This is the time of year where retrospectives of the previous twelve months become all the rage as content providers have column inches and/or pixels to fill up while skipping out the door on holiday breaks. As an independent blogger, I have no such requirements and the topic of this post, while retrospective, has nothing to do with 2014.
Earlier this month, the eight-year anniversary of this blog passed. I have recently been getting more comments from people who read it along the lines of amazement that I have been able to continue to blog for so long. The concept of blogging is quite long in the tooth now, but it continues to be a medium that works for me. I am appreciative that this blog has found an audience.
Looking forward before looking back: Next year, I will take an initial foray back into the world of academia after a 22-year hiatus by teaching an online class for Salisbury University. The preparation for the course has gotten me digging into a lot of concepts, but one thing that has been consistently coming to the forefront is my experience with zigGIS. For those who may not recall, it was an open-source/commercial/almost-open-source-again extension to ArcGIS that enabled direct access to PostGIS from ArcMap. It was my first real exposure to the open-source world and my participation in it was well-documented here during the first few years of my blog.
zigGIS was originally developed by Abe Gillespie, one of the best developers with whom I’ve ever worked. I ran across the code on an obscure site sometime in 2006 and it had been laying fallow for a while. Coincidentally, Paolo Corti stumbled across it at about the same time. Through the magic of blogging and the state of pre-Twitter social media, we all found each other and began to tidy up the code for up-to-date versions of ArcGIS and PostGIS.
This story could go on for some time so I’ll try to keep it brief. As the original author, the bulk of the work continued to fall to Abe. zigGIS was small, but just large enough to remain outside the effective grasp of three guys with day jobs. Along the way, we got support from great people like Steve C-P and a host of users but development was hard to squeeze in at night. The work would either need to start eating to billable hours or the project would need to take more of a back seat. Given that choice, and after much hand-wringing, we all agreed to take zigGIS commercial, which meant taking it out of the open-source world.
Looking back, I realize we didn’t need to do it that way, but we simply didn’t know that at the time. I’m not sure that the move was well-received by the open-source users of zigGIS and it was never terribly viable as a commercial product. I think one big lesson was that we gained no time as we had to begin dealing with licensing overhead and infrastructure. I expect those costs (in time and money) would flatten out for a product with a larger user base but that’s not where we were.
Eventually, we decided to make version 3 open-source again, but technical advancements by Esri obviated the need for a product like zigGIS, so the project was shuttered. While the official EOL was only three years ago, the project had been limping for some time so its effective end was much farther back. Of course, I got to work closely with Abe and Paolo, and I established relationships with many others that I now consider friends, so the experience was quite profitable.
I find that I consistently mine the zigGIS experience for a wealth of lessons. As it recedes farther away in the rearview mirror, I recognize it to be one of the most instructive experiences of my career and a turning point in my professional maturation.
If you are interested, version 1.2 of zigGIS, the last open-source version, is still available and enjoying a stately retirement here.