Tuesday, October 07, 2003

Bad User Interactions or Why Geographic Information is Important
The web is fascinating, chaotic, confusing and filled with examples of bad user interaction design.

In the days of mainframes, hierarchical menus were the bane of users. And if you have forgotten how bad they are, all you have to do is look over the counter of reservation agents at the airport or bank tellers. How many times have you heard the new employee asking an older employee: "Is it pF7 followed by PF3 to look up the account balance?" or some equally obscure key combination?

A classic problem that has reappeared on the web is the need to descend hierarchies to find what you are looking for? The problem exists everywhere. It takes numerous clicks to descend Yahoo's hierarchy to find a suppplier. And frequently you have to back up to go to a different leaf of the tree.

And from a use-case perspective, it's even uglier if you want to find two suppliers. Then you have to descend to two different leafs at the end of different trees.

In a more competitive Internet, where quality of user interaction is important, more companies are going to have go the route that firms like iiMap (www.iimap.com) have gone. (In the interests of disclosure, I sit on the company's board of advisors.) They start off with the question, "What does the user want?"

In many cases, the answer is that the user wants to know about location. Where is my closest supplier? What is near me?

In other words, I am here. What's nearby? This is revolutionary for the Internet, but actually it's returning user interaction design back to the early days of windowing. The breakthrough represented by windows and graphical interfaces, made popular by PARC, Apple, and eventually Microsoft, was that the grammar of interaction got simplified.

Object -- Verb - Action

I could point to a location, get a list of verbs or functions and immediately see the result.

Contrast the grammar to

Verb - Verb - Verb- Verb - Verb - Action - See result

in a hierarchical menu scheme on a mainframe or in Yahoo

Category - Category - Category - Category - See result

About the only simplification represented by Yahoo is the almost universal understanding that clicking or double clicking causes something to happen.

In a world where interactions need to minimized for commercial, customer satisfaction or user interface constraints (e.g. with cell phones or PDAs), simpler interactions are needed.

I am here.
I want a restaurant.
Show me what's close.

might be a prototype of such an interaction.

Interactions can be voice, keyboard or pen driven, but behind the interaction is the need to use geographical information to link information stored in other ways.

Alistair Davidson

Thursday, October 02, 2003

The Future of Data

Data used to sit in islands of information.

Today networking has caused organizations to want to connect information. Much of information management is concerned with managing the portfolio of information sources, processes, projects and outcomes of old and new systems and changes.

Middleware, object oriented programming, software and design reuse, integrated ERP packages, portals -- all target the objective of tying together information in many places. Sometimes the tying together is done in a central database or information repository. Sometimes it is done dynamically and on the fly using newer middleware technologies such as application servers.

But there is another parallel trend occuring - the increase in the amount of digital sensor data. Current hot areas include:

- digital sensors
- RFID (radio frequency identification)
- barcoding (with a new standard in bar codes about to emerge)
- locational capabilities in cell phones and automobiles
- track and trace capabilities in parcel and letter shipment

It's an interesting question to think about what the world is going to look like with an explosion of such data.

The first key insight is that the amount of data in the world is likely to explode by several orders of magnitude. What this means is that all our assumptions about the appropriateness of current technologies is likely to change.

Second, in a world filled with digital sensor data, managing the difference between useful and archival data is not just a question of good economics, it's required for performance.

Third, it's likely that scientific and business breakthrough performance will require drawing conclusions about highly granular data. Figuring out such conclusions will likely require a lot of smart people, sophisticated analysis tools and a different class of information warehouses. These new information warehouses will be ones that can manage archival data well, ones to which software agents can be attached, because the sheer volume of data will be too large for human beings to deal with.

An example of this new class of tools is Alacrity Results Management described under Performance Management at www.eclicktick.com/id90.htm.

Alistair Davidson
Consumers Leading the Way

Last year I predicted that the consumer area of digital technology would be a growth area. My logic was that as in the early days of CD-ROMs, home users had more reason to upgrade to new technology than businesses. And for a period of time, home users had better machines with CD-ROMs than you typically found at the office.

We are going through the same process again of course. Drivers of this process are the digital camera and its unending appetite for storage, the digitization of music from the popularity of downloaded illegal music and the emergence of good legal services such as iTunes, Rhapsody and MusicMatch. But if anything, the trend is likely to speed up. My evidence for this includes:

1. Apple's repositioning of itself away from the education market towards the digital imagery and home entertainment market.

2. Sony, Gateway, Dell, and HP all emphasizing the home entertainment market. Each is trying to build on their traditional areas of strength or areas of current investment: Sony with its music and home entertainment equipment, game machines and movies, notebooks and PCs; Gateway with its retail outlets; Dell with its low cost sales channel; HP with a broad array of products and dominance in printers.

3. Microsoft's continued investment in expanding Windows usage with a home entertainment versions of Windows.

4. The explosion of digital camera usage particularly low res cameras in cell phones.

5. Continued growth in digital TV recorders such as TIVO and Replay.

6. Growth of 802.11b and g wireless networks as easy ways of linking up your home network.

In fact, as I sit here typing, my notebook is downloading and playing music off the Rhapsody network and I have hooked up my notebook to my twenty year old stereo.

So while HDTV, huge plasma screens and satellite radio are all interesting technologies, we are likely to go through another period where the consumer leads the technology way in spending.

And in terms of processing power, there are only three things driving me to replace or add additional machines.

1. Digital imagery editing and processing for video and still pictures.
2. Voice processing for dictating longer memos and articles.
3. Running so many applications on my machine that the 512 meg. limitation with my notebook may be a problem in the future.

So for the first time in my life, I am actually contemplating a traditional desktop machine in addition to my notebook for my personal use.

Alistair Davidson