ten most persistent design bugs... or, how many top ten lists can we make

Bruce Tognazzini fuels my speculation that usability pundits have an obsession with creating top ten lists of design flaws, heuristics, etc.

All sarcasm aside, Tog has captured some fun design flaws in his latest AskTog posting.

cameron sums up necessity of discovery

Director (and new Guest Editor of Wired Magazine) James Cameron shares his views on the frontier of outer space and the need for cultural change at NASA. Read on.

google gets scholarly

Yay! Google has released Google Scholar, a tool for searching scholarly materials such as refereed journals and conference proceedings. For example, here's a search for tree visualization.

gaim rocks

Our team at work used to coordinate production deployments using MSN IM. When we got our new PCs and laptops, we were "liberated" of the need to do administration on our boxes (read: we were stripped of admin rights). MSN IM is disabled so I've been unable to use it for deployments ever since.

Then, I found Gaim. Gaim requires no admin rights and no registry additions on a Windows box. In addition to MSN IM, Gaim also supports Yahoo IM and AIM. Viva la Open Source.

little hard drives

I bought a Lexar JumpDrive this past Friday. I love the idea of having a mini hard drive on my keychain. Even better, I like being able to carry my Masters research materials and resume/CV with me wherever I go.

I often wonder if, in my lifetime, I will see a race to make computing a more pervasive and hidden part of everyone's lives. Technology has become so ingrained in my daily life that I ponder a day when I will go about my business with hundreds of tiny computers doing my bidding both in an explicit and implicit (i.e. predictive) fashion.

cleverness in slashdot sigs

"He uses statistics as a drunken man uses lampposts...for support rather than illumination." - Andrew Lang

I discovered this quote as the signature (sig) to a post on Slashdot. My sig, by the way, is favorite quote of mine made by Dijkstra.

"Computer Science is no more about computers than astronomy is about telescopes." - E.W. Dijkstra

new ian photos

There are new photos of Ian from Halloween and early November over on Ophoto. Highlights include Ian in his sheep costume and various milk-drooling pictures.

sql: rows in one table but not in another

This comes up a lot in database programming. How do you write a SQL query that selects rows that are in one table but not in another table? Here's the answer:

select a.field0
from table0 a
where not exists
        (select b.field0
         from table1 b
         where b.field0 = a.field0)

cyclones win

Pigskin nirvana? You bet.

design principles in infovis

In an attempt to organize my thoughts, here are several principles to consider when designing visualizations for specific task types.

1) Information scent - How does the user know how to find the information he seeks? What directs his browsing choices? How does he know what to look for? How does he know when he's found it?

2) Cost structure - How do we optimize the cost of knowledge access? How do we make commonly needed information immediately accessible to the user?

3) Information density and the useful field of view (UFOV) - What is the locus of attention? How much information can be processed within a given area? When looking at a tree of information, the UFOV depends on the nature of the task. If the task is to answer a question about the tree's structure, the UFOV might be the whole tree or a certain sub-tree. If the task is to answer a question about a tree node's attributes, the UFOV should be centered on the node itself.

4) Navigation/browsing - What actions does the user have to take to retrieve information? Is this done in the simplest way possible? Is the act of navigation too intrusive? Does it strain the user's cognitive abilities? Is the user forced to recall information, or can they simply recognize it in the interface?

third CS monthly departmental colloquium @ uno

Dr. Victor Winter will be giving a talk at this month's UNO CS colloquium entitled "Higher-Order Transformation and the Distributed Data Problem." It will be held at 12:30pm at PKI-279. Check it out if you're able; Dr. Winter's research areas (program transformation and high-assurance computing) are interesting ones.

It's 12:38 AM and I'm still at work jockeying around with the backfill processes. Pushing around hundreds of thousands of large records is hard work. :-)

tonight's big release

We are releasing the latest version of our application tonight. We have implemented a new security model that is very fine-grained and flexible. It will enable us to "turn on" new user types in a way that is (for the most part) purely data-driven.

From a release management standpoint, tonight's release is a behemoth. The new security model involves a multitude of code and database changes. In addition, there is a lot of data we are now summarizing, so old data must be backfilled appropriately. I have written a cool little data-backfill utility with a GUI front-end for starting and stopping the update threads as needed. It has several configurable parameters that allow the backfill program to rest at certain points (so the database can take a "breather").

It should be a very interesting night.