finally, they’ve “fixed” it for the Mozilla browser. Here’s the response they gave me:

 Adam 
 Thank you for your continued contribution regarding these issues. Rest assured that your feedback is received and considered. Following the August release the Tab issue was corrected for the Mozilla Browser. A user should not have to "spoof" NS to get Mozilla to function properly regarding the tabs. I also checked Opera and the tabs are all working on that browser as well. Both Opera and Mozilla combined comprise less than one half of one half a percent of user access for lexis.com. Linux currently comprises .06% of OS access to lexis.com. While we try to address usability and functional issues that effect our customers, there is an analysis involved regarding cost of fix v. benefit returned. The Linux OS, as well as the Mozilla and Opera browsers, are not included in our testing programs at this time due to their low use. When issues are subsequently discovered we examine them on a case by case basis to determine feasibility and cost of implementation. Thank you again for taking the time to contact us and for using LexisNexis. 
 Regards, 
 lexis.com product management 

All I was asking for was to have the

More fundamentally, if these companies adhered (at least generally) to the HTML standard (any version!) then many of these problems would go away. The whole point of standards is to reduce development costs; you shouldn’t need to test your product on one hunderd and one different browsers. If it passes an HTML validator test, then you can call it a day.

This whole episode also hammered in for me yet again the advantages of open source development. If there’s a problem and the developer refuses to fix it, then you can do it yourself. With a proprietary system like Lexis-Nexis, you’re at their mercy, and even a one minute fix won’t make it in if it’s not on their agenda.

I mentioned yesterday that the only way to use Lexis-Nexis with Mozilla is to “spoof” your browser agent, to pretend to be Netscape. Dan’s Web Tips has an interesting article describing the problems with spoofed user agents. He makes a very good point: certain websites only allow Internet Explorer (or maybe IE and Netscape), so in order to get in, you have to pretend to be using one of those browsers. The site administrators see that everyone trying to access their site is using IE, and so they get the impression they’re not excluded anyone. The result is a vicious circle where alternative browser marketshare is underestimated because of the unreliability of user agent strings, and site administrators don’t think they need to fix anything.

Many months ago, I wrote that Lexis-Nexis had agreed to fix their broken HTML, which prevented their research service from working with the Mozilla browser. At the time, they said it would be fixed “within a month.” Since then, I have received emails from many Mozilla users with similar problems.

I recently wrote Lexis to ask why they hadn’t fixed the problem, and got this response:

 Thank you for your message. The fix for the issue you are describing went in for Netscape earlier this year. All other platforms will be addressed in a release planned for late August this year. Thank you for using LexisNexis for your research needs. Regards, lexis.com product development 

The “fix” in question would be to switch

with

. I installed Netscape to see if their claim was true, but it wasn’t—still broken HTML, broken links. They wrote back again:

 I just tested linking on the tabs in both NS 7.1 and 7.02 using www.lexis.com and http://www.lexisnexis.com/lawschool/ and I can link on the tabs just fine. I tried the same on Mozilla 1.3.1 and the tabs do not work. This would be consistent with the fixes we have put in. As I mentioned, the tabs should work for all browsers after August. I don't know why you are still experiencing the issue in Netscape. Are you using an older version? Are you on a MAC? I do appreciate your pointing out the issue and trust that we are trying to address it. If you want to discuss or investigate the nonfunctioning tab issue further please call our Technical Customer Service Department at 800.543.6862. Thanks again! 

As it turns out, what they had done is checked to see if you were using Netscape 6 or 7 under Windows only, and if so, deliver content with the proper tags. Otherwise, you get the broken tags.

So the temporary fix is to spoof your user agent, which you can do under Mozilla Firebird with the User Agent Switcher. There are several other programs with similar functionality. If you set your browser to report to Lexis that it is:

 Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.0.2) Gecko/20030208 Netscape/7.02 

Then you will get the proper tags and Lexis will work!

I know this may only interest a small group of people, but I wanted to get it out there. It’s frustrating that Lexis hasn’t made more of an effort to provide functional HTML, especially when the fix is so trivial.

I was reading our free daily subway tabloid, Metro, when I came across an article about “Hunting for Bambi.” The article describes what appears to be an unbelievably offensive sport, where men with paintball guns pay thousands of dollars to hunt down nude women in the forest. It immediately struck me that there was something awry.

It turns out it’s a hoax.

But Metro wasn’t the only newspaper taken in; try a Google news search for “hunting for bambi”, and you’ll see dozens of reputable news outlets covered the story. Of course, at this point, the stories are being overwhelmed by the discovery that it was a hoax. Here’s one typical story as it appeared in the Washington Times:

Jockstrip: The world as we know it
By Alex Cukan
United Press International
NEW ADULT ‘ENTERTAINMENT’
Men are paying thousands of dollars to shoot naked women with paint ball guns near Las Vegas.
Hunting for Bambi is the brain child of Michael Burdick. Men pay anywhere from $5,000 to $10,000 for the chance to come to the middle of the desert to shoot what they call “Bambi’s” with a paint ball gun, KLAS-TV in Las Vegas reports.
Burdick says men have come from as far away as Germany. The men get a video tape of their hunt to take home.
Burdick says hunters are told not shoot the women above the chest, but admits not all hunters follow the rules.
The women get $2,500 is they don’t get hit, something they admit hurts, and $1,000 is they do get hit, according to KLAS-TV.

I think the problem here is that once a story appears in one “legitimate” mainstream news outlet (in this case, KLAS-TV), all the other media think they can just carry the story without rechecking the original facts.

I’d hope to see a retraction in the Metro, but I’m not going to hold my breath.