David Duey

Email Triage

I’ve never given much thought to improving my email system until I watched a video of a lecture on time management given by Randy Pausch where he lambasted email systems.  Everyone is concerned about filtering spam, but amazingly it seems like little has been done to filter and categorize all incoming email.

Microsoft has an interesting research project named Snarf (I like that name) that addresses some of the email triage issues.

SNARF was built around the notion that social network information that is already available to the computer system can be usefully reflected to the user: a message from a manager might be seen differently than a message from a stranger, for example. SNARF applies this idea to email triage: handling the flow of messages when time is short and mail is long.

Other good email triage information can be found here.

Crunchies: Best Design

I took a look at the sites nominated for the Crunchies Best Design award; most of the nominated sites seemed to be relatively simple and nondescript.  Netvibes has a fairly complex UI, but it's not overly cluttered. 

I think that's a good thing.  I'm not a UI or usability person, but I sometimes obsess over the user interface.  I like clean, uncluttered designs.  In fact, I like white.  I like plenty of white space.  I like a very focused and simple user interface. 

Here are the Crunchie Best Design nominees:

• Etsy
• Jackson Fish Market
• Netvibes
• SmugMug
• Songza

The Changing Face of the User Interface

A few days ago, after reading a TechCrunch article about Microsoft Tafiti, I played around with Tafti for a short time.  It has a really cool and impressive interface.  I don’t think Tafiti is something I’ll be using on a regular basis, but I think it's a good example of the direction that user interfaces will be going in the near future. 

  Based on my experience, hardcore techies often seem to be indifferent (at best) when it comes to the user interface.  During the early to mid 90s, I remember the shift from DOS character-based applications to the graphical user interface (GUI).  At the time, it seemed like the technical people preferred the character-based applications over the GUI.  The techies’ argument was that Windows required too much overhead to run (it did) and that DesqView (or whatever favorite TSR utility) was a superior environment.  But, users obviously embraced the richer environment of the GUI. 

During that period, I remember some of the software companies seemed to make the shift to Windows/GUI very slowly.  WordPerfect and Lotus 1-2-3 were slow to change and ultimately released buggy Windows versions of their software.  I think part of their failure in the marketplace was because they didn’t move quickly enough to improve their UI.

I think things are a little different now that we’re in the internet era; we have a legion of web designers with a specialized skill set that is a mixture of a good sense of aesthetics, art, usability and technology.  With web applications, I don’t think we’ll see the huge shift in UI paradigms like we saw when we moved from character-based applications to the GUI, but I think we’re going to be seeing a new class (for lack of a better description) of web applications appear that are based on Adobe AIR, JavaFX, and Microsoft Silverlight.  AJAX has gone a long was to making web applications more appealing to use, but the new tools for building rich internet applications are taking web UI far beyond the clunky realm of HTML, CSS and JavaScript.

I think there’s currently a lot of opportunity to take some of the applications of years past (desktop or web) and update them with a new and better user interface along with some collaboration/networking included courtesy of the internet.  Some examples are a web version of WinZip or possibly a screen snapshot utility.  I can’t wait to see the next round of rich internet applications.

Visualizing Time

I’ve been discussing a UI/programming issue with a colleague and an interesting question arose:  “If you could visualize time, what would it look like?” 

It may seem like a dumb question, but I think it’s a good question.  Of course, we have calendars, which deal well with days weeks and months, but we have to deal with time of day, days, weeks and months over a long period of time.  Graphs and charts are fine for viewing aggregate data, but we also need to allow user input. 

I did a quick search on the internet (nothing exhaustive yet), but I didn’t find anything new or unique.  It may simply be a dumb question, but I wonder if there’s not a better alternative to a calendar, particularly when viewing, entering and editing data over long periods of time.  If you could visualize time, what would it look like?

Decade Old RIA

Ten years ago, when I was working for one of the big six accounting firms and one of the senior partners was working on a proof of concept for an online “accounting toolbox.”  I was selected to help him develop a prototype.  We’d decided to use Microsoft’s new ActiveX Documents to develop the prototype.  The cool thing about ActiveX Documents was that they were relative easy to code; it was just like using VB form and you code them in the VB or Visual Studio environment.  Also, ActiveX Documents weren’t limited by the standard HTML forms. 

Now that I look back, ActiveX Documents were really a predecessor to today’s rich internet applications (RIA).  It was all new to me at the time, and I thought it was really cool.  I created a user interface that was, in my own opinion, much better than the typical web site of that era.  Of course, there was a huge downside to ActiveX Documents; they required IE and Windows, and more importantly, ActiveX Documents were COM components.  I liked COM, but downloading and installing COM components was always a problem, and was very fickle (too fickle, as turned out, to use for a production application). 

At about the same time that I was developing the proof of concept, I went out to San Francisco for a VBITS conference and I heard VB superstar Dan Appleman speak in one of the sessions.  He said that he didn’t know where ActiveX Documents where going, it might be a good technology, but he seemed to think it might not go anywhere.  I remember being very disappointed in his lack of enthusiasm about the technology.  Of course, shortly thereafter, I learned from experience; it was a cool technology, but trying to get it to work consistently even in a relatively homogenous environment proved to be impossible.

I’m glad to see some of the RIA stuff finally taking root; MS Silverlight and Adobe’s Apollo will hopefully take user interfaces to a new level (beyond Ajax) and will hopefully avoid some of the issues that arose with past attempts.