Techfoot

Speaking of Scott Nelson…..

I was on a task force studying digital imaging with Scott and he often talked about the value of of a “digital dogpile” as a collection of high quality images that could be freely accessible to members of the community who needed them to enhance communications. The reference generally made me a bit uncomfortable–sort of like, “darn, I need a scraper; I stepped in the digital dogpile.”

Actually, the term has a (somewhat) more refined etymology–most commonly used in the old Usenet days days:

When many people post unfriendly responses in short order to a single posting, they are sometimes said to “dogpile” or “dogpile on” the person to whom they’re responding. For example, when a religious missionary posts a simplistic appeal to alt.atheism, he can expect to be dogpiled. It has been suggested that this derives from U.S, football slang for a tackle involving three or more people.

(dogpile. (n.d.). Jargon File 4.2.0. Retrieved September 20, 2008, from Dictionary.com website: http://dictionary.reference.com/browse/dogpile)

My own personal dogpile is the unsorted collection of index cards, magazine articles, photocopies, books and other artifacts that have seemed important enough to me to save, but not important enough to actually do anything with. A recurring fantasy of mine is that I actually make my way through that basket of stuff and figure out why it is that I put it in the pile in the first place.

Dogpile is also the name of an aternative search engine that attempt to aggregate information from multiple search sources to create a single view of the results. While I still use Google most of the time, I give dogpile a shot every once in a while. They explain the origin of the name this way:

Oh, and the name Dogpile?

Well, that’s a funny story. You see, we love Rugby. It’s traditional in Rugby for players to come together and pile on one another. This is exactly what Dogpile metasearch does-it brings together the best results from the Internet’s top search engines, including Google, Yahoo! Search, Live Search, Ask.com, About, MIVA, LookSmart, and more.

Photo Credit: http://www.flickr.com/photos/soggydan/
Creative Commons Attribution License

Like many academics, I think that many security policies and procedures are a tad draconian and based on superstition rather than evidence. One of my pets that I often rail about is the requirement that individuals change passwords on some fixed schedule; I’m still looking for any evidence that these requirements actually make our institutions more secure. In my own case, I’m much more likely to try skimp on password complexity or write the new one down in those cases where I’m forced to change.

Every once in a while, though, I get a graphic reminder of why folks with more daily responsibility for security are more paranoid (which may not be too strong a word) than I am. Several of those reminders have been delivered this week as faculty and staff have been hit with a barrage of phishing schemes. At least seven members of the community, including at least a faculty member or two, have succumbed and provided their userids and passwords. Almost immediately their accounts were attacked by zombie armies, hundreds of sessions were opened and hundreds of thousands of spam messages were generated.

A Botnet (also known as a zombie army) is a number of Internet computers that, although their owners are unaware of it, have been set up to forward transmissions (including spam or viruses) to other computers on the Internet. Any such computer is referred to as a zombie – in effect, a computer “robot” or “bot” that serves the wishes of some master spam or virus originator. Most computers compromised in this way are home-based.

In these seven instances, millions of messages were generated. Cleaning up the resulting mess takes lots of engineering time–though unfortunately with practice we’re cutting it from days to hours. Mail response for local users slows dramatically and huge internet service providers like AOL and Comcast blacklist the college domain as part of their spam management process. Reopening delivery may take a couple of days and untold amounts of mail from college addresses may be dumped to the bit bucket.

If you care about your colleagues and being a good citizen of the community, don’t provide your id and password by using a link in an email message.

Humanities High Performance Computing: “”

For the last three weeks I’ve been immersed in the world of HPC–High Performance Computing. HPC is that parallel universe where researchers run programs that take five days of processing, where tiny jobs only require 12-15 processors, where terabyte drives fill up in matters of hours and where shouting at and threatening colleagues is considered a perfectly acceptable way of communicating. Now humanities scholars are being invited play in the HPC sandbox too.

The NEH Office of Digital Humanities has just launched a resource page for Humanities High Performance Computing. This new resource is designed to attract scholars in the humanities and social sciences who have masses unstructured data that needs to be sorted, mined, or visualized to be better understood. Programs include a series of grants (deadline is July 15th for award in January 2009) and invitations to access to the National Science Foundation’s teragrid.

William and Mary has a HPC operation that has recently become a part of our academic and research support for faculty. Like the folks at the NEH, we’re hoping that a broader range of faculty will take advantage of the college’s investment in these high performance tools.