Dan Gilmore has commented on Dennis Kucinich's criticisms of voting machine manufacturer Diebold.

Voting transparancy seems to be a big issue with Kucinich. It's great that someone is talking about this.

People look at the Florida 2000 situation and think technology is the answer. And certainly technology can help. A touch screen can be a more accessible mechanism for chosing from a menu of choices than a punch ballot in small print.

But there is a big difference between clarity in voting, and confidence in voting. You may be more sure you chose who you meant to choose, but how certain are you that your choice was properly counted?

I think our first steps towards introducing more technology into voting should be to address clarity for the user.

But for voter confidence we should still have physical tokens generated by the voting process. These tokens should be able to be examined by the voter at the poll. And of course they should be countable and recountable by hand and by machine.

Check out these recent entries in my Apache log files:

Both of these entries have a few interesting points. They are from the same ip address, which does not resolve to anything. They both report a user agent of "MSIE 6.0", which is not the actual user agent string of any version of IE. And they both point to fake blogs.

Neither of these sites has an actual link to my site. So the referrer strings are fake.

Looking up the domains reveals that both sites have a lot in common:

Domain Name: kwlablog.com Name Servers ns1.kwlablog.com 141.85.3.109 ns2.kwlablog.com 141.85.3.109 Domain Created: 11/8/2003 Domain Expires: 11/8/2004

Domain Name: jennifersblog.com Name Servers ns1.jennifersblog.com 141.85.3.106 ns2.jennifersblog.com 141.85.3.106 Domain Created: 11/8/2003 Domain Expires: 11/8/2004

But the sites have different names/addresses for Administrative, Billing, & Technical contacts.

Visiting the sites reveals that they have blog entries that are just links to news stories. And most of the other links on the sites (like the archives links) don't actually do anything.

What is the purpose of these fake blogs? And what is the purpose of scattering them around webserver logfiles in the form of referrer strings? Are these sites trying to take advantage of the recent trend to post referrer links within page content? Are they using this fact to try to boost their search engine pagerank by appearing in this generated page content?

I've found some blog entries on the referrer spamming I mention in the previous floppaganda post. You can find some good discussion at net warriors.org blog.

Check out these blog entries in particular: More Referrer Spamming, Referrer SPAM updates, Referrer Spamming, wrap up.

I've expanded the discussion of ad blocking css on the home page, and tweaked the look of the entire site slightly. Next up is a Python starter project to do some simple Apache log file analysis.

I now have instructions for using my ad blocking userContents.css with the Safari web browser from Apple. Find the updated instructions here.

So much for lying in a beanbag wearing tighty whities and weilding a remote. I've decided to join the world of the employed once more. A9 looks like a great opportunity to learn a whole pile of new things and try to solve some fun problems. A9 is Amazon's search technology arm, spun off into a subsidiary down here in Palo Alto. There have been a few articles in the press already, even though A9 isn't really started yet.