Wednesday, May 31, 2006

CNN - Making Cell Phones Simple is Hard


As the universe of people who want a cell phone and don't already have one gets smaller, wireless carriers are counting on advanced services to generate the bulk of new revenue in coming years.

...

That has providers working hard to make their devices easier to use -- fewer steps, brighter and less cluttered screens, different pricing strategies -- so consumers will not only use data functions more often but also be encouraged to buy additional ones.

...

"IPod was not the first MP3 player on the market, but once they figured it out (the user interface), they became the predominant one overnight"


http://edition.cnn.com/2006/TECH/biztech/05/29/cellphone.usability.ap/

Wednesday, May 24, 2006

SATs poor predictors of performance?

From the Chronicle of Higher Education
http://chronicle.com/jobs/news/2006/05/2006052401c/careers.html?top20


According to Human Capital Research, a college-admissions consulting firm, rated on an index from zero to one, SAT test scores predict a freshman's grade-point average at 0.03 to 0.14. "I might as well measure shoe size," the firm's president, Brian Zucker, was quoted as saying in "The Best Class Money Can Buy," one of a series of thoughtful articles on the admissions process that appeared in the November 2005 issue of The Atlantic Monthly. Zucker argues that the SAT's have "made schools lazy and stupid at the same time." The opportunity costs involve "looking past literally millions of kids who would do a great job."

Wednesday, May 17, 2006

Java 2 AJAX

Google has just released a web toolkit that lets you convert Java GUIs to AJAX.

I proposed something like this as a class project for SAUI (Software Architectures for UIs), and even thought about doing this as a long-term research project. I'm now glad I didn't because I would have been seriously scooped.

http://code.google.com/webtoolkit/

Whole Food Watermelons

Whole Foods' watermelons are simply amazing, the best I've ever had. That is all.

Friday, May 12, 2006

Rainbows End

I'm about halfway thru Vernor Vinge's new book Rainbows End. What's really amazing is how he ties together a lot of technology trends in a single coherent story, including:


  • wearable computers with contact lens displays
  • sensor networks
  • augmented reality
  • robotics
  • new biomedical therapies (I'm sure there's a sexier name for that)
  • silent messaging (I'd love to know how people could do text entry that fast!)
  • paper displays
  • autonomous vehicles (self-driving taxis)
  • micropayments
  • holographic displays
  • digital libraries


There's a funny section early on where the protagonist, who has missed the last decade due to Alzheimer's, is asking where all the robots and flying vehicles are. Vinge also does a good job of tying things to our present world by forecasting how existing entities such as Bank of America, Google, and Homeland Security fit in.

Vinge also seems to subscribe to Brin's notion of a transparent society, where everyone can easily find almost any piece of information about another person. There was a frightening section about on-demand information retrieval, where you simply stare at a person to initiate retrieving a person's biography, and can drill down on past speeches and current hobbies of that person (from Google no less!)

One thing that really bothers me, though, is that the book is supposed to be set in 2025, just 19 years from now, and it just doesn't seem plausible that all of these advances would happen, seamlessly mesh together, and be economically feasible in such a short timespan.

It's also not clear to me that people would accept wearable computers so easily, that our educational system could be so radically restructured (elders and school children in the same class?), as well as the dramatic loss in privacy (though given today's polls that most people support the NSA wiretapping, maybe I'm wrong).

At any rate, it's a really fun read, and really underscores the question of where we are going, and whether that's something we want and can be proud of.

Wednesday, May 10, 2006

Anatomy of a Nigerian Email Scam

Very interesting look at Nigerian email scams, takes you thru one person's story (who has a PhD in psych, no less).

http://www.newyorker.com/fact/content/articles/060515fa_fact

-------

The scammers, who often operate in crime rings, are known as “yahoo-yahoo boys,” because they frequently use free Yahoo accounts. Many of them live in a suburb of Lagos called Festac Town. Last year, one scammer in Festac Town told the Associated
Press, “Now I have three cars, I have two houses, and I’m not looking for a job anymore.”

...

When I asked Worley what he wished he had done differently, he didn’t answer directly. Instead, he spoke about hoping that the Abachas would get back in touch with him. However, before they could resume work on the multimillion-dollar transfer, he expected them to send the six hundred thousand dollars that he needs for restitution.

Tuesday, May 09, 2006

Phishing Scams up to $1000

All right, the phishing scams have moved up from $20 for surveys to $1000! Now that's progress!

-----

• The Washington Mutual Bank Online department kindly asks you to take part in our quick and easy 5 questions survey. In return we will credit $1000.00 to your account - Just for your time!
...

Web 2.0 and Research

I've been chatting with many of my friends and colleagues about an issue that's been bugging me for a while, namely whether academic research has any role to play in the emerging Web 2.0. I've been slowly coming to the conclusion that the answer is not much.




I had a similar discussion with other researchers at HotMobile a few weeks ago. When the web first came out, pretty much every systems researcher ignored it because it was so ugly. The web was not very sophisticated in terms of distributed systems, HTTP lacked elegance, HTML conflated many different ideas, and so on. There were also not any really new ideas with the web, as evidenced by the fact that Tim Berners-Lee's first paper on the Web was (probably rightfully) rejected from an ACM conference on hypertext.

I'm sure one thing that really irked researchers about the nascent web was that it completely ignored the large body of work in hypertext and distributed systems that had preceded it. Even in 1997, as the web was rapidly expanding and well after Netscape's explosive IPO, Infocom (one of the leading conferences on network communication) only had one paper about the Web. However, by this time, it was already too late, and the Web had taken on a life of its own.

The main lament here is that, if only we researchers had engaged with the early developers of the Web, we could have avoided many of the problems we face today. I'm not entirely convinced of this, however, since researchers really like to explore the full design space of things and be highly rigorous rather than letting things be ugly and good enough. (But that's another rant for another day.)

(Another fun one I hear often from computer science researchers is, why didn't we invent the web? I've heard this from digital library people, systems people, and HCI people. It's also funny to see how many books and articles I see from researchers saying that they had anticipated the web... No, you didn't, otherwise you would have gotten it out first. It's like the number of people that worked on the original Macintosh: for some reason, the number seems to keep increasing with time.)




So the question comes up again: is there any role for the research community for Web 2.0? I'm increasingly thinking that the answer is no, because the cultures, goals, and incentives with these two communities are far too misaligned.

Most of these Web 2.0 web sites are from small startup teams that care about making a successful product that lots of people use. They have the time, money, and resources to engineering that research teams do not. Most Web 2.0 teams also don't care about novelty, but rather the best implementation of something. For example, when some developers felt that del.icio.us sold out by going commercial, they just set up a clone site called del.irio.us (though the site seems to be down now).

There's also no incentive for web sites to do fully rigorous evaluations like you would see in academic papers, because there's no time, resources, or credit for doing so. Likewise, there's no reason to publish papers at all, as it might help your competitors.

Lastly, it's not immediately clear to me what research papers could be published in this domain. Pretty much every single Web 2.0 interaction technique was done decades ago on desktop computers, the only thing that's new is being able to do it in a web browser with Javascript and XML. There might be some new things, like tools for helping with development, but it all seems well within the state of the art, and is a standards issue rather than a research issue.

There could be papers published about how people use it and the community that develops around such web sites, but that seems less about anything in particular about Web 2.0 and more about the general usefulness and utility of the web site.




So, in summary, I think that the research community will have little to directly offer to the emerging space of Web 2.0 apps, but may have some things to contribute with respect to evaluating and understanding how people use these kinds of apps in the wild and how to improve the user experience. You know, the stuff that we've already been doing at CHI.

Now, I just have to figure out what I'm going to teach Web 2.0 in the Software Architectures for User Interfaces (SAUI) class this fall.

Six cool Quicksilver plugins you might not know

I can't take any credit for it, but two of the six that this person lists were developed as part of the Software Architectures and User Interfaces class I taught last Fall.


Just a quickie to bubble up some novel Quicksilver plugins that are new-ish or even a bit esoteric.

...

3. Abracadabra triggers - Associate a mouse gesture with any command. Seems especially cool for people using pen and tablet. (more)
4. Constellation Menus - Love it, hate it, or just find it wildly geeky and beautiful, this is kinda A1c0r’s version of right-click on graphical steroids. (more and more)



http://www.43folders.com/2006/04/13/qs-plugins/