 |
February 2007 |
|
| E-News |
|
|
|
| In This Edition: |
|
|
2/19/2007 1:00:14 PM, by Jacqui Cheng
The 911 system in the US works well under an average day's load but comes to a crashing halt under widespread emergency situations, according to two lecturers from the University of Maryland. To combat this, Ben Shneiderman and Jennifer Preece are proposing (subscription) a nationwide network of 911 web sites—911.gov—that will include input from the community as well as information from the government's emergency services.
During incidents like Hurricane Katrina, the Kobe earthquake, the British foot-and-mouth disease outbreak, and the Indonesian tsunami, communities collaborated via the Internet to provide supplies and assistance to those in need of help, which is what Shneiderman and Preece are looking to create. Under the proposed project, the web sites will be run by "trained volunteers with a few professionals, much as volunteer fire departments now operate," according to the proposition in Science magazine. The sites would accept text, video, and photos from the community to help emergency crews react more quickly to disasters as well as inform others of the status of the situation via the web. "Citizen reporters would report to a centralized authority who will take care of emergency response coordination and allocate scarce resources of police fire and medical services," Professor Shneiderman told BBC News.
Professors Shneiderman and Preece believe that such a project could succeed due to the popularity and growth of other social/community-driven sites, such as Craigslist and the Amber Alert system. With so many bloggers and "citizen journalists" out en masse during emergency situations—remember how many new sites and photos were online immediately after Hurricane Katrina?—such a coordinated effort could empower the community to support and protect each other. The site itself would be a national effort, but offer community-based sections for reports on each city, like Craigslist.
Such a project would not come without challenges, of course. Just the current 911 system is subject to pranksters, the 911 web site would also likely be a focus for attacks. "You would have to pre-register, the system would not allow anonymous entries," Professor Shneiderman said. "You must get people engaged in advance, to try it and learn it and be part of it." But enticing community members to be active before a disaster strikes could be of a challenge, as it's easy to mindlessly go on with everyday life until it gets interrupted. Also, the success of such a program will depend highly on the level of freedom they allow the citizen journalists to have on the site. If there are too many restrictions and lockdowns on content, users will just seek out other mediums for organizing their efforts, defeating the purpose of a centralized, government-sanctioned web site to begin with.
As for costs, Shneiderman and Preece expect that a 911.gov service would possibly double the current 911 phone center budget, which seems like a lot for what amounts to be a glamorized wiki. They suggest that the community response groups could be funded and maintained through fees from ISPs, similar to the fees on phone lines to support 911 service. If implemented, they expect development, pilot testing, and implementation of the project to take three to five years. For now, the professors may launch a pilot of the program at the University of Maryland later this year, with possible funding from the National Science Foundation.
2/20/2007 8:28:49 AM, by Ken Fisher
Viacom has announced a deal with Joost that will see the new P2P-based video start-up "broadcasting" Viacom's music videos, TV shows and other materials on the service. Viacom will provide Joost with content not only from MTV, but also from BET, Comedy Central, and even movies from Paramount. The news comes less than two weeks after Viacom backed out of talks with YouTube and demanded that their content be removed from the site. YouTube pulled down more than 100,000 videos.
According to the Wall Street Journal (subscription required), which first broke the news, Viacom usually takes a 2/3 stake in the advertising profits stemming from such deals. The terms of this specific deal have not been disclosed, but we believe it's in that general ballpark.
"We’re extremely pleased to be working with Joost, and couldn’t be prouder to be a key partner in the launch of the next generation in broadband video technology," said Philippe Dauman, Viacom President and Chief Executive Officer in a statement. "In addition to strong partnerships we have with traditional distributors, we will continue to seek out partners like Joost, which has created an exciting breakthrough platform that represents not only a fantastic user experience, but one that is built on a compelling and sustainable business model that respects both content creators and consumers."
Viacom CEO Philippe Dauman told the Journal that Joost's promise to protect their copyrights was a major factor in their decision, and also a stumbling block in their discussions with YouTube/Google. At the moment is it quite easy to download and store video content from YouTube, but no such exploit for Joost is known to exist. It is difficult to know exactly what Dauman means by protection, however.
Truth be told, Joost is nothing like YouTube. Joost is all about TV-length programming, although it can show shorter clips and even feature-length films. Most importantly, Joost is focused on commercial video content, not the user creations that have made YouTube so popular. To wit, you cannot upload content to Joost, making it a "secure" distribution medium in the eyes of many in the entertainment industry. Joost's founders don't have to deal with promises of filtering software or any of the other problems that stem from allowing anyone to "broadcast yourself (or someone else's content)." In this way, the two services are quite different.
This does not mean that YouTube and Joost are not competitors, however. The video advertising market online is very young, and a king has yet to be crowned. YouTube is undoubtedly feeling the heat, as the latest round of talks with Viacom ended rather poorly. Not only could the two sides not reach a deal, but Viacom decided that it would promote its own (Viacom-controlled) video sharing site and promote it vigorously. Similar responses from other major media companies could severely harm the site's ability to become a commercial success.
Started by Skype founders Niklas Zennstrom and Janus Friis, Joost has plenty of eyes watching it. Zennstrom and Friis presided over one of the most lucrative acquisitions in history, and there's no shortage of people who believe that the pair could do it again with Joost. The P2P online television service has been tested extensively here at Ars, and most of us are eager to see what the service can do once it is out of beta. Right now, the programming is extremely lacking. This Viacom deal gives Joost a hefty catalog of programming, and the company already has a smaller deal with Warner in place as well.
According to the statement from Viacom, the initial programming offerings will be as follows: MTV will provide shows, both past and present, including Laguna Beach, Beavis & Butthead, Real World, Punk’d and My Super Sweet Sixteen. Comedy Central's offerings will include episodes from Stella, CCP’s and Freak Show. Nickelodeon, CMT, MTV2, Logo, Spike TV, mtvU, and Gametrailers.com will also be onboard. VH1 will offer episodes of Flavor of Love, Surreal Life, and I Love New York. BET’s offerings will include Beef, DMX: Soul of a Man, Comic View and recent smash hit American Gangster.
2/20/2007 12:58:10 PM, by John Timmer
As we've noted in the past, the literature regarding potential connections between violent video games and actual violence is tenuous and filled with apparently contradictory results. In situations where the research is somewhat confused, a meta-analysis, in which results from independent publications are pooled and analyzed collectively, can sometimes identify clear trends within otherwise ambiguous data. The journal Aggression and Violent Behavior has accepted such an analysis for publication; its author, Christopher Ferguson, was kind enough to provide Ars with a copy.
The new analysis breaks down the literature in several ways and identifies a number of potential problems with what's been published. One of the clearest problems identified by Dr. Ferguson is the reliance of many studies on a nonstandardized test for violent tendencies. Many of the studies that showed a link between violent games and actions in laboratory settings rely on a test called the Taylor Competitive Reaction Time Test (TCRTT), in which subjects set voltages of shocks administered to an opponent while receiving similar shocks in return. That test, however, has been around for decades and currently exists in a number of variations. As a result, "the modified TCRTT is employed differently in almost every study conducted with it, and provides multiple possible ways of measuring 'aggression.'" It's also never been shown that aggression, as measured by the TCRTT, is linked to any real-world violent behavior.
There's also an discussion of the statistical analysis used in many studies, in which Dr. Ferguson notes that, although some results linking violent behavior and games report a significant linkage, other statistical measures did not. The confidence interval of those results, a different measure of their statistical strength, appeared likely to include a noneffect, according to the author.
The most extensive part of the analysis explores the possibility of what's termed "publication bias." This bias arises from the fact that it is easier to publish a positive result (something that supports a link) than it is to publish ambiguous or negative data. Although this is a widely recognized problem, the difficulty with this sort of analysis is that it has to cope with the fact that positive and negative results are not necessarily equally likely; some things truly are causally connected. Nevertheless, some of the tests used in this analysis were informative, as in some specific contexts, publication of a single negative paper would have been sufficient to tip the overall conclusion away from one that suggests a connection between violence and gaming.
The overall conclusion is that there does appear to be a connection between violent games and violent thoughts in a laboratory setting. But the connections between such thoughts and violent behavior in the lab or elsewhere are tenuous at best. The studies that try to address those questions currently suffer from a lack of a standardized measure of violent behavior and a lack of sufficient background on other potential influences on the test subjects' tendencies towards violence, such as family environment.
The author makes a number of suggestions on how to improve the quality of the studies in this field and cautions that "it is incumbent on researchers that they not let themselves be side-tracked by a priori hypotheses that may distract the scientific community and the general public from the real biological, social and family influences on violent behavior." Until more extensive data are available regarding how gaming fits into this list of influences, the use of caution in interpreting existing results appears to be necessary.
That sort of caution is widely ignored when legislation that regulates the sale of violent video games is passed. These laws assume that the science linking violent gaming to behavior is conclusive. So far, the judges that have ruled on the constitutionality of the laws have determined that the link is anything but. This paper suggests that these rulings have been appropriate, and makes it likely that any future legislation will face even longer odds against holding up in court.
February 20, 2007 @ 5:12PM, by Nate Anderson
Every few months, someone wonders aloud whether local and vertical search providers will one day kill off Google. These companies offer specialized searches within specific locations or industries; if targeted search results are what you seek, these may provide better answers than a general query. The most recent proponent of this theory is Jason Prescott, who argued last week that "specialty search engines could one day become more important than Google."
Prescott's main point is that general search doesn't always work very well. Using other data, he claims that only four out of ten business professionals are satisfied with the results they get from places like Google and Yahoo. This creates a niche that can be exploited by more targeted search engines that index only specific businesses, for instance, or that concentrate solely on local search.
It's a claim that's been made before, in 2006 and 2005, yet niche search products have done nothing to stop Google's growth or market dominance.
On the other hand, that's not the point for most of them. By its very nature, vertical search is a niche product that will never replace Google's main engine. Instead, the goal is to whittle away at the edges of the Google empire. The obvious danger to the smaller firms is that Google will simply enter the local and vertical search markets with a vengeance if they pick up much traction.
Google has already made these sorts of moves, though still in a limited way. Google Co-op has the beginnings of a vertical search engine orientated around broad categories like "health," and Google Maps has already turned itself into a nice tool for doing local searches.
While these tools (especially Co-op) pale in comparison to other targeted search engines like Zillow, which allows users to search for property values, they do show that Google has its eye on the upstarts. Will they one day dethrone the king?
[ Discuss ]