Andrew's ramblings

Opensource CFML Engines and what it means to the Community

Tags: General, Coldfusion, OpenBD, Railo

I have kept rather on the fence for awhile on this subject for a number of reasons, one of those was I was interested to see how others would begin writing about the impact it might have on the CFML community as a whole.

The one thing that we can not argue is that the future of CFML is NOT going to die, provided that the language heads in the direction that Adobe should have been taking it 10 years ago.

If you haven't been reading Peter Bell's blogs of late then you wouldn't know the drive to let them know what it is you want from the language, and I do mean CFML and not ColdFusion. If you haven't read these blogs then you can begin by heading on over too, Ralio: What Shouldn't We Do?

I think it should be noted what Peter says in that post as well, and one of the things that I will emphasize on is the why each other exists and that they should find a way to co exist. Peter also states that they both have selfish reasons that are independent of the interests of the community, and he is right I personally have seen first hand the interests of Adobe for now 12 plus years. In that time I have also seen how Adobe don't really care what the more serious developer needs, and by the time they get the push to do so it is usually too late and most people have moved onto another language that does what they need.

Should Adobe continue to adopt that attitude?

No they shouldn't for some very good reasons too, we always hear how their sales are still climbing and that the future of ColdFusion is as Adobe put it Rock solid. But I disagree with that and I will explain that in more further on, but first lets look at why the likes of The Smith Project began, and then Blue Dragon, openBD and Railo.

There was a time when Allaire saw that you could easily make the Application more RAD (Rapid Application Development for those who don't know), and ColdFusion began to attract some developers and the product became a hit because it was RAD. However Adobe can no longer lay claim to the product being RAD, nor can they claim that they wish to keep the product there either. I have had many discussions on what should be included in ColdFusion, and have always been politely told that is not our core market. Without going into details I will tell you what I have been indirectly told, we know what we want to do and what you are asking for does not fit into where we think we need to be. That attitude has never changed by Adobe nor has it swung their decision in anyway either, because years later when Adobe gets a bit more pressure it eventually gets added, not when it should have but when there is pressure, and Marketing and ROI has a big play in that decision too.

And that's the problem with Adobe.

The only reason Adobe still has a product called ColdFusion is that there are other die hard people, who have gone and written their own CFML engines. So the question is why?

The reason why is for the same reason that just before The Smith Project announced its open source version, I too was also working on creating a version of the CFML engine. I stopped for two reasons and the first was because I did not have the time to keep it going, nor did I have the time to try to make it into a product that others like New Atlanta and Railo have become. But it was an interesting learning curve along the way. But why did I begin to write one in the first place? Well that's because I wanted the language to move into the direction that Adobe is only now putting into the language, and is why Railo and openBD also exists for that same reason.

So when Peter states that they both have selfish interests that's because they do, I can't speak for Railo as I am not privileged to the behind the scenes there and what direction it will head in. But one can guess. As for Adobe their primary interest is to tie their products into a backend solution called ColdFusion, nothing more nothing less and that can be backed up by conversations from Adobe engineers and some of the things that gets left out of the language.

Now I can go into a huge rant on why Adobe has no interest in what we really think, and that is because the more serious developer that is now working on Enterprise solutions has moved on to other languages, that are more RAD and a hell of a lot cheaper than the Enterprise version that Adobe offer. Adobe claim that the Enterprise version is doing very well and that sales are above expectations, but the question that I would like to see an answer too is how many of these versions being sold are NOT hosting companies?

And this is where I think the likes of openBD and Railo will play a big part in CFML and its direction, and for those who don't know both these products are geared towards the more serious Enterprise solutions that Adobe have ignored all these years, and continue to ignore. I have made it no secret that the best thing that Adobe can do is become part of a consortium of companies that add support to the core language, and move into what Adobe wishes to do and that is provide modules to integrate their products into the CFML engine that is open sourced and free for anyone to use and deploy. Then Adobe can focus on spending money to make money, and leave the likes of integrating 3rd party products to those in the community.

When you break the nuts and bolts of ColdFusion down it is because of licensing costs that ColdFusion is so expensive, the CFML engine is very easy to write and is why there are others out there doing it now.

One last thing before I sign off on this post, and that is why I believe that ColdfFusion is no longer a RAD application anymore. You only have to look at the likes of other scripting languages to see that the title has now been taken away from Adobe, and Adobe don't understand the market as much as they think they do. Because the people they pitch ideas to is their core market, and their core market is not people who will be deploying Enterprise solutions and those will not be using ColdFusion from Adobe because of the overall cost that is involved. Thank god for the likes of Railo and openBD for the ability to do this, but even these languages are becoming hindered because they struggle to move forward because of having to be compatible with the CFML language that adobe puts forth.

And this is where Peter states that there needs to be a co-existence between everyone.

Do not get me wrong when I say the above statements, it is an observation from programming ColdFusion since 1996. I came from background of programming Perl and found ColdFusion easier than Microsoft's first attempts at ASP. And I believe ColdFusion still has a place in my heart or should I say CFML, but I can't see myself developing serious Enterprise Applications with CFML because the core things that are required for me, is that all engines support hibernate and closures, support more dynamic coding styles and more importantly ColdFusion adopt a MVC pattern to help lock down the application from time spent on security exploits.

And until that happens I have better solutions available to me to get me my pay cheque, but will always remain a ColdFusion developer because I believe that in time the language and not the product can evolve to be what it once was, a leader in its field and not a follower.

Tweet