- My facebook-recorded prediction was 43-29-28 (D-Mc-Mo), which turned out to be off by a bit, but not embarrassing. The final numbers will probably be 49-27-24. Hey, at least I got the tens digit correct.
- Nate says McAuliffe's numbers look like Dean's from Iowa 2004. I disagree -- they look more like Clinton's from Iowa 2008. Clinton actually had a lead in that race (as did McAuliffe) -- Dean never had a solid Iowa lead.
- Ed Kilgore thinks the results spell trouble for McDonnell and I'm inclined to agree. If Deeds can beat Moran in Arlington County, he clearly can develop broad-based support.
- I disagree with Ed that the WaPo endorsement wasn't everything -- it was. Check out the polls before and after the day of the endorsement (May 22). That endorsement allowed Deeds to corral the momentum and the anti-McAuliffe vote. For voters who wanted an experienced state legislator running Richmond instead of a fat cat outsider, WaPo forcefully directed them toward Deeds.
- Well done VA Board of Elections for getting 60% of the vote in within the first hour after polls closed!
- Turnout (probably just
shy ofover 300,000) decimated predictions and the 2006 primary turnout of 155K. Go VA Dems!
[This blog post is adopted from an email conversation with Chicklet. Coincidentally, one of Google Wave's features is to convert an email to a blog post with just a couple clicks. Disclaimer: I'm not a Google employee so I know nothing about Wave but what's in the media; however, I am currently working as an independent contractor on a Google project.]
I mentioned to Chicklet that I thought Google Wave looked cool, and certainly others agree. She responded that she might find it appealing if she knew what Wave actually was. So after watching (some of) the Google I/O video, here's my attempt at explaing Wave briefly:
Wave is basically one step closer to bringing the whiteboard to virtual communication. Think of it as shared docs in email/IM. Here's an example of what it does well: at Woody Woo, when they bring speakers they post a blank list with 15 slots for students to sign up for a discussion with the speaker. First 15 to sign up get into the seminar. Email can't replace this functionality because (for one thing) you can't know how many slots are left (without manually counting reply-all emails). Also, if you don't have conversation threading, people's inboxes flood.
Wave solves this by combining email and shared docs. Basically the admin would send out an editable email with 15 slots and the recipients would be able to edit it. (It tracks changes so you could tell if Doug erased Amy's signup.) And of course all the edits are in one "wave" so no inbox flooding. The challenge for Google, I think, will be interoperability. What happens when only 5% of your social/business network uses Google wave. Does each edit of a wave with MS Outlook user flood the inboxes of the outside users? Is Wave worthless to you if only you use it? Google appears to be trying to solve the adoption problem by having Wave appear on public sites like blogs where comments become wave-y. That should help non-adopters see the benefits of wave without changing their behavior. I also think people who (like me) have their gmail window open all day will quickly adopt wave as their inbox and gradually utilize its more advanced features. Of course, all this assumes that Wave lives up to its hype.
We got home so smoothly that we had turned on the TV to see that Obama's congressional lunch hadn't even started yet. (They were quite behind schedule!) After a relaxing afternoon, we got all dolled up for two unofficial galas: the Pennsylvania State Society Gala and the Google party. On our way to the parties, we took the Metro -- I always love seeing the tuxes and evening gowns on mass transit:
Seventh graders who pay attention in civics class know the difference between "impeachment" and "removal from office." The Washington Post had a bit of trouble on that front today. The Grey Lady did much better:

The change in the House (colored by majority of state delegation) is especially stark: everything east of the Mississippi except for the deep south is blue.
Tomorrow is all about Obama and rightfully so (I'll post pics asap), but tonight I just want to make sure Governor Dean is not forgotten.
Yes, okay, I'm at least a year too late with the joke. And yes, okay, no one laughed. But, I get home from NYC and check my RSS feeds and what is the XKCD comic for the day?

Freaky.
I think it's ridiculous that a bare majority of voters can amend a state constitution, but that's not the question at issue today in California. Instead, Prop 8 opponents (i.e., the side of justice and equality) have filed suit saying that Prop 8 is a "revision" to the constituion rather than an "amendment." And since I figure you all are just as ignorant as I am about the distinction, here's a good primer.
<b>Update</b>: After mulling this over for a day, I'm against this lawsuit. I think the appropriate action is to put the issue on the ballot in 2010 (either in the primary or general, whichever one gay rights activists think will be better). And in that case, we would want to argue that striking down Prop 8 would be an "amendment" and not a "revision" (so that we wouldn't have to garner support from 2/3rds of the legislature). Filing this suit undercuts that claim and also undermines the credible of the movement with the voters. Convincing 52% of Californian's to support gay marriage in 2010 will be a lot easier if we don't attempt to throw out their votes with a lawsuit today.
For the better part of a year I worked for Google but couldn't tell people the details of my work because of an NDA. Over the past few months the details of the project have emerged and this weekend the true promise of the project came to fruition. So, let me introduce you to the Voting Information project.
The goal of the Voting Information Project (VIP) is to free election data from the confines of state and county databases. This pursuit fits nicely with Google's mission statement of "organizing the world's information" and the project is a partnership of the search company, Pew, and JEHT. We started in early 2007 with the long-term goal of creating an open format that states could use to publish their election geographic and ballot information, and that organizations like the League of Women Voters could layer candidate questionnaire responses over.
For this election, we focused on election geography, specifically informing citizens where their polling location is. Past attempts at a nationwide polling place locater had failed because not all states or counties were willing to share information. However, with the advent of nationwide voter files, one step of the process was complete: every registered voter in the U.S. could be matched to a precinct. Google devoted many resources this summer filling in the second piece of the puzzle: matching precincts to polling location. Now, every registered voter should be able to find his or her poll location here. (The Obama campaign also used this strategy.)
But, what about new registrants (or even same day registrants)? If they live at an address without another registered voter, Google's strategy will fail. That's where VIP comes in. Participating states and counties, distribute their street segment infomation (in a common format), which matches all addresses in the state to a precinct and then a polling location. These street segments are exactly how registrars themselves determine the precinct of new registrants, so they are very reliable.
I'm happy to announce that this past weekend, Google combined these two strategies at maps.google.com/vote. Test it out! One of my Princeton colleagues hadn't realized that her polling location had moved until she tried the application, You can even put the app on your own website as a gadget. Also, the website got some press play from the NYTimes and Te The Today Show <video>. So that's what I've been up to.
Also, go vote!
| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | |
| 7 | 8 | 9 | 10 | 11 | 12 | 13 |
| 14 | 15 | 16 | 17 | 18 | 19 | 20 |
| 21 | 22 | 23 | 24 | 25 | 26 | 27 |
| 28 | 29 | 30 |
