Marco Polo ^{(freebase, versiontracker)} is an app that’s all about getting out of your way. With Marco Polo you describe certain contexts where you use your computer, and what you want to happen when you’re in that context.

I’ve got three main contexts where I frequently use my laptop, and one default one to cover everything else. For example, if it’s between 5:30 and 9:15 on Tuesday and I’m on the campus network, I’m in class. If my laptop is plugged into a particular external monitor and keyboard, I’m at work. Likewise, if I’m on the wifi network at my apt, I’m at home. You can get as complicated as you please here, giving each source of evidence a confidence rating, setting a confidence threshold needed to automatically switch contexts.

Marco Polo has been flawless so far at immediately recognizing where I am.

Then comes the fun part: telling it what to do when your context changes. When I’m in class, my speakers are automatically muted. When I’m at work, my wifi is turned off. When I’m at home, my screensaver password is disabled.

The biggest place where Marco Polo can grow is its selection of actions you can tell it to perform. I imagine a bit of integration with Quicksilver would be fruitful there. On the other hand, as I was writing this blog post, Marco Polo performed an automated update that added the QuitApplication action, apparently in response to a feature request… so perhaps they’ve got that under control.

Of course, since one of the built-in actions is to run a shell script, I’m happy as a clam either way. Marco Polo’s abilities outpace my desires for the moment, which is definitely a kick-ass place to be.

There are plenty of examples of disruptive ideas and social phenomena that were so in conflict with those of their time that they were only really adopted in large numbers by the next generation, those that grew up with them. Take evolution for example. Many of the leading lights in biology at the time were never convinced of it, and went to their graves arguing against it. They missed out on an opportunity to make significant groundbreaking contributions because of their prior ideological commitments. It was the next generation of biologists, coming into the field without having made any commitment to a biological ideology that were able to look at the various explanations of life and recognize evolution’s explanatory and predictive power in contrast to alternative ideas. So it is with many social movements. In America, a large number of people three or more generations back were and are quite racist, largely because racism was built into their world views at a young age. It is my understanding that racism has largely died out in America not through convincing people who already had a strong opinion, but through a generational change.

The interesting question here on a personal level is how recognize this sort of change and how to consider these ideas fairly when they go against deeply set beliefs and understandings of the world. Certainly it appears that the older you get the rarer this ability is, but there were a number of people who were well established in biology who accepted and argued for evolution. If there is a radical disruptive idea in my field or in society in general, I want to be able to consider it on its merits, rather than discard it for not cohering with other ideas I hold to be true, and I want to know how to be prepared to do so.

Keeping an open mind and being open to questioning of any aspect of your world view is one pretty sensible response, but I think there’s an additional element.

There is a well known principle of psychology that we are loathe to question critically something we’ve made a visible, tangible, or most of all monetary commitment to. This principle is exploited to powerful effect by everyone from scam artists to cult leaders, stringing people along far beyond what they would have accepted otherwise. An excellent case study in this is Scientology, wherein large commitments of time, energy, and money are ensured before one learns of their more ridiculous beliefs. Therefor, a level of humility is necessary, I think, to be willing to admit that one has been wrong and that one has publicly and repeatedly acted and spoken based on false assumptions.

SSL of course is the ideal solution for encryption on the web.  The place where a good javascript-based approach could make an impact is in web apps meant to be deployed by users.  Blogs, bulletin boards, wikis, photo sharing apps… these kinds of web apps all commonly have, at minimum, an admin interface, and many times support user accounts.  These sorts of apps are also often deployed on shared hosting where SSL is expensive if offered at all.  A nice javascript authentication library that has been vetted by security experts could make a difference in this space, offering a step above simply sending passwords in plaintext.  The difficult part wouldn’t be coding the library, but ensuring compatability and correctness in a variety of environments.

Any security geeks interested in giving this a shot?

I wrote a bit of ruby (along the lines of gstakeout) that ties together the OSX notification manager Growl with Ruby’s unit testing framework. It’s just a little hack that adds a couple of lines into the command line test runner of unit/test, but I found it to improve TDD on little projects so much that I made a short screencast showing it off:

This could be improved by integrating it into autotest, which has the nice property that it only tests methods that need to be tested.

edit: Usage: Add this to the end of your test script:

require 'growlrunner'
runner = Test::Unit::UI::Growl::TestRunner.new(YourTestClassGoesHere)
runner.start

If you do this, it’ll work with autotest, and in Rails projects.

Requirements: RubyCocoa
Download growlrunner.rb (updated 5-2-2006)

Perhaps The Conductor will smooth this over, if it’s ever released.