September 5, 2008
Classic Colbert on Wikipedia
Earlier this week I mentioned Stephen Colbert and how he got fans to manipulate Wikipedia to change the reality of the status of the African Elephant. Here's the clip. [MG]
More Buzz on Chrome
Google's new browser, Chrome, is a traffic generator as people look for the download. Information Week has the story. There was some backlash over harsh terms in the Chrome EULA. Google has said the language was a mistake and has changed it retroactively to cover all downloads. That story is in PC Magazine. The most interesting story comes from CNET, where Stephan Shankland documents commands that can be entered in Chrome's omnibox which causes the browser to give up details about its current configuration, such as memory usage, statistics, and other useful information. That article is here. [MG]
September 3, 2008
8th Circuit Says No to Wikipedia as Evidence
The Eight Circuit is not a fan of Wikipedia, at least when the online source is used for the legal basis of a ruling of the Board of Immigrant Appeals. The case is Lamilem Badasa v. Michael Mukasey, (07-2276, 2008 WL 3981817) on appeal from an order of the Board of Immigration Appeals. Badasa applied for asylum under 8 U.S.C. § 1158 and Article III of the Convention Against Torture. Badasa's documents were deemed not credible in establishing her identity. She moved to reopen her case as she possessed a new document, a laissez-passer, issued by the government of Ethiopia. The Department of Homeland Security, as a party to the immigration proceeding, submitted a definition of laissez-passer sourced from Wikipedia which the Immigration Judge accepted. This document is a single-use, one way travel document issued on the basis of information supplied by the traveler. The IJ affirmed the denial of the first examination.
The case went to the Board of Immigration Appeals who said the decision that the laissez-passer document did not establish Badasa's identity was not clearly erroneous. The BIA also noted that it was uncomfortable with Wikipedia as a source, but there was other evidence supporting the IJ's conclusion. The Eight Circuit looked at this and said the case had to go back as while there might be other evidence beyond Wikipedia, the BIA never stated what it was.
From the Court's opinion:
We know only that the BIA thinks that if, hypothetically, the IJ had not considered Wikipedia and reached the same conclusion, then that conclusion would not be clearly erroneous. But we do not know whether the IJ would have reached the same conclusion without Wikipedia, or whether (and, if so, why) the BIA believes that the IJ’s consideration of Wikipedia was harmless error, in the sense that it did not influence the IJ’s decision. Because the BIA’s ultimate conclusion that Badasa failed to establish her identity is not adequately explained, we must remand for further proceedings. See Shahinaj v. Gonzales, 481 F.3d 1027, 1029 (8th Cir. 2007).
The Court had a few scathing words for the use of Wikipedia as the basis for a legal decision. Wikipedia is not good for establishing facts in evidence used as the basis for a legal judgment given that, by its own admission, Wikipedia content may be in flux, not verified, or written by individuals with agendas. Recall Stephen Colbert urging his viewers to change the entry on the African Elephant to say that the population had tripled and was not endangered. Nothing was further from the truth. Wikipedia editors had to lock that particular entry from further vandalism. Colbert proved the point that saying something and reality may be two different things.
Federal courts have cited Wikipedia 167 times as of today as referenced in Westlaw. Most of the citations are used to elaborate on a definition or explain a concept. However, almost all of these references cite other sources that are fixed or substantial enough that the issue in the Eight Circuit is not present. Some of the courts do give disclaimers about the viability of Wikipedia as a source even though they use the web site. One litigant tried to use an article from Wikipedia as evidence of ineffective assistance of counsel after pleading guilty in an arms possession case. See United States v. Allen (10th Circuit, July 29, 2008, 2008 WL 2894006). The Allen Court didn't bite on that one, dismissing the appeal out of hand. The lesson? Wikipedia is not the best source for information used in legal arguments. [MG}
September 2, 2008
Google's New Browser -- Chrome
This post is coming to the blog via the new Google browser, Chrome, almost. See below. When AT&t decided to build a browser, many commentators wondered why. When Google builds one, the online world is filled with intrigue and speculation. As most people know, Microsoft's share of the browser market is on a par with Google's share of the online advertising market. So what does a Google browser mean? Google's own statement is that the company wanted a browser that was optimized for running web applications. From Google's perspective, the browser is less a piece of software than a place to run applications coming from the web. Applications such as word processing, spreadsheet, or presentation software, perhaps? Microsoft take note.
Chrome has some interesting features which it details in an interminably long online comic book hosted by various Google staffers. The most interesting features include running each web tab as a separate process. This means if one tab crashes, it doesn't take the entire browser down. Chrome has a privacy mode similar to that in the IE8 beta. It's accessible via a tiny wrench icon in the upper left hand corner of the screen.
Chrome takes the application base seriously by eliminating the traditional home page. The opening page collects thumbnails of most recently viewed sites and pages. In fact, there is no icon that takes one back to a home page. It's just the browser. The problem is that there is no way to edit this list, meaning that duplicate pages or page no found links will sit there just as much as any other page. One assumes that these will disappear over time as other content takes their place.
Another big change is that the address bar and the search box has been combined. Type an address, go to a page. Type a search, get search results from Google and go a link from the list. Chrome is not locked into Google, by the way. It defaults to Google but asks if you want to change to another search engine during installation. It's not as if key words in the address bar are anything new or unavailable in competing browsers. The integration with a specific search engine is fairly tight in this implementation. One other feature that will be noticed by those who enter text in online apps such as web mail is that misspelled words will be flagged through the traditional red underlining. This feature shows up in IM messages as well.
The browser is written in the open source WebKit, which underlies the Safari rendering engine. The pages come up fast and accurate, or at least the ones on the sites I have mostly visited show no problems. One exception is Typepad, which hosts this blog. As I attempted to insert links, they did not format in the text as I had expected, which forced me back to Internet Explorer for the final post. We'll see what else shows up.
More on Chrome as I use it. [MG]
Get Chrome here.