Thursday, August 5, 2010

FuseboxFramework.com Domains Are Going Away Soon...


Where did you get those URLs? The fuseboxframework.com domains are going
away soon (because no one in the Fusebox community was interested in
taking them on - and they expire in two weeks).

via groups.google.com (a comment by Sean Corfield)

This is in reference the domain being used by the what appears to be the "defunct" FuseNG project that forked from the Fusebox project. While these aren't the main fusebox.org domains, this prompted me to look at the status of the main Fusebox project.


I was surprised to learn that the last stable release for Fusebox was in March 2008 (version 5.5.1). Has it really been that long? It will be 2.5 years ago in just about a month or so without a release is stunning. Time flies.  Also, I quickly checked the Fusebox Trac site. Only one new ticket other than spam tickets has been filed or commented on in the past 6 months and that one ticket is just a question on syntax (it should have been sent to a list).  What is the state of Fusebox these days? Is TeraTech really "driving Fusebox forward and you can expect to see major improvements to the web site and the documentation in due course"? I don't see evidence of it on the site.


What really saddens me is the state of the CFML community. Have people not learned to pitch in and help their open source projects? Clearly not because there is still the glut of new one-man projects that never leave the ground or barely hover. CFML community members need to band to together instead of re-inventing the wheel. They need to learn to contribute (which I'd say that 99.9% of them do not) and realize that there is only a handful of people contributing to their open source project of choice. You don't have to pay for software with money; you can pay with your time, talent and expertise. Contributing does not always mean code but help on lists, documentation, sample applications, etc. There are so many things to do on open source project other than the next generation of code.


This is a call to arms! If you use open source, donate some time back to it or you might sadly find yourself with a defunct project and no maintainer there to help you. The great news is you can save yourself by contributing your time now. I urge all CFMLers to donate just 30 minutes a week to ONE project of your choice (if you don't know what to do, contact the maintainers -- I'm sure they have a laundry list of things to do). Just 2 hours a month would change the state of affairs in the CFML community and propel our language forward.


15 comments:

  1. Thanx for picking up on this Peter! Your call to arms echoes my Open Source Landscape talk (at cf.Objective() and CFUnited this year) where I tried to get folks excited about *collaborating* on open source projects.

    It's true: no new Fusebox releases since I stepped down two and a half years ago. I bought the fuseboxframework domains (all three: .com, .org and .net) back when we were wrestling with The Fusebox Corporation to get more control into the hands of the community - and it looked like we might have to fork Fusebox to keep it going. They (Hal Helms and John Quarto-von-Tividar) sold Fusebox to TeraTech and at first Michael Smith was very gung-ho about moving Fusebox forward and the shiny new website was the main product of that, along with the push to support an XML-less dialect of Fusebox.

    I stepped down because I hadn't used Fusebox on a client project for a long time and felt it would be better for someone actively using Fusebox to lead the project. When Adam Haskell got frustrated with TeraTech, it looked like forking was the only way forward and I pointed the domains at his new wiki / trac setup in preparation. As we now know, nothing happened with FuseNG after the initial flurry of controversy.

    Over the last year there have been various discussions on the Fusebox mailing lists about "getting things moving" again but the fact is that Fusebox is stable and mature and does what most people need so there's not much incentive to change things.

    4CFF looked at taking over Fusebox but discussions with Michael Smith didn't get anywhere (he wants to be paid quite a bit for the Fusebox name - which is fair since he paid Hal and John for it - but coming up with $5,000 for it just isn't going to happen!).

    When I asked the Fusebox community if anyone wanted to buy the domains, no one showed any interest. I've continued to pay for them for two years, even after leaving the project, but I've decided to switch them off this year since no one seems to care.

    ReplyDelete
  2. To be clear the fuseboxframework URLs existed for a long time before the FUseNG mess I caused. I made an official post about FuseNG ages ago, the interest was not there and, more importantly, life got in the way. Barney B made some updates at one point (to Fusebox), not sure if it ever made it into an official release, just like updates I made were not made into an official release. Personally I think all the frameworks are just silly. What the community needs are good solid open source projects like Razuna or what could have been Codex wiki, or Mura.

    ReplyDelete
  3. I thought I was clear about the timeline with this comment:

    "I bought the fuseboxframework domains (all three: .com, .org and .net) back when we were wrestling with The Fusebox Corporation to get more control into the hands of the community."

    But since Adam raises the issue, I'll concur and make sure there's no confusion: I bought the domains back in the Fusebox 5 days - 2006 - so we could fork Fusebox and release 5.0 if Hal and John didn't "get with the program" about openness and community involvement (to be fair to Hal it was mostly John objecting to the level of openness I wanted).

    ReplyDelete
  4. You were Sean, I was commenting at the same time as you. I only saw your comment after I posted :)

    ReplyDelete
  5. @Sean, I don't think that Adam was questioning the time line. Actually, I think his comment only served to reinforce that those domains were purchased before FuseNG was forked.

    One of the problems in the CFML community is the concept of openness. Even to this day most open source CFML projects are back door dealings. This is why Team Mach-II publishes our meeting minutes openly on Trac.

    ---

    @Adam, I don't agree that frameworks are silly however I do think that all the bickering about which is "the one to rule them all" is silly. That's why I don't compare frameworks with each other any more. I'd rather spend my time contributing to open source.

    I think you bring up a question indirectly of where should we spend our time as a community?

    ReplyDelete
  6. Adam - to your 2 point comment:

    1) Documentation: Great point. Model-Glue did a huge push on this about 6 months or so ago and I think the results were great.

    2) Rest Wrappers: Heh, as someone who has done a bunch of these (and enjoys doing it ;) I disagree. It's a minor point I guess - but I do think they are helpful. Rest is pretty darn easy - but not even having to parse the XML into CF is even easier. I wouldn't necessarily call that a project though. It's more like a fancier UDF. Also, I've seen some Rest APIs grow pretty complex, so having a simpler interface to them can be useful.

    Also - I'll add to Peter's point. I've got a lot of people _using_ my own projects. The number of people contributing? Next to none. I greatly appreciate those who do - but it is a rare occurrence. :(

    ReplyDelete
  7. I'd argue there is no excuse for a project not having a wiki, forums, etc. With RIAForge, GitHub, SourceForge, there are multiple places where a person can get support for stuff like that.

    ReplyDelete
  8. @chortlehood, it's all about initiative -- I've seen many people excited but you have to twist arms to get people to even blog about their accomplishments or what they did with XYZ.  However, we need to figure out how to motivate people.  We are all in trouble as the heavy contributors to open source in the CFML world won't want to be doing it forever.  I suspect Mach-II would have a hard time if Matt, Kurt and myself needed to hand the keys over.  I'm not sure who has the motivation to be a project leader and from personal experience it's hard mostly thankless job.  Most of the feedback we get is negative -- there is bug with this and shame on the team for not finding it or feature XYZ doesn't work the way I'd would have designed it.  The biggest thank you anybody can give me is to contribute their time back in exchange for all the time the team spends on the framework.  For example, Mach-II is valued at $1.25M or 22 person years using the COCOMO cost system using $55k per man year of work (http://www.mach-ii.com/index.cfm/go/about/).  This shows that nobody should say today that they want to re-invent the wheel!

    ReplyDelete
  9. @Brian, "the ColdFusion community doesn't contribute any less to open source projects than any other community" is a rather baseless assumption -- there is no way to back that as fact at the moment.  As for paying for CF, I threw that out as an idea because unlike all other community (minus .Net which is an unique beast itself) -- CF started and is a "pay for" product (only in the past couple years we have OS alternative engines).  There is nothing wrong with the fact that it's a pay for product, but I do think that does have influence on how the CFML community was formed and the "types" of people it attracts. Other languages like PHP, Ruby/Rails and Perl all started as free and think they initially attracted people that a) didn't have any money and b) they had to contribute to the language / project because that was the only way to get things done.  I think we can all agree that CF started in an entirely different manner and therefore the people that it attracted were of a different "type".  It leads the question of "why build your own if you can just buy it?"  An analogy could be that we all could have large gardens (like they did in pre-WWII times) to grow food, but these days we just go and buy it instead.  I'd say the CFML community are not the "gardeners" type of people in general and that is why we have problems with contributing to open source projects.

    ReplyDelete
  10. This is why it was presented as an "assumption" and not as fact. However, its no more or less baseless than assuming that they do contribute less - that is my point. :)

    ReplyDelete
  11. So obviously I'm biased, but I take issue with this statement, Peter:

    " As the CFML community, we're behind the times compared to PHP / Ruby / etc for great applications with *full* feature sets. "

    Define "full". You didn't call out BlogCFC, but let's say you did. It may not have the features of Wordpress, but it's been running my blog, successfully, for 5+ years, and many others as well.

    Is there some magical limit where we decide an app is 'complete' or 'full featured'? I would define a successful app as one that solves a problem and enables our clients to be successful. If an app does that, but doesn't equal something in the PHP world, does that mean that that particular CF OS app is somehow a failure? Or not a "real" OS app?

    ReplyDelete
  12. @Ray, I wasn't trying to single out any specific open source project but I think the real question is why would you choose BlogCFC over WordPress?  Getting down to brass tacks, but that choice would only matter to somebody that is using CFML.  Asking that question is *not* meant as slight against BlogCFC -- it's just about leaving languages on the side lines.  The point I'm making is we have very few (basically Mura and Razuna) open source projects that have the ability to compete against other open source project built in other languages when you looking at *all* the available solutions.  I like think about it as: we have very few *turnkey* open source applications when you take the which language part out of the equation.

    ReplyDelete
  13. To be clear, I'm not taking it as anti-blogcfc - just using it as an example. ;)

    So I think the issue is - you say it (or any CF os app) isn't "turnkey". Who defines that? Who says this app is as good as another? We could measure pure users as a metric, but really, as I said before, if an app is _working_ and filling the need, isn't it successful? By that metric, I'd say we have a lot of successful CF OS apps out there. _As_ successful as other languages, no, there is always room for improvement. But I don't think it makes sense to say there is some magical judgement by where an OS app is now considered turnkey/successful or even "done" really.

    ReplyDelete
  14. @Ray, no application is ever done -- come on -- we all *know* that. If you were looking the apps in the CFML in a "world" view, then I would agree that there are many "successful" applications. However we're lying to ourselves if we don't consider the "world" in a broader sense -- all languages.

    If you're interested, I'd check out a book by Geoffrey A. Moore called Crossing the Chasm:

    http://en.wikipedia.org/wiki/Crossing_the_Chasm

    In general if you apply this concept: "Moore believes visionaries and pragmatists have very different expectations, and he attempts to explore those differences and suggest techniques to successfully cross the "chasm," including choosing a target market, understanding the whole product concept, positioning the product, building a marketing strategy, choosing the most appropriate distribution channel and pricing."

    A nice visual to go along with that:

    http://en.wikipedia.org/wiki/File:Technology-Adoption-Lifecycle.png

    Suffice it to say some CFML open source apps have crossed the chasm in the CFML world; but that is not true when you apply crossing the chasm to in a broad IT sense (all languages and available apps).

    None of this is meant to diminish what's been accomplished in the CFML world so far, however we have a long way to go when you compare us to others. We have to look outside our little fiefdom or we're all screwed.

    ReplyDelete
  15. First off it's really not easy to see the comment box here(FF 3.6.8).

    Anyway I just wanted to say that I'm still down with Mach-II. M2 is no frills at least that's the way I see it.

    Build your library and be happy. A ginormous library.

    ReplyDelete