Form and Function

Problem Domain:
During the course of the day I probably work with and open 75-100 files, all organized into the dozens of client folders that I have. Each client might have mulitple projects etc, so repeatedly getting to files requires this endless drilling through the directory structures. I find this task repetitive, tedious, and annoying. Each time I search for a file I have to consciously read each directory label since they each look the same visually, once a directory unfolds I then have to repeat the process. I estimate that I spend probably about 4-6 seconds to find a file on average. I probably repeat this task 100 times a day. The cost to me is anywhere between 400-600 seconds a day... multipy that out over the year (assuming 250 work days) and you are talking conservatively over 25 hours of my year spent searching for and opening files !!!! Now I am not sure what your time is worth to you, but I certainly have better things to do than spend what could amount to a 3 day vacation opening and closing files.

A Potential Solution:
During the course of the day, there are some files I use over and over again, whether it is particular to a client or project I am working on, or a certain technology or package. Most of these are located in the same proximity to each other on the file system as I organize my files by related subject areas.

What if the File Explorer (Windows) remembered which files I openend most recently, and how often I opened them? It could remember this over the course of minutes, hours, days, weeks. Now, if the File Explorer knew what I had openend most recently, and how many times I opened it, it could use some deductive logic and mark these directories and files for me visually. What if the directory or file was a different color/size/shape based on how recently I had been there, or how many times I opened it? This would allow me to use my naturally given parallel processor (visual cortex) to instantly see these patterns and guide me to the most likely navigation path without me to have to constantly read directory and file names. Even if I just shaved 2 seconds off my time, that would amount to 100 hours in my pocket :)

I have done googles across the web to try and locate such a tool, because to me it would seem that someone smarter and more adept would probably have already built one. But alas, I could find none. This is a product waiting to happen!

My software development experience dates back over 20 years, but in the past 8 years I have worked almost exclusively on the Microsoft platform. One thing that I thought Microsoft did that was absolutely ingenious (and viral) was to practically spoon feed its development community. With their MSDN subscriptions and robust API documentation you could practically teach yourself to do anything with their technology. They also made it a primary focus to improve their development environments to make things as easy as possible. This is not to say their technologies are not without their problems and flaws, only that compared to any other platform I have tried, MS really got it right about catering to the development community.

Lately I have been exploring new technologies to use for data visualization applications and have been looking into the wave of technology around Rich Internert Applications. After setting up OpenLazlso, Microsoft Atlas, and Macromedia Flex in a lab, It started to dawn on me that Microsoft has a good chance of loosing the desktop completely.

It appears that Microsoft is betting the technological farm on Vista. All of the new and interesting desktop development efforts all will require Vista to support them, specfically Avalon or the Win FX platform. The challenge I have as business person and software architecture with that approach is that it would require me to develop applications around a platform that does not exist, and would require a user or company to upgrade their whole OS to take advantage of those new and importan features. In the best case scenario, if Vista ships on time, I could realistically expect only 50% of my target market to actually have it installed in about 3 years from now. That is quite a gamble for developing cutting edge client apps today.

Then comes along Flash. As a programmer at heart, I always considered flash and action script to be a bit of a toy versus a real development platform, but you can't argue with some of the incredible things people are doing with flash. After a couple of aborted attempts to trying to program within timelines and layers I pretty much set it aside as I found the programming paradigm to cumbersome and akward to be productive in. But, I kept on getting drawn back to the fact that I was seeing amazing things being done. Specifically, amazing visual things, that could really open doors from a data visualization perspective.

Along comes Zorn, Macromedia's merging of the open source Eclipse IDE with Flex, their flash based development platform. Now were talking.... When I found out the Macromedia hired away one of the guys who started and ran the Microsoft .NET product team, I really started to get interested. When I started poking around and realized that Flash has a 98% install base, across ALL types of devices (desktops, mobiles, even cameras) it really got my attention. Then when Macormedia announced last week they were dropping the price point of Flex from the enterprise level down to a price that pretty much any developer could afford AND they were going to release alpha bits to engage the developer community I realized they were going for the jugular at Microsoft. If Macromedia succeeds in winning over the development community they have a very strong chance of obviating the desktop, or at least the Windows one.

The new flash player 8.5 that is planned on being released by Q2 2006, has a complete virtual machine encoded into it, allowing run-time debugging, exception handling, and support for an ECMA standard OO language. While a majority of the people who run flash are running off a windows desktop, it does not require one, and since the flash player footprint is so small the deployment concerns usually faced when you pair up your application with a development platform are pretty much mitigated. Try leveraging MS .Net with a desktop or server app, and you basically force your user into a 30MB download of the .net framework, if they don't already have it installed, or if they don't have the correct version.

All in all, I will be very interested to see how much traction Macromedia can get with the developer community around Flex and Flash, because if they succeed and are able to leverage their amazing market penetration with their Flash player, it could really change the way we approach client application development, and the way users view their desktop.

My name is Tom Gonzalez and I am the managing director of a software development firm that specializes in a very small niche of the business intelligence community, dashboards and data visualization. In the course of delivering projects for clients and creating unique software products, I spend a major portion of my time working with new technologies and trying to predict what turns the industry will take and how different technologies will fit together.

I have been skeptically sitting on the sidelines for the last several months, pondering the benefits of blogging. After reading the article in Business Week a few months back I realized that this medium is here to stay. What has prompted me to start my own blog, is that over the past several months I have gotten some very valuable information and insight from reading other blogs.

I have a background in computer science and visual arts and one of my primary interests is to create technologies that make life easier for people and improve the aesthetics of their environment in a way that brings them more pleasure and satisfaction with their surroundings.