From North Montpelier to 10 miles south of Hardwick, Route 14 is a mess. However, it’s worth it if you’re heading north of Hardwick. The road and the views all along Route 16 are simply wonderful. Turned around at Barton to turn the harsh headwind into a tail wind, sailed all the way to Morrisville. Turned the corner and up over Elmore and back down to Montpelier.
RiseUp! has just announced their Freedom Summer of Code which seems ideally suited for developers who are interested in helping to develop code that both helps people and organizations to not depend on closed source code. For Drupal developer’s I think this could be a great way to develop and improve some functionality that is not dependent on third parties (ahem, Google).
The Freedom Summer of Code aims to advance critical movement technology projects and tools that benefit a wide-variety of radical social justice organizations and movements; inspire developers to become more interested in directly participating in social-justice tech organizations; contributes back, for the benefit of all, to the free software world which sustains us while simultaneously honoring individual’s labor; increases the social ownership and democratic control over information, ideas, technology, and the means of communication; empowers organizations and individuals to use technology in struggles for liberation. We are developing software that is geared specifically to the needs of network organizing and democratic collaboration, providing new services that greatly enhance your security and privacy.
They solliciting applications now, as well as additional donations to help the project along.
The climb up Route 17, above the base of Mad River Glenn and then up to the top of the ridge (Ap Gap), passing one of the tops of Mad River’s lower lifts, is a spectacular climb. Not only is the road in good shape (a rarity in Vermont this year), but the river which you climb along side is a clear crystal green, and the ridge that Sugarbush and Mad River stretch along is just a fantastic view. The decent offers some great twisty turns, and some nice straight aways that quickly rocket you up to 40mph and beyond. Heading out the Mad River valley, you meander through a beautiful pastoral valley.
The climb up 100 toward Waterbury opens with a fantastic view of Camel’s Hump. This road was paved last year and again is a gem in road quality. While it is more trafficked than other roads in Vermont, there is decent shoulder. This is a series of pitches- not the steady pace that Ap Gap requires, but 4 short climbs, but the reward is several miles of down toward route 2.
While this ride is somewhat of an out and back, it’s got good variety. I think the only real bummer is Route 2 which has a narrow shoulder and some dubious pavement.
Sunday proved to be a less sunny and surprisingly more windy day that Saturday. Out RT2 to RT100, up passed Mad River Glen, over the Application Gap, and a fast down toward South Starksboro. Then climbing, slowly at first, pavement changing to dirt, and pitching upward fast. At the top of the gap, there is still snow pack on the road, but there’s a bare spot to sneak past. And then down, down on the breaks the whole way. Compared to the western side, the road is in vastly better shape, but the turns are fast and there is still lots of debris on the road, and frost heaves on every drain crossing. Spills out onto gravel again, and then a fast descent back to RT100.
Saturday was the old stand by of around the Woster mountain range. Nice tail wind heading North, but rounding the corner to Elmore was a steady headwind the 26 miles back to Montpelier.
One of the core features that Media Mover has offered from the beginning is harvesting files from an email account and enabling those files to be transcoded as any other file in the Media Mover system. This has provided people the ability to take posts from a variety of different devices (think cell phones here) and get that content directly on their website. In many ways this is a pretty amazing feature- while there are existing email gateways, there aren’t many options for processing the incoming data.
Unfortunately, when Media Mover was first being built, we made the classic programmer move of “go it your own”- we didn’t look for prior art and instead did it our own way. Whether hubris, bad research skills, or dissatisfaction with what existed, this seems to be a common theme for programmers in the open source community- scratch your own itch, if you will. I think there are good sides to this- sometimes, its just what’s needed. On the other….
It didn’t take long for this approach to mail handling to seriously bite back. The code we used couldn’t support SSL mail connections, the code is tangled and not easily extensible. There was no reasonable possibility for adding in support for commands for taxonomy terms, user ids, etc. Suddenly, I realized I had an unworkable solution for our client.
While I’d argue that the core of Media Mover is the other side of this problem- there is a need for a file processing system that can handle a variety processing requirements in Drupal- the mail solution that we implemented failed in two directions. For one, it was hurried and did not anticipate other use cases, for two while it solved a problem that there wasn’t a current solution for, it could have been part of a contribution back to other Drupal projects, making it more viable in the long run.
Fortunately, there are folks like Moshe in the Drupal community who have already built solid mail handling libraries- mailhandler to be precise. While Moshe’s module doesn’t directly integrate with Media Mover, it is built in an abstract way so that I can reuse many of its parts to leverage existing work, and help maintain a common code base for a variety of projects.
So I moved forward and ripped out the custom code from Media Mover and made the Media Mover Email functionality depend on mailhandler. I recycled forms and functions from mailhandler inside of Media Mover. I basically created a presentation layer over the standard mailhandler forms and use it as a wrapper around PHP’s IMAP functions. Now instead of trying to support people on the custom code inside of Media Mover, I can help submit bug reports and patches back to the community of people who are using mailhandler- something that hopefully is a win for everybody involved.
The notion of “scratching your own itch” is a powerful one, and I think its something at the core of the DIY sensibilities of opensource communities. But its not without peril. Having project after project duplicate portions of the same functionality takes away energy from creating solid solutions. Sometimes its necessary to attack the same problem from a different perspective- there is no doubt about it. Flexinode and CCK might well be a good illustration of that. But I think more often than not, those are the exceptional cases, not the rule. What I find attractive in the Drupal community is that there is more of an ethos of a “community itch”- that as programmers, developers, administrators, and users, we are moving together to make a fantastic environment, not just solve our individual issues.
You can checkout the latest and greatest Media Mover email handling code from cvs
Stephen Musgrave from Capellic was kind enough to video my talk at Drupal Camp a few weeks ago. I attempted (and succeeded!) to install the full Media Mover suite in under an hour, along with an introduction to the theory behind Media Mover. Check it out! You can check out the crib notes if you want to follow along.
DruTube - Building a video site in Drupal from Capellic on Vimeo.
As Media Mover continues to stabilize, this process is going to become easier and easier. I’ll be adding an install profile which will help to reduce the number of steps that are required to get things up and running.
