GPS's for bicycles are emerging, and they offer a lot. No more measuring the radius of your wheel ... just add batteries and go. Some include heart rate montiors (with the constrictive chest strap).

Though bike-focused GPS units do well as bike computers, in time they will likely do more. Imagine hitting a website and downloading a local route or waypoints from a friend or stranger to the unit. Then make your way to the starting point and hit GO to start the race. Nevermind that your competitor may have done his/her race weeks or years earlier ... you're going head to head on the screen! And you can upload the results for bragging rights.

In a race with waves, downloading the progress of others in real time (which goes beyond a GPS's capability now) would make it more of a head-to-head race.

What if data about poorly surveyed areas was aggregated to give a sense of the altitude of land masses?

Or if non-road routes were aggregates for map makers, who could analyze the data to locate popular trails that may have never been recorded (or intended by any authority!). Or imagine you were in an unfamiliar area and you could download non-road paths that were popular -- ie. to the GPS unit a field is a field, but if 30 people have ridden a particular route through that field, it might suggest the same to you.

They would likely also serve as Black Boxes in the case of an accident -- something that cyclists prone to breaking traffic laws should be mindful of.

I got excited about Web2.0 about two years ago; I read an article about flickr, which fawned over its novel photo "tagging" feature. That article speculated that OS's will adopt tagging in place of hierarchical folders.

Here's where I went askew. This article -- or how I read it -- led me to see tagging as a brainstorm-esq -- almost sub-conscious -- process. Ie. for each flickr photo, delicious link, or local file, you were to just spew a series of words (or hyphenated phrases), and in months or years to come, you or any reasonable person would spew a similar set of words when pondering the photo/link/file.

I even developed this into an idea for a service-business: Staples and Grand & Toy have notoriously difficult information architectures -- both in their physical stores and online. How do you categorize whiteboards and CD jewel cases in one dimension? But if a sample of the population assigned 6 intuitive tags to every Staples SKU, eventually, any non-caveman would be able to find stuff with a command-line-style search.

This is all interesting, but the point of this post is that tags should not be spewed in a brainstorm; they should be carefully considered. My delicious account has tags like: business, company, business-idea, start-up, entrepreneur, entrepreneurialism. So I feel forced to re-tag every cool start-up with six tags, where one will do.

No, the right approach to tagging is to assign the right tag in both your brain and the tagging system. Have one word for one underlying concept. Tag-clouds and other nifty features help with this ... you can assign what you have most-often assigned.

Tagging takes discipline. Rather than re-build my delilcious account (and not super-happy with Blink List), I'm going to take the opportunity to start-over with diigo.com, which techcrunch rates well.