Things ColdFusion is not... and... Why ColdFusion isn't free...

Just want to subscribe to comments.

But, whether or not I agree with all you said here, this is an awesome level of commentary coming out of the CF team. Thanks, Jason.

This post clearly describes everything that the ColdFusion platform has to offer. I think what Adobe needs to do is convince technology decision makers that when ColdFusion is purchased, they are getting more than a language. They are getting a complete platform of server side technologies that has a very low cost of ownership.

The community will always support ColdFusion but if technology decision makers are not behind ColdFusion, I fear that our ranks may start to diminish as some are forced to use other technologies on a full-time basis. I don’t want the future of ColdFusion to be limited to the independent contractor and small business market (not that there’s anything wrong with those markets since I’m a part of them). I would like to see the number of large enterprise businesses that use ColdFusion double or triple. For me, that would be proof that big business has a better understating of the platform and the value it has to offer.

@Ben

Thanks for your comments!

I wasn't trying to say that syntax can't improve productivity and make things easier... in fact this is one of the big selling points of ColdFusion. What I was trying to suggest is that a loop is a loop... a try-catch block is a try catch block. I think a lot of developers using other languages see the tag-based syntax of ColdFusion as different... and my point is... it's just the language that's different! Who cares if you say

<!---start code block --->
<cfif true>
<cfset bar = 1>
<cfelse>
<cfset bar = 2>
</cfifc
<!--- end code block --->

versus

<!---start code block --->
if true {
bar = 1;
} else {
bar = 2;
}
<!--- end code block --->

I recognize that the second one has fewer characters... but the real bang for the buck is when I say
<!---start code block --->
<cfexchangecalendar action = "get" username="foo" password="bar123" server="127.0.0.1">
<!--- end code block --->

Let's see you try to do that with PHP! :)

My point is that being tag-based is unimportant when you compare it to the productivity gained when you go to use the built in services.

I am completely for continuing to improve the ColdFusion core language syntax. ;)

Jason

I think it's also worth bearing in mind that ColdFusion (shared) hosting is pretty reasonably priced and the differential between CF hosting and the "free technology" hosting is often only a few bucks per month. Even if you're paying an extra $100 a year for a shared CF plan vs, say, a PHP plan you are getting a huge value-add with CF. And cheap CF hosting can be found fairly easily these days.

Looking over the list of features, not many of those directly affect much of what I've been doing with CF in the past (but AJAX will play big moving forward) but you don't mention the high-level SQL handling which is a big productivity boost compared to most "free" technology and you don't mention event gateways (which have been a huge part of several of my projects).

It's Friday so I'll play the 'Bad Cop'... :)

For one I'd argue that several of these 'features' very few people use. Verity? Too difficult to setup and configure (esp. vspider). Exchange? I don't know - maybe too new but I certainly haven't seen a deluge of posts about people rushing to implement it. Server monitoring - if I'm hosting - that is the hosting companies responsibility. PDF - again too new.

For my daily tasks at work where I've been using CF for awhile now I mainly use the "Basic CFML" for 90% of my daily tasks. Maybe some charting. But I work for a huge corporation so the price argument there is moot - we are entrenched in CFML and the cost of CF (however high) is negligible compared to what it would cost us switch languages.

Blogging? Shopping cart? CMS? I run all of these using free, open source platforms and have no need for any of ColdFusion's advanced features so why should I spend a few more bucks a month on something I'm not going to use? So for me - a less expensive "Basic CMFL" would be ideal.

If your application requires and makes use of these advanced features than I'd say it's well worth the cost - if not - then I'd say ColdFusion is overpriced compared to other languages available.

The open source matter is another can of worms entirely. :)

Oops, I meant to say "I rarely, if ever, use those features you mentioned /Jason/". :\

I use ColdFusion because I like the language. I use it because it was the easiest for me to understand, and to jump into and start writing apps. I've never used charting, the exchange functionality, the pdf manipulation, the reporting, or many of the other more advanced features of ColdFusion. But I'd still pay for it as the core language makes me more productive. I don't care that it's not free...I just wish Adobe would put more marketing dollars into helping people realize all of what you just typed up.

I can't believe I let this one slip under the radar for a full week!

Look, if you want CF without the great features, use BlueDragon. If you want something free to do a basic web site - use PHP. You're right, you don't need ColdFusion to do the odd SQL query.

However, if you want to develop high quality web based business applications, then I disagree with the assertion that you don't need all of the ColdFusion features. Paper, no matter what you've heard, is still alive and well in Corporate America (and in the home!), so PDF and reporting features are very much needed. And really, there isn't enough room for me to talk about the newest CF 8 features, and how they make us business competitive.

While I would LOVE to see all of these features included for free (or as close to it as possible), the truth is that none of these features would be available at all if there wasn't a revenue stream. Quality costs. I've done cost comparisons of CF Enterprise vs. other competing platforms for versions 6, 7 and most recently, 8 - each time the math adds up in favor of CF, and that's before we even get to the subject of productivity and steep learning curves.

+1 Andy. Let's ramp up the marketing. On second thoughts, maybe don't. Because using ColdFusion allows me to gain an edge on my competition. Better features, delivered quicker.

And the cost? We're talking single-digit %age (per annum - I'll be somewhat vague, if you'll allow me) of our total development team costs - and that's a fixed fee, as we grow (as we have been) the percentages come down. That's a small price for a large competitive advantage.

Ask yourself this - would you be better off without ColdFusion? I wish there was a tool out there that did everything CF did, and was for free, but not in this version of realty.

Great post Jason - always good to get some deeper perspective on CF.

Cheers,

Davo

Ziggy - not sure your location, or what you term "expensive", but a quick search of "ColdFusion Hosting" found me hosting starting at $5. Am I missing something?

Cheers,

Davo

Syed,
Here is the link to $5 hsoting: http://hostfolio.com/web-hosting-plans-coldfusion....

I must respectfully disagree with your assertion that "if coldfusion is free, it would be better, for
everyone. " The features that we talk about as being selling points would not be available, if it were free. There would be no revenue stream to pay for the kick ass features.

CF *is* good for the enterprise, and there are other languages on the same level, but those languages cost more money to install and deploy than ColdFusion. Look at the BEA and WebSphere productivity suites.

ColdFusion is niche - I don't have a problem with that. I think my investment in the language gives me a competitive advantage, and so I don't care if my competitors don't use it (I'm sure Adobe minds ;-)

ColdFusion won't go away just because it's niche - it will be dropped as a product when it stops earning money, and all signs point towards a healthy and expanding product base. So, with respect, I don't see how "free" CF would be better for everyone - I'm afraid I'll have to disagree with you there.

Cheers,

Davo

>>The AJAX stuff and the Flex integration stuff are going to keep ColdFusion current

Well, I still hate Flash apps and you don't need Coldfusion to do Ajax. And all of the built-in GUI stuff is terrible, unless you like simple pages weighing 700kbs. IF there were some well-thought out helpers, OK. I haven't seen anything yet that isn't much better done outside the CF code.

>>a quick search of "ColdFusion Hosting" found me hosting starting at $5. Am I missing something?

Yeah, like a lot of hosts and really good features for cheap just like you see with non-cf hosting. Especially overseas.

>>I don't see how "free" CF would be better for everyone

A good free version would generate a mass of users. Try to sell a CF app - who wants to buy it when they find out they have to hunt around and then pay extra for "special" hosting? Adobe can still do the enterprise thing for serious money. Best of both worlds?

Gus - I don't think anyone is arguing against the high cost for enterprises who use all the features. And yes - I've built sites for myself, my wife, and several non-profits - all for free.

What would be nice is a lower/free alternative so people could build basic personal, non-profit, or small business sites on ColdFusion. This would also tie in well with the educational, young developer market (see Ray Camden's recent teacher interview).

Personally I run Ubuntu, Open Office and the GIMP because I don't need (or care to pay for) the features in the software you mentioned. Am I missing out on anything because it's all free? I don't think so.

I REALLY like the idea of Adobe supporting a project like Smith - that would be the best of both worlds and would earn a lot of kudos from the open source world as well.

As Sean mentioned - free != open source. But free does not necessarily mean you give up your revenue - look at companies like MySQL - they offer a free alternative but also charge for support and additional features. I don't see why the same business model wouldn't work for ColdFusion. Want basic CFML and that's it? Free. Want PDF, Exchange integration, reporting, etc. Then pay up and we'll provide support as well.

I remember on one of the "ColdFusion is dead" threads I mentioned a-la-carte pricing and I think someone (Tim Buntel?) indicated they had looked at that - and no one liked it... (I can't remember his exact response).

So Adobe - throw some support at Smith, and release a free, basic, open source version of the base CFML application server. Develop an educational program - get it in the hands of high schools and colleges. These will be the people when they graduate and get hired - that will come back and shell out $$ to run CF in their companies.

So everyone is pushing for an Adobe free version of CF - but hasn't New Atlanta Blue Dragon cornered that market? Hands up - everyone who says they just need the bare bones CFML syntax, and uses Blue Dragon. John (Farrar) - wouldn't this put you on an equal footing when bidding on a contract? (Quote: "When we bid a contract we loose sales because there is no free version."). Oh, and BTW, we use BlueDragon in my company, for an internal application, that doesn't need all the bells and whistles. Works out great. Why would Adobe compete in that space?

@Ziggy - I did a VERY quick Yahoo! search and found $5 ColdFusion hosting, I'm sure with a little time, I could find many more.

@Gus - +1 dude, no make that ++1 - great points.

David... you better review your "Free" BlueDragon lisc. It isn't universally free unless it changed. If that is incorrect then that is something worth noting for sure. (Yet, somehow it doesn't seem likely that BD is kicking out a 100% free product to boost Adobe sales.)

Now BD does offer some excellent deals where you write products written with BD server and you can distribute them as a package for very reasonable price. Heh, I think we have a better chance getting Adobe to give us a free version of the Pro/Std CF then that happening. Yet, I am open to the possibilities!

New Atlanta changed their license around the 6.2 release so that it is no longer free for commercial use, only free for non-commercial use but it's not quite clear what they mean by that:

"...development, testing, non-commercial and non-governmental deployment purposes."

"BlueDragon Server may not be: i) bundled or distributed with yours or any third party application(s), ii) used to provide commercial web hosting services or, iii) used to provide commercial services in an Application Service Provider (ASP) environment, ..."

They do specifically call out SSL in relation to commercial use but it's not clear whether you could legally run a commercial / business website on the free edition if you had no secure sections of the site...

Dugg: http://digg.com/programming/Things_ColdFusion_is_n...

I'd like to bolster Mike Kelp's arguments about CF in education. Mike graduated from UNT within the last few years. I left UNT in 1999 for a full time ColdFusion job.

I started using CF in 1997 while working for the School of Visual Arts. It wasn't affordable for SOVA to buy their own CF server, so I was given an account on the single CF server the computing department had. In the years since, ColdFusion has become more widely used around UNT (check out http://unt.edu), but I'm sure that FrontPage, or whatever MS product has succeeded FP, is still the website product of choice for the majority of campus departments.

Before I left UNT, Microsoft had a blanket license with the University. Students could buy most any MS product for the cost of burning a CD at the computing center. The bulk Adobe licenses made it very affordable for the University to install Photoshop and other Adobe programs across multiple computer labs using some kind of concurrency system to make sure only X number of users were running the program at any time. The "Academic versions" available to students (full versions not for commercial use) were very affordable too.

If you go to http://www.adobe.com/education/purchasing/educatio..., you'll see that Adobe still has some serious discounts for students and academic institutions. But even at discount, CF Standard (at $860) isn't going to get it installed on the student development servers. Not as a standard program, at least.

When most students leave UNT, or most any other University I'm sure, they know how to use MS-Office. Many even know how to use Photoshop or at least what it is. Ask any group of students about C++ or Java and I'm sure you'll have at least one that can tell you its a programming language. Mention PHP and someone will know you can make websites with it.

Ask any group of students about ColdFusion and they'll tell you about nuclear energy or start quoting Star Trek.

Hi Sean - I've read many sections of the site, and it says everywhere that deployment of BD server is "free" - I can understand them not wanting to allow you to package and sell products based on their free CFML engine.

Still, I'll have to investigate more to make sure we're in compliance - thanks.

@Scott - As a matter of interest, what order of magnitude are we talking about for these intranets? 5k, 10k, 20k? And maintenance/upgrades?

DW - great points - I see a free version of BD that's just plodding along, and is devoid of any major innovation. For a while there it was just trailing CF, and billing itself as a cheaper alternative. But CF8 just blows it away. It's hard to justify major investment to a product thats free, right? So, why do some people think it would be any different if Adobe adopted the same business model?

Cheers,

Davo

I found the post a bit arrogant. The rhetoric makes me believe that asking for a free ColdFusion version is an absurd. Something that we, as CF developers, should forget for good. It sounds that Adobe (and Adobe representatives) are telling us to forget everything we know about ColdFusion, how to market and sell it in the "battle field". It sounds that we should stop complaining and accept ColdFusion and its price as "gift", not as any other Adobe product (lets say... Flex). If ColdFusion is so good and so cheap, why its not the leading server technology tool/product/whatever (and I'm not considering the open/free ones)? Are we so stupid that we can't see the true behind the numbers?? Are our customers wrong when they say CF is expensive?? According to what I've read here, yes, we are all wrong!! That's what this post makes me feel about... and thats sad, especially when it cames from the CF product manager.

As for the math, prices, listings and the whole logic/arguments behind this post I have to say that IMHO (read again: IMHO):

1) Application Server (Jrun): I don't need it (specially after Adobe dropped it...errr, sorry, "ended new feature development plans"). As long as CF stays a J2EE compliance application, I can run it atop of Tomcat/JBoss/etc;

2) Reports (Jasper): What about DataVision, Pentaho, OpenRPT and many other similar products (cheaper or opensource)? Can't they be bundled with CF? I'm sure it can, just as other opensource products that are already bundled in ColdFusion - ie. Apache Axis).

3) PDF Generation. Seriously? There are plenty of alternatives out there (expensive/cheap or even free)!! PDF Forms? Sorry, I never used and I don't see it many people filling PDF forms... Sell it for the Enterprise, they love paper... And don't forget: what about LiveCycle? Is ColdFusion going to replace it as the Adobe PDF server solution? I don't think so.

4) RIA Development? Same question: is CF going to replace Flex? ExtJS for $6,999??? Are you kidding me? Sorry, but for me, as a commercial developer, it costs only $219 for unlimited use/clients... maybe for Adobe it costs $6,999, but isn't because Adobe is selling it bundled with a product like ColdFusion? I don't now, but I know much does a CFEnterprise licence costs. And finally, lets be honest, you can't use the full ExtJS features with CF. ExtJS Implementation in CF is very basic in many ways. What about Yahoo! User Interface Library (YUI)? Free. JSON? Free. Spry? Free (as long as Adobe keeps it that way). Flex? Well... "free". A parenthesis: why not make an free "Coldfusion SDK" just Adobe did with Flex? What happened to Macromedia/Adobe to change their minds so radically about Flex licensing and market positioning? What happened to the technology after that? Did it grow up? Did Adobe made more or less money with it? Can the same happen/be applied for Coldfusion? Why not?

5) Server monitoring? Why do I need this? As a developer my concern is to write good code, not to monitor a server. Sell it to the hosting companies or for those companies that can't make their developers write good and steady code (or use a stable product)... Ouch! Yes, despite of the many critics in that particular area, I'm not saying CF is a unstable product, it is not. But I'm pretty sure that Server Monitor feature in the CF8 is here because CF gets a lot of trouble when it cames to performance. And it's not me that is saying this bul*!! Is the people that buys ColdFusion and asks for a server monitoring tool. For what? Ask them...

6) Debugging? Hmmm, nice one, but sorry, this is not a "plus". Ok, CF has very good debugging capabilities compared to other "Basic interpreters/compilers". But the truth is that any decent language interpretor/compiler offers debugging.

7) Search Engine? You mean Verity? Sorry, I do prefer Lucene and other better (and newer) products/technologies. I never used Verity or K2, seriously.

8) Encryption algorithms? Like in SSL, if i want a SSL "brand" to make my customers feel safer, I'll buy a CertiSign certificate. Sell it for those who want (and need this). And can provide a good level of security using the basic/open/free encryption algorithms available.

You forgot to mention .NET integration. But again, that's something I personally don't need and I'm sure we can make it in different ways (otherwise Macromedia will not make .NET integration as a selling point/argument for previous CF versions as it did it - as long we have webservices, CF can be integrated to any technology)... (and be sold like that)

Bottom line is: ColdFusion is definitively a great product. But it's time for Adobe to rethink its licensing, versions, prices and market positioning. I also think its time for Adobe to listen to what developers has to saying (and they do it for very long time. For me, this post is an indication that in fact the prays and complains bothers Adobe executives somehow and its a good thing (not exactly the way it was done by this post).

I believe that what they made with Flex is a good start. Why not?

Sorry for the irony and sarcasm, but I didn't like the approach Jason took. Maybe it's because my English (and understanding) is not that good. Its just that it doesn't feel right for those (like me) who ask for changes in that particular area. We're not that wrong. Trust me.

@all:

My experience is in Enterprise intranets, and my perspective is completely different. We don't use all of the features of Coldfusion, but I can safely say that we made back the cost of the Enterprise licenses within a few weeks. The simple reason is that labor costs are ongoing, while the license purchases were a one off. With CF we do not need J2EE experts, security experts, Lucene experts, etc. We only need to employ CF developers (even they are tough to come by, however).

@Alex:

"Server monitoring? Why do I need this? As a developer my concern is to write good code, not to monitor a server...Is the people that buys ColdFusion and asks for a server monitoring tool. For what? Ask them..."

For what? To isolate the crappy code that you just wrote because you didn't use the same tool to profile your application! :)

Particularly in response to Mke's comment about use of CF within educatioal sector...

As far as using CF for development, CF Server is free for access from 2 IP addresses.

Best practice would dictate that CF development should be done against a local server anyway - rather than a shared development server.

Is there any reason why this could not be used within the educational sector. Addmitedly a slightly better spec of machine is needed for local dev than remote - but not that much better (certainly I normally find CFEclipse far more of a drain on resources than CFServer on development machines)

I have set up a virtual machine (using VMWare Player) running ubuntu, CFEclipse, Apache and CFServer 8 - and this runs quite happily with 750 mb RAM assigned to the virtual machine, on a 3 year old centrio 1.6Ghz laptop. This runs everything I need, except a DB server - I connect to a shared development server for this.

Could a solution like this potentially be used within the educational sector?

@Alex Hubner

Coldfusion is not stable. i have yet to install Coldfusion and have even acceptable stability. I am forced to use Coldfusion because of the company I work at and would happliy move to another language if we could. Some issues I encounter are:
1. Persistent JRUN errors (memory usage is terrible)
2. null null errors. Seriously, how do these exist. Sometimes i feel that we upgrade to newer versions of Coldfusion (forking out silly money) just to get some updated error catching.
3. Output buffer locking. I am glad that we are running Fusion Reactor now because before when the server hung we had to restart the service. Now I just have to kill all the processes. All the time I get pages that would never time out.

Now here's some good things I like about Coldfusion.
1. cfquery is damned easy and other languages should strive to implement queriying a database this easily.
2. The integrated apps such as cfchart and cfldap. Although they are not as robust as I would liek them to be they make simple things simple. I would recommend that Adobe find some way of letting developers extend these features in someway.
3. cfloop over a query. Another forehead slappingly easy way of working with data.

Things I don't like about Coldfusion:
1. Why doesn't cfloop have the group attribute? It really sucks to have to close out a cfoutput just so that I can use cfoutput query,group.
2. Verity really sucks. Sure it's worls for indexing some files but vspider is what any site really needs and it's implementation bites. As they say, s**t or get off the pot. How about some kind of extensibility in integrating other products?
3. The community. Where is it? I always feel like I have to reinvent the wheel when writing code. cflib is nice for finding simple udf's, RIAForge can be cool 1% of the time but how come when I use PHP I can find anything I want to do what I want? why is it that the CFDocs are so incomplete? Just so Ben forta can sell another few books?
4. Has the cfloop inside of a cfloop bug ever been fixed?
5. cfmail is so unreliable. One error log is the only info you'll ever get info from on what's going on with the e-mails your trying to send. Adobe should just buy activmail from zrinity already.
6. cfpop so slow. cfx_pop3 anyone.
7. why are all the includes in the cfide directory? Might I suggest creating a thord cfincludes directory cause it just seems a little unsecure to me.

Not really a Coldfusion problem but def a macromedia/adobe problem. (I'm on a ranting roll)
Dreamweaver is terrible. Stop creating bloatware. If they took half the time they spend figuring out new useless features for a silly WYSIWYG editor fixing the bugs they might have a usable product. I'm a coder so I'm a little biased and I'm sure that more money comes from the WYSIWYG side.

I have tried using Dreamweaver several times and each time (not counting the failed installs and constant crashing) I uninstall it out of frustration. I currently use Homesite because it is atleast a little more stable and a lot more code centric. Plus FTP doesn't die whenever I try to use it.

Come one guys give us Homesite 6+ please, and don't you dare bundle it with Dreamweaver again. I'll give the developers a hint. Take what you currently have and perfect it. Look at Textmate for ideas on how to write an IDE. Listen to developers, Homesite and Dreamweaver are two seperate products and should stay that way (unless somehow the Dreamweaver team makes a good IDE) Oh and eclipse is not a webapp IDE I'm sure it's great for compilable projects but trying to use it for CF is not that fun. Maybe just take Visual Studio's code editor, add in some Homesite bug fixed file management and a little bit of Textmate's extendability and yay, the perfect IDE.

Sorry for the off topic, omce I pop I just can't stop.

If ColdFusion were a girl, I would marry her.

I have to agree with Dan Lancelot in addressing Mike Kelp's concerns about CF in the educational sector. Why do you have to buy a "real" license to teach Coldfusion?
I took a ColdFusion class as part of a web dev certificate program a few years ago. All of the workstations in the class had the (free) CF developer license installed and we all worked on our local machines. It worked out great.

@All

These are spectacular comments... please continue to post... I assure you this post will be read by influencial people...

Also, I wanted to make it clear that I was in no way suggesting that there could/would never be a free edition. In fact, the first part of my discussion around it was "I can see Adobe making the core language for ColdFusion free some day". Anytime I post, it is not to show a decision has been made... it's to get feedback to help solidify or adjust my opinion... and to get from you the arguments that I need to make if I want to help make these changes.

Here are the arguments I see for a commercial free edition:
- ColdFusion would be used more in basic applications
- and it would be used more in contract settings (you need to bid out developing a site for someone... and can't eat the cost)
- The hope is that this would increase the number of developers using it over time...

Here are the arguments against a commercial free edition:
- CF would lose revenue from hosting markets, and from basic sites running Standard today
- The tag-based nature of CFML is fundamentally different from the script syntax used in many other languages... so we're not going to get the PHP developer just by making it free
- Macromedia did a CF Express edition (and CF Standard for $699)... and they did nothing to increase the number of developers... just reduced revenue...

To be clear of my intentions... I AM A CF DEVELOPER!!! As a CF developer, I want CF to be free too. But more than wanting CF to be free, I want the business of CF to be healthy so that we can all enjoy having this language around for years to come. I'm being honest when I say, as the ColdFusion Product Manager, I am on the fence as to whether a commercial free edition would be good for the health of ColdFusion. It seems like a very grey area to me...

There were a lot of things that cost quite a bit of money that I didn't mention for the simple fact of keeping the article shorter. One was mentioned by Dan Wilson... WebCharts... there are several others and we are still increasing that list... we also leverage several other Adobe products that we still treat with the same respect... Adobe needs to make money to pay for that development as well... not just the core CF development.

Someone pointed out iText... well, you can see an iText.jar in the lib directory so that's clear that CF does use iText... but the PDF features of ColdFusion use CONSIDERABLY MORE than just iText. I think some of you may be underestimating the effort that the CF team puts into this product. CF is not an inexpensive product to produce... in any way.

Doing a free edition the right way would take a lot of time and effort that would detract from innovation within the product. You can't just say "here's a free version"... editioning is very complex... among other things, it would require touching almost every file in the product. Some of the other languages or clone products out there don't have that complexity, either because they were able to build from scratch with that in mind, or don't have existing contracts or production costs to deal with...

I don't have an answer... I do think a free edition of some sort would be very expensive to create (which is ironic really).. and would be helpful in some areas and harmful in others...

Please keep posting your thoughts... and helping me form mine.

Thanks!
Jason

@Troy: Adobe released Flex SDK for free. Who do you argument that in terms of revenue for Adobe?

@All: I think Flex is a good example/path to follow. Why Adobe released the Flex SDK for free?

What about making the same for ColdFusion? Release the core/compiler with some basic features that depends only on opensource 3rd party products/technologies for *free* (not opensource - see ahead!) and sell one (or various) "enterprise" package(s) containing all the advanced settings (including security, searching, JDBC drivers, RIA forms, Exchange, etc, etc), much like what happens with Flex and it's components (Data Components for instance). Ah... And of course... this also makes a good excuse to build an IDE (a long-time CFdevelopers demand) and sell it, just like Flex Builder... ;-)

Regarding Opensource ColdFusion: I don't think thats a good idea. Unfortunately I think that our community is not big enough nor have the technical expertise (of course, there are many exceptions) to maintain a product such as ColdFusion whitout a heavy dependency on Adobe or any other big software company. CFEclipse as a good example. It's opensource and its widely used/adopted among us, but it gets very few effective (read: people who know how to code and incorporate new features, fix bugs, etc) help from the community. Mark Drew is the one that makes almost everything on the project.

But when it cames to release ColdFusion as a free compiler/library just as Flex SDK is, I think it's a very good idea. Microsoft did that with .NET framework, many others did (including Adobe with Flex)... why not making the same thing with CF? I believe that the immediate loss of revenues can be compensated on the medium and long term (including support, although I don't think this is expressive). If Adobe see a future for ColdFusion as a product/technology, not as another good-selling product, I think thats the best it can do. But that's is only my opinion, and I might be wrong. Would be nice to see other comments.

@Jason,
Thanks for the post - it's great to get your insight into the matter. Interesting about CF Express - I never knew that. Obviously "free" isn't enough, it needs to be free AND kickass to attract developers. Beware the "yeah but" people out there. Every positive can be spun as a negative - I can see the headlines now "Adobe releases sub par web server product - thinks it can attract developers". It was like the linux zealots, when Flash/Flex was released (late) for Linux - "Yeah but...it's not 64 bit" was all they could say. It's interesting the scope of the task, you mentioned. Is the juice worth the squeeze? If the upgrade from 8->9 is as good as 7->8, charge me!

Having said that, there were some great suggestions made here, the two I like best:
1) Free licenses to Education - with some obvious caveats, of course - as in, it needs to be used for education!
2) Support of the Smith project (or something like it).

I'm sure devoting real resources to the Smith project would benefit it immensely, be some goodwill to the community, and give Adobe the chance to point developers to a "free" version, without having to go through the "edition" process you mentioned in your post, with the existing CF codebase.

It would also make a statement - Adobe isn't afraid of a scaled down free version of its product. The real meat is still in the boxed product, and when your business needs mature past the basic CRUD level, you can easily upgrade.

Just MHO.

Cheers,

Davo

I think the real reason why there isn't a free version of CF was not discussed in the original post, but in a later response. Allaire/Macromedia released a free version (CFExpress) and didn't make any money from it. I'm sure short-term sales reflected this, although I'm not sure it accurately reflects the long-term benefits for CF. We'll have to accept that statement and assume past is a predictor of the future.

So is there a legitimate business case for a free/cheap CF version now? The CF clones, BlueDragon, Railo, etc. are already out there so why should Adobe even bother? If I start my project in a cheaper clone and it works, why would I move up to CF. And if I run into compatibility issues when I try to move to CF, how do I solve that problem - rewrite a bunch of code? or abandon CF and work with the clone vendor will implement whatever feature I looking for from CF. Adobe loses these customers that become entrenched in a clone.

Until Adobe releases a language standard for CFML (that'll be the day) then CF is the reference standard. Sorry clones, but I'd rather go with CF than a company attempting to reverse engineer. Not to mention, I'm pretty sure Adobe will be around in a few years. I'm not so sure about some of the clone projects out there.

So back to making money - or at least not losing money. What's a basic version of CF worth to me? Well PHP is free but I think CF is much easier so even for a basic project that's worth something more to me. ASP/ ASP.net is neat (especially with the free Express IDE thrown in), but that is somehow worked into the cost of an OS. Lees than $200 for IIS on XP Pro or Vists Business with 10 concurrent users, or $400 without user limit. If I'm just looking to interact with a database on an intranet, then in an office that is deploying MS Office, the incremental cost for an Access license is in the $100 or less region. (And that's actually pretty easy to get basic functionality out of.)

So, lets start the bidding. I'll pay $100 for CF with basic tag support like database (CFQUERY) and file (CFFILE) interaction. No usage restrictions (i.e. full commercial rights) but throttled to 10 concurrent users. When I want to scale up, because the application is indispensable but overloaded, I'll pay to move up to Standard and know I'm 100% compatible. What's your price point and feature set?

David B

@David,

You successfully made and refuted the clones argument in the same post. Sure, we can use one of those guys for free, but there's always the compatiblity issue. Not to mention, those guys are not known outside of the CF community. If you ask a PHP/ASP developer if he's heard of ColdFusion, most will say yes. Ask them about BlueDragon or Railo, and you'll get a blank stare.

The reason this is relevant is when you start talking about market share. ColdFusion has a very small slice of the market. Every time I have presented ColdFusion to a group web developers, they ALWAYS say it's too expensive. If Adobe truly cares about market share, they are going to have to get over that hump, some how. Is free the best answer? Not sure, but it seems like it. And I'm not even sure Adobe cares about ColdFusion's market share...

The argument about ColdFusion being 'too expensive' is spurious.

*Anything* that is not free can be looked at at 'too expensive'. In marketing, they teach that unless the customers think the price is too expensive, it isn't priced high enough.

ColdFusion might be 'too expensive' to dominate the Web Programming landscape, I'll agree to that, though it is quite late in the game to try and take over the Web Programming world. ColdFusion fought that battle in the late 90's and did not win. It has been relegated to a Rapid Development Framework with corporate support and handily wins that category.

It makes sense to give a product away when a competitor relies on a comparable product as a strategic product and damaging that competitor reduces their ability to attack ones one strategic products. Right now, web programming platforms and languages are already commoditized. Adobe is one of the few vendors that makes money on a Web Programming platform and one of the most to lose by making ColdFusion free.

If I ran Adobe, I would not focus on the individual web page builder who needs a contact form, I would focus on the enterprise that has money to spend, and lots of disintegrated systems. The kinds of developers needed for this aren't coming from the 'I build Contact Forms' type developers. Trust me, Adobe has enough swag hungry people clamoring for T-Shirts.

I would keep giving hooks into such compelling products as Flex LiveCycle DataServices expecting to drive sales of these more expensive and unique products in the marketplace.

If, by chance, Adobe saw strategy in the "I build Contact Form" guys, buy out Railo, make it the R&D wing of ColdFusion and use a model such as Open Solaris. Sure the code isn't as concrete and solid as Solaris, but man does it get awesome QA and a rabid fanbase.

Adobe isn't a loose knit Open Source organization, it is a publically held, for profit corporation. It must act as a responsible publically held corporation and stay accountable to the shareholders else risk being bought out by the competition. If you don't agree with what Adobe does with ColdFusion now, what would you expect Microsoft to do if it purchased Adobe?

DW

Why did I choose Coldfusion?
I chose Coldfusion in 2001 when I decided to change my Flash website from a static website with some simple information, to a website with a database, where people could add and update content.

What does that have to do with Coldfusion?
In 2001, I found a great Flash/CF example+article on the Macromedia website with well-written documents and sample code. CF was the fastest way for me to pick-up a scripting language and build a website. I had a great first experience with Coldfusion. It made me feel smart and capable.

7 years (and 25lbs later), my experience isn't as positive as it was back then.

I'm a Flex Developer, and the tools and experiences AS 3.0 and Flex has taught me have influenced my preferences.

IDE:
I value a good IDE. It saves me time. For the $1300 I just spent on CF8, I want an Eclipse IDE that stops me from making mistakes, just like the Flex Builder IDE does. For example, a CF IDE should:
- validate the scope of my variables.
-tell me what class paths are invalid before I run the script.
-provide name-spaces to classes as I type.
-give me real refactoring because refactoring saves me time, especially in a real Agile environment. I want to drag methods and classes around when refactoring, and all of my name-spaces should update themselves. I should be able to rename Class names, and a real IDE should update class references.

Syntax:
CF is verbose, but that's okay, provided the verbosity provides clarity, but with CF, it doesn't always make sense. Many of the CF Tags need to be updated (CFSET and StructNew() don't make sense to developers coming from other languages).

Syntax:
The syntax was cool when HTML ruled the web.

Now, HTML is out. XHTML is in. XHMTL is XML, and XML makes sense to everyone, including entry-level programmers and veteran programmers. By migrating CFML to an XML syntax, the language would be easy to learn and would provide a visual hierarchy about what code is doing.

Attributes and Features
CF has too many attributes that no one uses. CF has too many features no one uses (Flash Forms, anyone?). Backwards compatibility slows me down when I have to scroll through tags and docs that are, and will always be, irrelevant to me (Who uses the Java applets?).

CF Admin and Configuration:
Configuring CF takes too much time. It's too complex. It frustrates me. It makes me feel bad. I want configuring CF to be simple. I want combo-boxes and check-boxes, not XML files and hours pouring over docs. I want to define a virtual local site in ten seconds, just like I can with Ruby.

When it comes to cost, I'll pay well for any platform that helps me make a better website, faster. Flex helps me deliver a user experience that is richer than what my competitor offers. So does Spry. For me, I need a better user-experience with my middle-ware.

Cost:
If you're stressing over the $35 a month for hosting a website, you should skip the $90 a month Latte habit or drive slower to work. Trying to point-fingers at cost makes no sense to me, since the tool we're choosing is chosen based on features and productivity. I paid more money for my car because I wanted more features. If cost ruled every purchase decision I made, I'd live in a box and take the bus to the office. People purchase products based on status, convenience, and experience.

Price does not tell the CF story. Price is not the problem. The developer experience is the story and the story is the problem.

Only two points to make (and they may well have been made already)....

1) We all know the details of CF, and how much value it provides, but no-one else seems to... GET THE MARKETING DEPARTMENT TALKING ABOUT COLDFUSION. Get the general opinion of CF in line with what you describe above. Do this, and not only will you sell more, but maybe even reduce the price even more! I'm fed up with reading non-CF devs talk about CF in a way that shows they've not heard about it since CF4.5. This is really a big problem for CF today.

2) Give options to remove functionality. For instance, I might not want Verity or PDF functionality - so give me the option of knocking off 20% and give me a special license key.

For me, the problem we have at the moment is that to the ignorant (i.e those that pay the bills), there is still a several thousand dollar difference between CF and something like PHP/ASP.NET, and both are better known and widely available. This needs fixing from an awareness point of view, before CF becomes truly marketable by the small web shops where it will make the most difference.

Ran across this article that's relevant to this discussion:
http://www.news.com/Software-licensing-costs-predi...
A salient quote: "Although Gartner says open source won't topple the likes of IBM and Microsoft, the firm believes that it will put pressure on traditional software margin structures, particularly in areas such as servers, operating systems, development tools, and database technologies."