Mach-II: End of Life Cycle

On September 3rd, Peter Farrell announced that the team behind the Mach-II ColdFusion framework (himself, Matt Williams and Kurt Wiersma) will no longer be working on or supporting the framework.  With a 10 year run, those who initially made it and kept going have much to be proud of.

When the announcement came across the Mach-II Google group, I was of course disappointed to hear it.  We've been using Mach-II in our shop now for several years.  We started with version 1.6 and in the last two years we upgraded to 1.8 as we built new apps.  So we had a vested interest in him seeing it continue to be carried on and improved.

At the same time I was not entirely surprised.  1.8 was released over three years ago, and while 1.9 has been "in development" since then, it didn't seem like there was a whole lot of activity on it after that first year 1.8 was released.  There wasn't a whole lot of activity on the Google group either, not even many "help me" discussions. 

There was some effort to move the Mach-II documentation over to Github from a wiki format, but despite an announcement saying it was completed back in April it really wasn't.  Entire chunks of the documentation were never moved over leaving current and potential new users unable to find some very basic information on how to use the framework.  And because Github documentation is not readily searchable what was there was practically unusable.

While the framework is open source, so it can be forked and continued on, I don't think it will be.
The signs have been there for a while that Mach-II was a dying framework.  If I needed any further proof, I would say it would, in the almost complete lack of reaction from the ColdFusion community, at least that I could find, about the announcement.  One other small blogger like myself who made a quick repost of the announcement and a brief mention on the CF Hour podcast, episode 193 (06:00-09:50), that focused more on how old it was and how anybody was still using it probably should upgrade to a more current version of ColdFusion and a better framework anyway. 

Now I will say I felt the hosts, who noted they knew little about it, were wrong to imply that framework must have tons of bugs and be completely useless because of its age.  For us it's been stable and it works perfectly well on ColdFusion 9.  I honestly can't give any bugs that we have personally run across, though we have occasionally wish there were some additional features that perhaps would've been there in 1.9 had ever been released to production code.

Still, I can't argue their point that it seems as if there are very few users of the framework left except folks like us who've been using it several years, and so continue to do so while working on their legacy apps.  I mean really, how many shops have the time or energy to completely convert an app from one framework to another for no other reason than to change frameworks.

Now will we continue to build brand-new apps with Mach-II?  Probably not.  If we build any new ColdFusion apps of all, we likely will explore current frameworks that are still actively being maintained and worked on. 

As I mentioned back in November, our shop is pretty much going to be moving away from ColdFusion for new development.  At this time we still haven't decided where the directions going, but as we've been busy working on the apps that were in progress when this was discussed.  But it's likely that decision will be made relatively soon and the next new app we build will not be in ColdFusion.

Now I don't just mention that to indicate why we probably will not continue to build new products of Mach-II. I mention it because of the reason the Mach-II team has said goodbye, so to speak.  That is because all three of them have changed from ColdFusion to other languages: Peter and Matt now do Python and Django while Kurt switched to C#.

These are not nobodies in the world of ColdFusion; they're well known to those of us who've been around for the last decade or so.  When I was a younger, fresher developer, they were some of the guys I looked up to as "real" developers, the pros, the folks I hoped I could be as good as one day.  And in many ways that sentiment is still there, just as it is still there for Ray Camden, Ben Forta, Scott Corfield, Adam Cameron, and many others, even though I'm no longer that newbie developer who was terrified that someone would realize I had no idea what I was doing.

So to hear that three of these guys have walked away from ColdFusion, to me is an even bigger nail in the coffin of the language I love that anything else.  It hurt more than hearing that Mach-II was going away, it was more upsetting.  And yet I can understand it, while my more stubborn side will continue to go kicking and screaming, the more rational side of me wonders how long before the rest go too?

And after that dip into more sentimentality than I expected or planned, I'll end this by saying Peter, Matt, and Kurt I wish you much success in your future endeavors.  I'll probably be watching for you online because I'm sure I still have much to learn from you as a developer.  And thank you for giving us a great framework that I feel confident will continue to keep our existing apps going for many years to come.