7 days left in Virginia

A week from today, Becky, Ethan, and I will be heading back west to Searcy, Arkansas so I can start teaching again at Harding University. The condo is sold, the boxes are starting to pile up, and we're getting excited about returning to Searcy.

At the same time, we are going to really miss Norfolk, especially our friends and church which have been such a great blessing to us. We'll also miss living in a neighborhood of beautiful, older homes surrounded by water.

God has really blessed us while we've been here these past three years. Becky was able to find a really great job at Regent University, and my Ph.D. studies have gone really well working with Michael. And now God has given us a beautiful son whose smile will brighten anyone's day. Like they say, Virginia is for lovers.

Citizendium or Wikipedia?

A few weeks ago I applied to be an editor on Citizendium, a new wiki project intended to be a more accurate and reputable Wikipedia. Citizendium does not allow anonymous postings, and they put new articles under a review process. My application was accepted a few weeks later, and a user account was created for me along with a user page listing my brief CV.

I decided to warm up with an article on digital preservation, a topic which I feel qualified to write about. Rather than start from scratch, I imported the Wikipedia article on digital preservation. Although Citizendium frowns upon importing articles from Wikipedia, I had previously written a large portion of the Wikipedia article, so I didn't feel too bad about doing it. I cleaned up the article by focusing the definition, cleaning up the references, and removing the numerous external links.

I didn't spend a whole lot of time editing the article because I got to thinking... can Citizendium really compete with Wikipedia? Larry Sanger (Citizendium's founder and co-founder of Wikipedia) seems to think so. Although I agree that Citizendium's policies in theory would result in more reputable articles, I don't think Citizendium can possibly scale to Wikipedia's size. First, there are a number of people who want to make a minor contribution to a Wikipedia article, a slight correction or clarification, for example. And since they don't have to register, the barrier to entry is sufficiently low enough for them to contribute.

Second, there are a number of people who, for whatever reason, want to remain anonymous or known by some alias. They are not likely to sign on with Citizendium and convey to the world why they are qualified to write about XYZ.

Third, if someone wants to make a contribution to an article on a particular subject, now they have to decide do they make the contribution just on Wikipedia, or on Citizendium, or both? Do they monitor both sites for changes to an article that is important to them? If a Citizendium article is actually better than the Wikipedia article, what is stopping Wikipedia from just importing the entire Citizendium article?

Forth, choosing the name "Citizendium" was a poor choice. It doesn't exactly roll off the tongue.

I really do hope Citizendium takes off, but I think its going to take a very, very long time before they have anything near the number of articles that Wikipedia has to offer. In the meantime, I would much rather put my efforts into something I know is going to rank high in search results and gets far more page views. I'll be watching Google to see if my Citizendium article will ever beat out Wikipedia's (currently ranked number 8) in the search results for "digital preservation". If it ever does, I'll defect to Citizendium.

Fav5

My pick of the week's top 5 events, articles, or items of interest:

1. Researchers at Harvard have developed a robotic fly which could some day be used for spying, detecting harmful chemicals, or just annoying families out picnicking.


2. What comes next after Vista? Windows 7.
   
   
3. Here's an interesting article about Luis von Ahn's latest foray into utilizing humans for AI. The reCaptchas project is brilliant, but what caught my eye was a new game he is developing called Matchin' which enlists humans to rate the attractiveness of people's photos. The results could be used to make image archives searchable by attractiveness. Sorry bon Ahn, but I think this has already been done.


4. A new search engine called Lexxe is using natural language technology to answer user's queries (think Ask Jeeves). I tried the query "why should I use lexxe instead of google?" and got my answer on the first result. Not bad.


5. Becky finally joined Facebook this week. I'm a little jealous though because I've been on there for several months, but she already has more friends than I do. Just a thought, but why does Facebook limit you to just "friends"? What about "casual acquaintances" or "person I saw on the bus"? I guess it's up to me to invent the next Web 6.0 social networking site.

Note to Acer

To the Acer Desktop Design Team:

Guess what happens when you put a power button that activates at the slightest touch on a tower which sits by the operator's knee? You get one very irritated user who feels like hurling your tower out the window.

Regards,

Frustrated in Norfolk

Fourth wedding anniversary

Becky and I were married four years ago today in Searcy, Arkansas. We celebrated today by hiring a babysitter and having a nice lunch at The Trellis in Williamsburg (and their famous Death by Chocolate cake) and then walking around and enjoying our time away.

I cannot thank God enough for providing such an encouraging and loving spouse. I honestly don't think I could have made it through the Ph.D. process without her support. Although I could go on and on about Becky and embarrass her endlessly, I'll just say this: someday I pray our son will find a wife as incredible as his mother.

Cool visualization of the day

I came across this today. It's a figure from The Political Blogosphere and the 2004 U.S. Election: Divided They Blog (2005) by Lada Adamic and Natalie Glance showing a citation network of the liberal and conservative blog communities.


There are 676 liberal (blue) and 659 conservative (red) blogs represented, and the size of each blog point reflects the number of other blogs that link to it. Blue links connect liberal blogs, red connect conservative blogs, yellow links go from liberal to conservative, and purple from conservative to liberal.

Fav5

My pick of the week's top 5 events, articles, or stories:

1. The CACM has several excellent articles on the science of gaming. Once of the articles discusses Carnegie Mellon's masters degree in video games which sounds like something I would leap at in a heartbeat if I was a graduating senior. I also enjoyed Kelleher and Pausch's article on Storytelling Alice which I would seriously like to investigate using at Harding in the near future.


2. While on the subject of gaming, who is preserving our video games for future generations? An interesting article in the Guardian laments the fact that, besides a few game archives (like the new one at Univ of Texas), there is no industry initiative to preserve video games and consoles like the Amiga CD32, Pioneer LaserActive, and Bandai Playdia. The article concludes with a statement that may have been tongue-in-cheek, but there's some truth to it:
   

   The Virtual PC thing is worthy, but are you really telling me that the preservation of obscure census data is more important than saving the software catalogue of the Sega Dreamcast?
   

3. I stumbled across a link to Pagefactor while scanning the Wikipedia article on link rot this week. Pagefactor is attempting to use humans to locate pages that disappear. I'm not sure if the approach is going to work (hmmm... I can't find the web page I'm looking for, oh now I found it, and now I'll go do some work to let others know the new location), but it's interesting nonetheless.
   


4. Six months into Vista, and users are still griping. So am I. I got a new PC in June loaded with Vista, and although I have enjoyed the much improved graphics and speed, I've been annoyed by two blue screens of death, video that won't play smoothly in QuickTime, and constant pop-ups asking if I really meant to do what I asked the computer to do. So why don't I give up and buy a Mac? To quote For Your Consideration:

   You can't throw the baby out with the bathwater because then all you have is a wet, critically injured baby.

   Actually there are two main reasons: 1) I've invested an incredible amount of time learning Windows and don't want to start over, and 2) I want to be running what most people are running. It's embarrassing be asked a computer question (by family, friends, or students) that I know nothing about. And as a CS faculty member, I should be very familiar with how the majority of the world is interacting with their computer.
   


5. This week some masked bandits cracked a safe and stole $12,000 with the help of Google. After struggling with opening a safe for over an hour, the buglers went to a nearby PC and Googled "safe-cracking" which led them to an article on "How to Open Safes". Soon they were in and left with $12,000.
   

Warrick queue is growing

Last week I wrote about the new web interface for Warrick. Well apparently the interface has made Warrick a very popular tool, because there are currently 18 jobs in the queue, 5 jobs currently running, and 16 completed jobs. Unfortunately, the jobs currently being ran are huge, most of them with more than 10K resources that need to be recovered. One job has over 136,000 URLs needing to be recovered!

I upped the daily queries to the Internet Archive from 1000 to 2000, but the changes only get read the next time Warrick is ran, so the current jobs can't be sped up. Not sure what to do but hope some of these jobs finish soon... there are possibly numerous web pages falling out of search engine caches each day for those jobs waiting in the queue!

I guess it's better to be too popular than the shy kid in the corner of the room.

Oh... Ethan slept 8 hours straight last night! Not bad for a 3.5 month old!

Ethan update

It's 7-7-07, so I had to post something on this statistical anomaly of a day. Here's a few pics of Ethan from the past month.

Looking good in my golf outfit!	With grandma and mom at church
With dad and grandpa	I'm not amused
Happy with Mom

Fav5

TGIF! My pick of the week's top 5 events, articles, or items of interest:

1. The Internet Archive just increased their web archive (the Wayback Machine) by 25%. Brad Tofel, an archivist at IA, told me that from here on out the Wayback Machine should be 3 months out-of-date rather than 6-12 months like they have been in the past.


2. It's a great time to be a talented college student looking for an internship.


3. Some brilliant blogging advice from one of the most prolific bloggers at Google:
   

   Don't post when you're angry.

   These and other wise sayings from Matt Cutts were in response to a Google employee's blogging about her dislike of the new movie Sicko. I'll add to this...
   

   Don't post a response to someone else's blog posting when you're angry.

   I've learned this lessons the hard way.


4. Dell is warning their customers about moving from XP to Vista. I recently made the move a few weeks ago when I purchased a new desktop for home use. It hasn't exactly been the smoothest transition, but I'll blog more about that later. I would certainly advise anyone making the leap to be on their guard.


5. If you haven't played the Wii yet, find a friend who owes one and give it a shot! On the 4th our friends Brad and Katie threw my family a going away party (we're heading back to Arkansas in Aug). Brad owns a Wii, and it must have been the most popular activity at the party! No wonder the Wii is currently out-selling the PS3 six-to-one.
   

Link rot irony

What do you get when you do a search on Google Scholar for the paper titled

Runaway Train: Problems of Permanence, Accessibility, and Stability in the Use of Web Sources in Law Review Citations

(a paper about how problematic link rot is) and then try to access the link? Irony at its best.

Finally, a web interface for Warrick

Warrick finally has a web interface. Actually it's more than just an interface, it's a queueing system that we call Brass. You can read more about Brass in a paper I presented at IWAW 2 weeks ago. We're hoping the interface is intuitive enough for those who are not very technically inclined to be able to recover their lost website (or someone else's lost website) without too much difficulty.

We had a masters student (Amine) working on Brass for quite some time, and over the last few months I've been overhauling the interface and tightening up the security. (I'm sure there's still some holes in it, but it's much better than it used to be!)

Below is a screen shot of the opening interface.


After submitting the requested info, the user is sent a confirmation email (to avoid recovering websites for bots). Once the user confirms the request, the job is queued and later dispatched to a free machine. The user is emailed when the job completes, and then the user can download the recovered set of files from the lost website.

Go ahead and give it a spin, and let me know if you have any problems.

*By the way, the really cool Warrick logo was created by two graphic design majors at Harding: Andrew Murray and Luke Jones.

Save your Yahoo! Photos

I just received this email yesterday from Yahoo! Photos, telling me that my photos were going to disappear unless I took action. I couldn't help but wonder what if I my spam filter had eaten the message, or what if I wasn't using this email account anymore?

Considering that just 57% of individuals actually backup their personal data, how many people do you think are depending on Yahoo! to preserve their photos indefinitely? You can image the horrible feeling of logging in and realizing all those photos of your newborn child have ended up in the big bit bucket in the sky. Sigh...

Dear Yahoo! Photos user,
...

We will officially close Yahoo! Photos on Thursday, September 20, 2007, at 9 p.m. PDT. Until then, we are offering you the opportunity to move to another photo sharing service (Flickr, KODAK Gallery, Shutterfly, Snapfish, or Photobucket), download your original-resolution photos back to your computer, or buy an archive CD from our featured partner (for users of the New Yahoo! Photos only). All you need to do is tell us what to do with your photos before we close, after which any photos remaining on Yahoo! Photos will be deleted and no longer accessible.

...

Please give us your decision by Thursday, September 20, 2007, at 9 p.m. PDT. After that time, any photos remaining in Yahoo! Photos will be deleted. Click here to make your decision, or review a list of our frequently asked questions.

Fav5

It's been a few weeks since I've posted a Fav5, but now that I'm through with traveling, I'm back at it.

1. Internet music is about to die. Thursday marked a day of silence in protest of the absurd fees that must soon be paid by Internet music providers. Very sad indeed.
   


2. The iPhone is finally here. Being the traditional late adopter that I am, I don't really give a hoot, but it's entertaining to see all the technophiles go crazy over it.
   
   
3. An interesting article in the NY Times discusses social searching and the new searching paradigms that may challenge Google's lead in the Web search field. Matt Cutts says (unofficially) that Google is not opposed to leveraging the wisdom of the crowds in the future to stay in the lead.


4. Researchers at Northeastern University have demonstrated the minimum number of moves needed to solve a Rubik's Cube is 26.


5. I came across Citizendium this week. It's essentially a Wikipedia where only the educated and pre-approved can edit articles. I submitted my credentials and am waiting to hear back. In theory, Citizendium should be of higher quality than Wikipedia, but I think the barrier-to-entry is just too high to make it a serious contender. Besides, sometimes anonymous contributions can be useful.

IWAW 2007

The International Web Archiving Workshop was held at JCDL this year (normally it's held at ECDL). Our research group at ODU was well represented with 3 presentations. I presented our new Warrick queueing system called Brass (slides are available here). This was the second time I attended, the first time being in Vienna 2005.


There were some excellent presentations by a number of people including the Internet Archive, UNC, CDL, and others. One that I found really interesting was about Preserving 2008 Presidential Election Videos (work from VidArch) where they have been archiving YouTube videos using their TubeKit. Overall I really enjoyed the workshop and hope to attend again.

I'll be heading home tomorrow!

JCDL 2007 - days 3 & 4

The conference opened on Thursday with a keynote from John Willinsky (Univ. of British Columbia). Amazingly enough, John gave an entertaining talk without the aid of PowerPoint. John focused primarily on issues surrounding the open access (OA) movement, exhorting the digital library community to take the lead. At one point in the talk, John gave a shout-out to Marko's poster, using the metaphor of "fingerprints" that define a variety of ways scientists leave their marks on the scientific world.

I attended several sessions, but one of the more interesting presentations I attended was on a paper called Measuring Conference Quality by Mining Program Committee Characteristics by a group from Penn State. The authors basically divided a large list of conferences into reputable and non-reputable by examining the quality of the conference's PC members. To verify that some of the conferences that were judged non-reputable were indeed non-reputable, they created 3 bogus papers using the SCIgen software which was used in the MIT prank. The 3 papers were submitted to 2 conferences, and 2 of the 3 papers were accepted! There were obviously no reviews returned for the papers.

I also enjoyed Cathy Marshall's presentation on The Gray Lady Gets a New Dress: A Field Study of the Times News Reader. Cathy's presentations are always filled with interesting photos and stories.

On Friday I gave the first talk in the User Studies and User Interfaces session on my paper Agreeing to Disagree: Search Engines and Their Public Interfaces. I got some really positive feedback, and was turned onto some related work by Jamie Teevan.

Right after my talk I caught Marko's, but I had to skip out on a few afternoon talks in order to complete my slides for tomorrow. I did attend the closing ceremony- looks like JCDL is going to be in Pittsburgh next year, and the conference dinner is going to be held on a boat! Should be another great conference.

JCDL 2007 - day 2

This morning Daniel Russell (Uber Tech Lead for Search Quality & User Happiness at Google) gave a fantastic keynote address entitled What Are They Thinking? Searching for the Mind of the Searcher. Daniel spoke about the great amount of research Google does to figure out how users use Google to search. Here are a just few points of interest from Daniel's talk:

• For North America, the average user performs 9.4 searches a week.
  
• 60% of users perform one Google query or less a day.
• Only 10% of users go to the next page of results.
• Half of all clicks to the Advanced Search page result in the user abandoning their search, most likely because they are overwhelmed by the Advanced Search interface.
• Most users look at 3 results before clicking on a result.
• Many users "teleport"... instead of using Google to make a search (like searching for a flight), they search Google for a website from which they can then make their desired search.
• The more query terms used in a search, the longer amount of time the user will spend examining the search results.
• Most people don't have a clue how Google works, and they don't really care. For example, many individuals Google only examines the first couple of lines of text from a page.
  

After Daniel's talk and a break, I presented my first paper, Factors Affecting Website Reconstruction from the Web Infrastructure. (You can see my slides here). Joan presented after me: Generating Best-Effort Preservation Metadata for Web Resources at Time of Dissemination. Both talks seemed to go really well, and we both got some positive feedback.

This evening I attended the Minute Madness where a representative from each of the demonstrations and posters got a minute to advertise their demo/poster. The poster reception was held immediately afterward, and Marko won top poster again (that punk won top poster last month at WWW07)!

I think the most interesting poster I saw though was Blogger Perceptions on Digital Preservation, a study conducted by UNC. The researchers performed a survey asking bloggers questions like, "What would happen if your blog suddenly disappeared?" Their work coincides with mine- I recently coauthored a paper where we asked individuals who actually lost their websites questions like "How did losing your website affect you?"

JCDL 2007 - day 1

I'm in Vancouver at JCDL 2007. I flew in yesterday with Martin, and the rest of the ODU posse flew in today. Between the 3 of us Ph.D. students, we'll be making 6 presentations at JCDL and IWAW this week.

This is my first trip to Vancouver, and I must say it's one of the most beautiful cities I've been to. Just wish I had Becky and the Bean here with me to enjoy the beauty. The photo on the right shows what the view looks like from my room at the Westin Bayshore hotel.

This morning I joined Kris Carpenter and Brad Tofel of the Internet Archive in a tutorial about researching the Internet Archive. I shared some of my research with Warrick and a recent study analyzing IA overlap with search engine caches. (You can see my slides here.)

I learned a lot about IA from Kris and Brad. Here are just a few items of note:

• The IA currently has archived around 96 billion resources (html, pdf, images, etc.), or about 1.9 petabytes of data, 51% of which is unique.
  
  
• Although the Archive's holdings are 6-12 months out-of-date, beginning July 1, the Archive will receive updates on the first of each month and will only be 2 months out-of-date.
  
  
• The Archive is currently working on adding full-text search to its contents from 1996-2000 in a project called 20th Century Find. (No URL yet.)

Tomorrow morning I'll be making a presentation about a Warrick experiment, and Joan will be presenting a short paper after me. Looks like I'll be burning the midnight oil getting prepared for tomorrow...

Ethan 2.0

Wow- what a busy week. I’m getting ready for JCDL 2007 in Vancouver where I’ll be presenting two papers, one on a large website reconstruction experiment I performed several months ago and another on the search engine APIs. I’ll also be talking about Warrick at the Internet Archive tutorial and the web archiving workshop. Plus I’m doing my best to get Warrick’s queueing system on-line so I can demo it at the tutorial. Note to self: always go with PHP over Java servlets when using Apache!

My parents are coming into town tomorrow to see Ethan. They got to spend a little time with Ethan 1.0 the first week he was born, and now they’ll get to see Ethan 2.0, the smiley, attentive, and cute-as-can-be Ethan. Just in case you haven’t seen enough photos of Ethan yet, here’s some more from the last month:

Pooped-out	With Mom
Now you're in trouble!	Just kidding!
What, me scared?	Catching some TV
Hey!	zzzzzzz...
Happy boy	Play ball!
Double chin action	Passed out

Fav5

1. Thanks to Mike Baur for sending me a link to this video of some really cool image visualization software being developed at Microsoft... I especially liked the aggregation feature which pulled images from Flickr together to form a complete image. If you thought the video was cool, you may also be interested in these other videos from the TED conference.
   


2. Vote for your favorite 1980’s arcade game. Although I genuinely loved most of the games on the list, I had to go with Tron.
   


3. There’s a good article from the New York Times about Google’s continued focus on search: Google Keeps Tweaking Its Search Engine. Matt Cutts also put in his 2 cents on about Google’s search focus.
   


4. Researchers are finding that bullying in the virtual world is becoming quite a problem, especially for newbies.


5. Ask.com has received a makeover. Ask3D is their new approach at integrating various vertical search results together in a single results page. Nice.