SitePoint’s August 2nd 2006 issue of Tech Times was released yesterday. In it, a reader asks what products and technologies she should learn (she mentions “AJAX, CSS, Flash, ColdFusion & MSSQL, PHP & MySQL, Microsoft technologies”).
I have a couple of issues with Editor Kevin Yank’s response, and just sent him the following:

I am writing regarding your August 2nd column, and your comments in your “What To Learn?” section. I quote:
“… and even the relatively stagnant ColdFusion have plenty of power to offer with a gentle learning curve up front.”
You are correct about the gentle learning curve, but can you clarify “relatively stagnant”? Just to be clear, the Webster definition of stagnant is “not advancing or developing”. ColdFusion was first released in 1995, ColdFusion MX 7 was released in 2005, 7.0.1 later that same year, and 7.0.2 in June of 2006. In addition, the ColdFusion team is hard at work on the 8th major version of ColdFusion (currently codenamed “Scorpio”), to be released in 2007. Obviously, we are both advancing and developing ColdFusion, and so the term “stagnant” is utterly inappropriate. As such, I must request that you update and correct your statement.
In addition, just in case you were unaware of this:
“But more heavyweight platforms like Java and ASP.NET can open the door to more complex and esoteric applications, not to mention higher salaries”
Yes, Java is heavyweight, I agree. But did you know that ColdFusion is a Java application, it runs on standard J2EE servers (like IBM WebSphere, BEA WebLogic, and our own JRun), all the runtime services are native Java, and it actually code-gens Java source and compiles down to Java bytecode (in fact, you can deploy your ColdFusion application as Java bytecode without needing the CFML source on the server)? ColdFusion is Java, no more or less so than if you had written your code in Java or JSP directly. So yes, Java is indeed heavyweight, and ColdFusion is as capable and heavyweight as the server it is running on.
— Ben

I’ll let you know if he responds.

74 thoughts

  1. CFEclipse/Eclipse platform is missing pretty much ALL of the features a serioius developer would expect – thats no response to the missing CF IDE problem as Dreamweaver is better than CFEclipse.

  2. "is the fact that developers are not supported with a professional level IDE"
    Agree totally. Dreamweaver is a design tool, and cfeclipse is just a code assistant. If Adobe cotton on to the fact that professional coders will never take coldfusion seriously until it has a fully integrated development environment, they might actually grab some attention with coldfusion.
    I must say that cfeclipse is okay, not slow on my machine. Its just rather, well, basic. What iirks me is that with teh money they charge for coldfusion they could easily have developed an IDE, possibly even shipped it with coldfusion server. Even if they did it at a loss initially, adoption would improve and it would effectively pay for itself.
    Come on Adobe, don’t let cfeclipse set the standard for development in cf. Give us some decent tools and lets get some pro coders behind cf!

  3. Just to add a piece of information, people are demanding for "Proof" Coldfusion is not "Stagnant". Well, here is one source:
    This displays a list of Programming Languages, and it includes their SERP (Search Engine Report Rank) and SERP rank change of a year. Coldfusion is currently listed as SERP: 16, with a SERP Change of +9 over the past year. This is the third highest rank improvement, after Ruby and
    I think jobs come with skills and experience. I currently work as a CF Developer (intranet) for a very large Cell phone Corp, and I CONSTANTLY get offers for employment for outside ColdFusion development. One thing that people are not mentioning is that many large corporations use CF for Intranet Development.
    BUT MYSPACE IS CHANGING FROM CF TO .NET!!!!!!!!! ARGHH!!! hmmm, ok… one company is making a technology change… That must be the signal of the apocalypse! Or… it might be just another company switching technologies. I wish I could find a listing of companies changing from one platform to another, I am sure hundreds do so weekly, from .NET to CF, from PHP to .NET, from xxx to yyy.
    Also check Ben’s extensive list of companies using ColdFusion:
    And possibly the biggest proof that ColdFusion is not "Stagnant" and actually is only starting to grow is: ABOBE SYSTEMS just bought it! Why would a corporation with $1.996 billion USD (2005) revenue purchase a clunker? One possibility is that this is only the beginning of the Golden Age of ColdFusion. In fact I am constantly amazed at how much is being added to CF by the company and the community on a daily basis. And when Scorpio comes out, I do not have a single doubt CF will kick even more butt than it currently does (which I can’t even imagine).
    So please, if you do not like ColdFusion, or you do not see it as a good bread winning technology, or you just cannot find a job in the field. Switch technologies, pick up the newest fad, or become another mediocre ASP drone, or a PHP Kiddy. The only impact it will have on me is more ColdFusion jobs for me to choose from! 🙂
    Keep up the great work you are doing Ben!

  4. re: Also check Ben’s extensive list of companies using ColdFusion:
    Thats teh problem, its extensive! With numbers like this, one wouldn’t want to base a ‘cf is not stagnant’ argument around them.
    Re: ColdFusion is not "Stagnant" and actually is only starting to grow is: ABOBE SYSTEMS just bought it!
    Talk about reaching – adobe did not buy coldfusion, they merged with Macromedia. I suspect you know this, and that you can see how your comment was blatantly misleading. This is just a sign of desperation to me.
    Re: In fact I am constantly amazed at how much is being added to CF by the company and the community on a daily basis
    The you would staggered and breathless if you looked at other technologies. In the scheme of things, not much has happned with cf in years. To a coldfusion loyalist, I have no doubt, it would seem not to be true. Sadly, CF has done barely enough to keep up with the vastly popular PHP. In fact, cf is still missing features that php has. (though you can download stuff for cf that will give it some of this missing functionality)
    RE: And when Scorpio comes out, I do not have a single doubt CF will kick even more butt than it currently does
    CF kicks butt now does it? Thats a new one – it is not even a mainstream technology any longer and here’s you saying it kicks butt. Interesting.
    rE: So please, if you do not like ColdFusion, or you do not see it as a good bread winning technology, or you just cannot find a job in the field. Switch technologies,
    Thats the whole point, people are switching. More importantly, they are rarely even considering cf, let alone adopting it.
    Nice post, I see where you are coming from. But the thing is you are in the minority. There are many many numbes of developers that would love to use cf, but just can’t – no money in it. Regardless, good to see its working for you.

  5. >>In fact I am constantly amazed at how much is being added to CF by the company and the community on a daily basis.
    Nice that you enthusastic – but might I suggest that you are rather easily pleased. First up, what is happening to coldfusion on a daily basis by Macromedia/Adobe? Not much. Relatively speaking, one of coldfusions recognised downsides is that there is not much third party support for it compared to what exists for other platforms, and that there is relatively no interest in it by vendors distributing software or the open source community.
    I am not saying you are wrong, but I am saying that your point is rather backward. Coldfusion sees much less happen on a daily basis than the likes of php, and other more mainstream technologies.
    It would be good to see cf folk make points that are not so obviously weak and not so easily disproven.
    I also agree with the other poster, if Bens list is extensive, thats a bit of a worry.

  6. I’d like to address the comment by Ali that MySpace "only switched to BlueDragon because it works with .Net, and they had already made the decision to phase out CF completely and become a 100% .Net site. So BD was only a stepping stone."
    This is simply untrue. MySpace did test CFMX prior to choosing BlueDragon, and decided to go with BlueDragon for improved performance, reliability, and the ability to integrate with ASP.NET. Yes, some of the site has been converted to "pure .NET", but the majority is still running on CFML (most on BlueDragon.NET, but some still on CF5).
    MySpace is currently beta testing the new BlueDragon 7.0 release (and is quite pleased with the performance improvements). I’m not aware of any plans to "phase out CF completely and become a 100% .NET site."
    Vince Bonfanti
    New Atlanta Communications, LLC

  7. I don’t want to rev anyone up. I work as a CF developer. In my neck of the woods – Queensland Australia, coldfusion is being used less and less. It is harder to fine job adverts for it and I am close to reaching the top pay rate that developers get with my company. On the other hand job ads for java web dev and dev are on an all time high. I wouldn’t say that coldfusion is stagnant as a statement by itself. What I would say is that coldfusion is stagnant relative to the other web languages which are experiencing massive growth in comparision. It would seem that there are pockets of work for developers and a lot of companies are moving over to asp or java because they cannot find enough skilled CF developers, and the cost of the ones they do get are extremely high. Its the old secenario of a little fish in a big pond (asp, java) or a big fish in a little pond. So in my experience, i am looking at moving away from coldfusion and venturing down another path that offers more job security. By the way, i find that most of the work in coldfusion and for most languages is just copying and pasting and that there are 12 year old kids out there that can do it, let alone the programming shops in Asia that have employees work for 2 dollars per hour. Anyone who believes that coldfusion will be around forever(any language) is kidding themselves and if you feel its stagnant now wait a few more years and it will get worse. Perhaps programming in general is stagnant, I mean all of us do it, we have a problem and just get on google and look for the answer, its already done, we grab it and make out that we coded the solution. The only good thing about coldfusion is that you get paid at the end of the week. But if i was you I would get out now. In fact i spoke to the well known Ray Camden the other day in a breeze session and even he told me to get out of coldfusion. All these guys that say coldfusion is great are the ones in management and sales, who make money of selling the product, or all the typical nerd that sits in front of the computer all day and codes, but had no marketing or business savy. I mean who really thinks that coding in CF requires talent, a monkey could do it. I mean look at what management call the programmings team – code monkeys. I have been in cf for about 3 years and there is no where left to go. I mean I have reached the pinnalce and life from up here is crap, I wish i hadn’t wasted my time. Now i have to go back and retrain in another field so I can have a career, not something that is a stagnant gimmick. I am so glad a job interview with a big engineering firm next week as a data analyst least i won’t be ctrl-cing and ctrl-ving for half my day and cfset-ing for the other half. So they way i see it is that cf programming in my part of the world is stagnanting. And its so funny this guy at work nows nothing about anything else except cf, he’s like a horse with blinkers on and cannot see that he is wasting his time being so focused on the one thing. And he goes and talks to the boss about all of this CF-talk and the boss just looks at him with that i don’t want to know how to look, just do it and get paid peanuts casue u got nothing. People who use CF become Stagnant

  8. Just to be absolutely clear – I never told anyone to get out of ColdFusion. I did speak with this person in the breeze room before my lst presentation, but the rest is a lie.

  9. Ray,
    Don’t worry. I don’t think anyone actually believed that post. And even if anyone took it seriously initially, they did not anymore once seeing that commented attributed to you. We know you better than that.
    — Ben

  10. Ben, I think you were total in the right to question Kevin. It is the responsible thing to do. Just because something is written does not make it gospel and therefore should be challenged. Ever heard of the Bill of Rights? Kevin was stating his opinion without a solid foundation to form a valid opinion and even admits he had to look into CF further. Unfortunately, reading other blogs does not constitutes quality research in my opinion. Kevin put way too much emphasis on judging CF “stagnancy” based on book sales. Just about any question I’ve had, LiveDocs has provided the answer or in the least, pointed me in the right direction.
    Perhaps I just don’t understand Kevin’s point. I’ve been developing with CF now for at least 6 years and I’ve made quite a career for myself. A reader posted about CF being used more in private domains versus the public domains. For the fortune 200 company I work for, CF is THE company standard for internal sites. Developing in any other language requires exceptions to be completed (which can be as difficult to learn as a new language 😉 but I digress). If Kevin’s main point is for people entering the web developing arena to avoid CF, my response would be then do not bother turning in a job application to my company. We use CF and there are no plans on changing company standards. What Kevin should have suggested to maximize your portability is to become comfortable in a several languages (notice I did not say proficient but comfortable).
    Bottom line, As Stacy Young stated, you should have more than one language in your tool belt.

  11. Wow, way to revive a conversation, guys! Great points on both sides!
    Truth be told, I’m not much of a blog reader. But this is a topic that I deal with regularly, and let me say that I’ve been a die hard CFer for a long time (that won’t change soon, it’s working for me), BUT there are some really good points from the opposition, and for the most part, they have been respectfully stated, so I can’t consider it CF-Bashing.
    As far as careers and profit potential go, that is so subjective to your location/industry/resources, so it’s not even worth discussing. Suffice it to say that you can make well over 100K/yr as a ColdFusion developer if the conditions are favorable; and most of the popular-technology developers I know have hit a glass ceiling in salary, due to the competition. My recommendation is learn the basics of several languages; learn how to interview well so you get the job; then learn as you earn.
    I think the Stagnancy debate is stupid and ambiguous. Let’s talk about something more relevant and measurable like COST. There are many factors to cost (Licensing, Development Talent, Support/Maintenance, and the all too important Time). While I certainly see the shortened timeframe of building applications in CF, I don’t buy the whole "Cost is not important to lots of people" mantra that my friends have been preaching. Sure in the old days the cost of CF was just a drop in the bucket compared to development costs and the insane rates firms were charging for programming talent. But that paradigm is changing, and off-shore talent offer the benefit of turning programming into a minor cost. So why would someone choose to pay $1500 for the server (per server), when they spent $10K to build the app (assuming shared hosting was not an option)?
    We can assume that the support costs would be comparable, so let’s talk about scalability. Not the scalability of the app, but of the license. Say you have an app that needs to be load balanced for reliability. Now your cost is about $3K for CF Licensing. Using Quad cpu servers? Double that! So, now the costs are piling up for basic CF use. But say you chose CF for one of the “unique” enterprise features? Well, then you have to quadruple the number you had before. It’s hard to justify an investment into a CF server farm for traffic that is “expected”. Now, imagine the number app servers myspace must be running. I know Vince hasn’t mentioned it, but could cost have also been an issue?
    I know this isn’t a cost discussion, but it always comes up as the first attack at CF, and quite frankly it’s a valid one. Now, Adobe is a Software Business, and it is up to the business people to decide the direction, here. CF may very well flop as a free or inexpensive tool. Or it may flourish. We’ll never know until it happens (and don’t give me that “it used to be free” crap, because times change). Personally, I think free would be a great opportunity to up-sell enterprise.

  12. I must say I got a kick out of reading this thread. A few posts into it, I was thinking ‘This sounds like our Andy Grant troll off the Adobe forums". And wouldn’t you know it, he even signed one of posts! For those of you that haven’t had the pleasure of his rantings on the forums, he consistently posts all kinds of negative stuff about ColdFusion, making up new personalities as he goes and often even debating amoung those aliases. His favorite topics include CF being non-existent in Australia, Tiobe index, lack of books, and lack of IDE. Poor spelling and grammar are another hallmark of an Andy post. He particularly loves to talk about how he *likes* CF, it’s a great product…but no one should waste time using or learning it anymore. I’m sure he will vehemently deny this, but looks to me like the troll has found himself another home.

  13. blitz,
    Not sure why you’re surprised; People generally respond to offending posts. On one end is Kevin opposing CF, on the other is Ben protecting CF. I’m sure Kevin’s readership is comprised of people who can not form their own unbiased oppinion of CF, so that is that place to post that stuff. Besides, Ben already knows we’ve got nuthin but luv 4 him!

  14. Here in DC the CF job market is very strong and I have never had a shortage of opportunities in the 6+ years I have been in the biz – even during the dot-com bust when I was relatively inexperienced. I agree that certain dev platforms are more popular in certain regions/market sectors and since CF is the de facto Federal Government standard for dynamic apps I don’t see there being a serious shortage of work anytime soon.
    Java and .NET are always going to be the twin towers of web development since C++ and VB have been around much longer and Java and .NET are naturally more appealing to developers who are well versed in those traditional languages. Instead of presenting ColdFusion as a direct competitor to Java and .NET, I hope Adobe will continue to market CF as a practical alternative for small to medium-sized RAD apps that need to be built quickly and modified often. After all, CF was created to alleviate the need to write complicated CGI scripts to perform simple tasks.
    For most organizations (like mine) with a small budget and a need for numerous specialized single purpose apps, the cost of a few CF licenses is relatively insignificant compared to the extra money it would take to hire more full-time developers to build in Java or .NET.

  15. I have to back Ben up as well on this.
    I have been a CF developer since 1998 and I don’t really even want to learn any of the .NET stuff, Ruby, or PHP (though PHPMYAdmin does rock for MySQL management and Media Wiki is the best Wiki I have seen and it is done in PHP). I use the best tool available…CF is usually that tool.
    One of the arguments was that CF is only for web development…well duh…I thought that was pretty obvious. Most of us are web application developers, not desktop application developers, so that is a pretty moot point.
    As far as being a language a newbie should learn or not…well if said newbie wants a language they can learn real quick, then CF is that language. The more advanced features, like using Java can be learned later.
    Job Market… Here in the Chicago area, there are quite a few companies…big and small…that use CF. I have been doing consulting for the past year and a half and I get several phone calls and emails every day for new positions. I turn down more positions than I can remember. I think the longest I have been unemployed has been a month…and that was a year ago. The company I am working for now is looking for another Senior CF developer…as is a large pharmaceutical firm in the NW suburbs(at 54/hr). There was a comment from a guy in Seattle about how there is no CF opportunities there. Well big shocker…that is Microsoft’s backyard…one would at least hope they were strong there. Other parts of the country can differ…as with anything. I think one thing that has been lacking is the marketing support. Macromedia was very guilty of this and so far, Adobe is guilty of it..though they are new to the CF game, so I will give them a bit more time to get thier act together ;-).
    I am also certainly making good money. CF positions here in Chicago will get you anywhere from 35(for less experienced developers) to almost 60 an hour. That translates to 70k-120k/year. That is certainly not chump change.
    As far as cost, Ben, you should repost your CFDJ article on that…I thought it summed up the cost of CF vs. the cost of ASP (I think it would apply to other .NET technologies just as well) argument quite well. The other side of the equation, the cost of developing or purchasing the features in ASP/PHP/.NET/etc… that are already there out of the box with CF is conveniently overlooked when this argument is brought up by CF bashers. When I first lost my job to the .com bust in 2000, I picked up an ASP book to learn it since I saw a ton of jobs out there looking for ASP developers. The first chapter of this book was "This is what ASP doesn’t do…" it was a laundry list of just about everything CF does out of the box. My reaction was, why would anyone use this POS? I never went forward with my plans to learn ASP.
    Books…CFWACK and Adv. CFWACK are the only books I have used. I think I bought the O’Reilly book for CF5 or 6 and I never use it. THE CFWACK and Adv. CFWACK books are awesome and besides the help topics in Dreamweaver, I hardly use anything else. That is also backed up by a very active community like the cftalk list at that is very helpful when one is stumped. So it is not surprising when i see that book sales are not as high as other technologies. Other books just are not needed.
    I think a lot of the perceptual problems started after the .COM burst in 2000. Many companies falsy believed that ASP…since it comes with IIS was free. So many companies, feeling budget crunches, went with what they perceived as a free product. Development costs and modular purchase costs didn’t jive with the perception, even though some held on dearly to it. If you compare the cost of CF at it’s most expensive, it is still a lot cheaper than most of the application servers out there…some which will cost 10’s of thousands of dollars just to buy…that;s not including dev costs. They you have the misguided and asinine assertion that because CF is cheaper, there must be something wrong with it. I think, sometimes, companies just like to throw money at projects.
    CF is just a superior product. I am not just saying that because I am a CF developer, I say that out of experience. No other language out there allows you to do what CF does and what CF can do with the ease of use and flexibility that CF affords. I think Ben once said it best…(to paraphrase) "CF is as extensible as your developers" meaning that CF’s only limitations are the skills of your developers. If you have developers that are skilled enough to develop what you need CF to do (if it doesn’t do it already)…then the sky is the limit. Even with developers that aren’t skilled enough to do that, you still have access to Developer Center and quite a few other sites that offer a plethora of tags, some free…some at cost, for you to make use of. Combine that with CF’s integration with Java…and you have a pretty kick ass tool to build your web applications that is on the other side of the universe from stagnant.
    I look forward to hearing you speak at the CCCFUG here in Chicago in May Ben! Keep up the good work 😉

  16. I love the code portability discussion.
    I am in an environment where I have seen Java applications written for WebSphere and almost everytime an upgrade to the next version of WebSphere happens, the applications seem to break and weeks if not months of work goes into making these existing applications work with a newer version of the same product line. Now, the fact that moving something from WebSphere to JBoss takes even longer and cost a lot more changes my perspective on portablity.
    Another thing I have noticed, applications that have run fine, when you upgrade to Java 5 or to the latest version of framework always seem to have some quirk that takes days or weeks to resolve.
    I want to say I am not a Java basher, I am learning and living it as the opportunities provide themselves. I am just noting if I have to buy a new server and and rewrite large chunks of code for an upgrade that does not fit my definition of true portability.
    I have also been part of a migration of 4 years of CF5 applications and moved them to a MX7 and it took 1-2 days to get everything working. Coldfusion’s abstraction layer really made the migration very easy (except for the LDAP sort not working the same, I wish that was documented better as things to watch for).
    If we are talking about how I would spend my money, I would spend it on a tool that allows me to hire the young and hungry to get the results of senior and expensive. Better, Faster, Cheaper usually wins.

  17. RE: "The OO elitists just wont stand for that kind of thing!"
    The OOP elitists are easy to stop: Ask them for coded proof that OO is better in a business application setting. They can’t deliver! OO is not objectively better. It is just not. They will either use over-simplistic sub-types (the real world does NOT change in a tree-wise shape), or they will build complicated "navigational" structures which can be attacked for their complexity or for being in-app self-rolled databases. The OO’ers got nothing but vague brochure-talk. The practicality of their designs is easy to attack by an experienced developer who can toss real-world change scenarios at them that tie the OO designs in knots. Their shape, animal, and device-driver examples only carry them so far. Stop getting beat up by them: ask for real evidence, not shapes and animals.

  18. The problem with coldfusion is when your director says that he is going to implement sharepoint portal server. I have seen several companies do this. It pretty much kills
    CF. There needs to be more integration with Microft Products. I understand CF 8
    will have better integration with exchange server.
    In addition, Adobe should not release versions of the Product that is riddled with bugs.
    The issues in the earlier versions of CFMX with the JDBC drivers was a nightmare. Our
    coldfusion apps were dying constantly, creating a bad impression with our management.
    As far as the job market, I would say that there are not that many coldfusion jobs as compared with the other languages. It also seems that the job requirements are now to know
    every language available out there to get a low paying job.
    Overall, I think the wages are going to continue to drop in the programming industry. The
    big push is to replace us with the massive guest worker programs.
    Don’t get me wrong. I love CF and I am very passionate about it. However, because of
    the global economy, I am working on getting out of the profession. I have started my own
    company. By the way, the site is run by CF. I would not consider anything else. It is nice
    for a change to make the decisions of the platform that I will use. I know the benefits
    of using coldfusion. I would rather have applications get done quickly to improve the bottom line.
    Best Regards,

  19. Hi
    Is it possible to do multiple left outer joins on query of query.
    Have asked on forums if ColdFusion supports this, although there are workaround for single left outer joins on query of query, it is not what is required.
    It appears that no one has attempted to do this on multiple primary keys before!
    Can you tell me if this can be achieved?

  20. If you would just stop and look at the problem!!!!!!!! Just try and see the -8-n if you can possibly comprehend!!!!!!!! Not all genius is cognitive*

Leave a Reply