Gary Bentley wrote:
Jonathan,

I was wondering (I don't have anything to do with the Velocity project, I receive this list because I am writing a tool that will allow one my OS projects to be used within Velocity), why not use your energies to lobby Apache to take FreeMarker on-board?

What for? I have no interest in that. And, especially, now that FreeMarker has the critical mass it has, it would be almost superfluous. There would have been more of a tradeoff, pro vs. con, back when FreeMarker was little known. But at this stage, there are mostly just cons.

And, in general, I don't want anything to do with ASF. I consider it to be an extremely dysfunctional organization, at least if you take the idea that it is a non-profit foundation that exists for the public benefit. (If the point of it is to promote the members and their work, independently of merit, then the organization works quite well, but that is mostly to the detriment of the public benefit.)

Oh, actually, I wrote a somewhat humorous blog entry about that subject of potentially joining ASF. See: http://freemarker.blogspot.com/2006/02/some-comments-on-joining-jakarta-or.html


You have indicated that FreeMarker is more advanced and mature than Velocity (and from the limited reading I have done it appears that way) but your efforts on the list seem like wasted effort to me.

Well, what's your basis for thinking that? I mean, there has been a huge migration of users from Velocity towards FreeMarker. Even other Apache projects, which are (though somebody may pop up and deny it) under some pressure to use other ASF projects, tend increasingly to use FreeMarker for templating. Struts 2, Open for Business, for example. I think there are one or two others.

Now, I'm not saying that we've reached this point because I occasionally speak my mind on this forum. But OTOH, I don't think you have much evidence that it is ineffective. There is not much evidence either which way on that.

From what I've read the meat of your argument is that you are unhappy that Velocity is considered the de-facto template engine to use because it has the Apache badge

Well, that's only the case among people for whom that Apache badge means a lot. And, basically, for technically informed people, it surely doesn't mean a lot nowadays. Probably, among people who do some due dilligence on investigating their tool set, FreeMarker is the overwhelming choice by now. You know, the fact that such projects as Netbeans, Hibernate, and Webwork (which is now "Struts 2") used Velocity at some point in the past and then switched to FreeMarker, this tells you something. Any minimal amount of investigation you do by googling around, I think you'll quickly see what's going on. You can find all kinds of cases of people (those very prominent projects among them) who have switched from Velocity to FreeMarker. And you cannot find a single example of anybody going in the other direction. Try it...


and that Joe User doesn't realize that there are better alternatives, if this is the case then an obvious remedy would be to get the FreeMarker project the Apache badge .

Well, there's some logic to that, but it's a perverse kind of logic. I mean, it's like you have a local farmer's market, and you can buy wonderful fresh fruits and vegetables there. You buy there, but your kids won't eat the stuff. They will only eat fruits and vegetables that come out of a tin can, made by some corporations, that market this tinned food during the cartoons during children's TV shows, using cute cartoon characters.

So how do you get the kids to eat the fresh local fruits and vegetables when all they want to eat is the crap from the tin? Well, you figure it out. You convince them that the fresh fruit and vegetables actually came from the tin can and then they'll happily eat it. Yum, yum. (I think it might be better to just try beat some sense into the little brats, but that's not politically correct....)

I mean, that characterization is about right. Look at what happened with the Struts project. It went nowhere for years, got ever increasingly obsolete. Even without improving for 4 years or so, Struts in 2005 was by far the most popular web application framework. A whole small industry existed around the darned thing: books, training courses, add-on plugins and so on... The problem was that ASF, being the dysfunctional organization it is, the project was basically in a state of abandonment. The committers did no real work on it, and if anybody showed up with a fire in their belly to do some work on it, they gave them the runaround. (You can verify all this, it's true.)

Anyway, what happened? They got the developers of a competing web application framework, one that was far technically superior to Struts, to donate their work to ASF, and it got relabelled as Struts 2. This was the deal the Webwork people made to get visibility for their work. How is this different from the fresh vegetables scenario I described? It's as if the majority of developers would only use the far superior framework if it got relabelled with the name of the far inferior one.

At the time that was happening, I wrote a blog entry with my reaction. That's here:
http://freemarker.blogspot.com/2006/03/musings-about-competition-ego-open.html

It's interesting to reflect on all the psychology behind this and the perverse group behavior that makes such a story possible, but one could go on about that and not fully explain the phenomenon. It's a tangent but I'll share my own theory on it. Basically, I think it's a certain kind of herding thing and the herd tends to herd around certain reference points, even arbitrary ones, and of course, once the herding starts in earnest around such a point of reference, call it a focal point, it becomes a self-sustaining mechanism. You know, people use something because the other people are using it, which in turn becomes a reason for other people to use it, and it becomes perceived as the safe, normal, default thing to do.

But whether my explanation for the phenomenon is right or not, it means that, for whatever reasons ASF has become this sort of reference point for so many people. And that is why something as technically inferior and obsolete as Velocity, say, can still attract a lot of users.

Now, the thing is that if you have a private company, and the goal of that company is simply to make money, it's great to have a brand-name like that, so strong that it allows you to get away with selling obsolete technology because people want so bad to have that brand name.

But if the point of something like the Apache Software Foundation is to be a non-profit entity for the public benefit, well, the ability to get people to adopt inferior, obsolete technology is not really a success, because you were not a company trying to make money and doing that by marketing a product. You were supposed to be operating in the public benefit. The public benefit would be served by ASF using its "marketing power" to promote things that are of high quality and state of the art in their space. And if, in a given space, they don't have anything remotely competitive to offer, they shouldn't. That would be operating in the public interest.

Really, if I were going to lobby ASF to do something, it would be to label honestly the various projects that they host that are effectively abandoned, and/or obsolete wrt the state of the art in their application space. My sense of things is that this could easily be 80% of the projects. But obviously, it would be utterly naive to think they would ever do that. So, given that, I'm not going to lobby them to do anything, I don't think.

What would really be in the public interest would be a foundation that does serious comparative reviews of different open source products. Maybe somebody could convince the big companies that this was something worth funding, but it would have to have the independence and integrity to really properly evaluate things objectively -- usability, code quality and so on. I don't know how you could ensure the objectivity and integrity of the thing. And to quote one of my collaborators, if something is shit, it would say it was shit, though using nicer words, of course.

Just a thought.

Well, thank you for that. I have responded (okay, maybe at too much length, but it's more time-consuming to write a shorter note) to the ideas you've thrown out there.

Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/

Velocity or FreeMarker: Looking at 5 Years of Practical Experience
http://freemarker.blogspot.com/2007/12/velocity-of-freemarker-looking-at-5.html




Gary


Jonathan Revusky wrote:
Matthijs Lambooy wrote:
Well Jonathan gives me another good reason to use velocity

How is that a reason to use Velocity? I know you're trying to tell me off (and suck up to the people you feel are the relevant authority figures) but it's unsettling that there isn't even the slightest attempt at logical (or pseudological) discourse.

Summary: A dipshit writes a response to my private message in public and I tell him off and another self-righteous ass says I should be banned for telling him off.

And that's a "good reason" for you to use Velocity.

I mean, there's no logical connection anywhere. It's like:

"The socialists were just reelected in Spain."
"That's a good reason to use Velocity."
"Damn right."

Why don't you just say "Yer mommy wears army boots!" or something. At least it's something of an insult (at the primary school level anyway...)

Or really, try to make some technical argument in defense of how great Velocity supposedly is, we could talk about that...


Thanks Jonathan!!

You're welcome, I guess.... <shrug>

Oh, here are a couple more good reasons for people to use Velocity. The price of crude oil is at a record high. Pamela Anderson still has big tits.

Jonathan Revusky


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to