You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@velocity.apache.org by Serge Knystautas <se...@lokitech.com> on 2003/06/25 21:24:18 UTC
DVSL status? (or got another XSL replacement?)
I've got a documentation system that's using Velocity for much of the
HTML output. A few of the HTML documents I want to output are coming
directly from an XML file, and so initially I'm using XSL, but I would
really like to use Velocity templating to expose some extra info via
Java objects.
Anakia looks somewhat abandoned, and DVSL seems to fit the bill pretty
well. But it looks somewhat abandoned as well with no commits in 9+
months. In fact the most recent commits were by Jason van Zyl to
"demonstrate memory leaks," not the most encouraging sign.
For using Velocity to transform XML documents, is DVSL the best thing
available, or are there other tools I should be considering? Thanks in
advance for any suggestions.
--
Serge Knystautas
President
Lokitech >> software . strategy . design >> http://www.lokitech.com
p. 301.656.5501
e. sergek@lokitech.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
RE: Alternate Language Engine
Posted by Gonzalo Diethelm <go...@diethelm.org>.
> > * Write a Velocity-compatible template engine in C, with
> > hooks for everything that could be environment dependent,
> > like how to access a context, how to call an object's
> > method, how to access an object's properties, etc.
>
> I like this because it makes it easier to hook into engine of the day,
> but the pain of writing a templating engine in C... It would be
> probably easier do separate Perl, Ruby, Python, and C# implementations
> than write one in ANSI C and write wrappers for each I think =/
But I wouldn't want to be the one maintaining the separate engines,
in several different languages... I have written a couple of compilers
in C before, I don't think it would be that hard.
The one thing I could never assess was how easy it would be to design
a generic interface for each language "skin". What would you need to
support in this interface?
* Access to a "context".
* The ability to introspect.
* The ability to call methods in objects placed in the context.
* Anything else?
That's about it. The rest of the code could be implemented ONCE in C.
> > One set of template pages, many possible "back-ends".
>
> That be the idea. Couple it with a Context Service (ie, extract the
> Context via JAXB and equivalents in other languages) and things get
> interesting. In theory you can define your services to be pretty
> transparent. As webapp contexts are pretty close to single-thread
> oriented most of the time you can cache them very aggressively and be
> okay - but this is worrying about performance way too early.
Yeah. It sounds interesting to me, no one ever agreed though...
> -Brian
Regards,
--
Gonzalo A. Diethelm
gonzalo.diethelm@aditiva.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
RE: Alternate Language Engine
Posted by Tim Colson <tc...@cisco.com>.
> but the pain of writing a templating engine in C... It would be
> probably easier do separate Perl, Ruby, Python, and C#
.Net/C# has already been done. :-)
http://nvelocity.sourceforge.net/
I wonder if there is an equivelant to JavaCC for the other languages.
-Timo
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: Alternate Language Engine
Posted by Brian McCallister <mc...@forthillcompany.com>.
On Friday, June 27, 2003, at 09:24 AM, Gonzalo Diethelm wrote:
> * Write a Velocity-compatible template engine in C, with
> hooks for everything that could be environment dependent,
> like how to access a context, how to call an object's
> method, how to access an object's properties, etc.
I like this because it makes it easier to hook into engine of the day,
but the pain of writing a templating engine in C... It would be
probably easier do separate Perl, Ruby, Python, and C# implementations
than write one in ANSI C and write wrappers for each I think =/
> One set of template pages,
> many possible "back-ends".
>
That be the idea. Couple it with a Context Service (ie, extract the
Context via JAXB and equivalents in other languages) and things get
interesting. In theory you can define your services to be pretty
transparent. As webapp contexts are pretty close to single-thread
oriented most of the time you can cache them very aggressively and be
okay - but this is worrying about performance way too early.
-Brian
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
RE: Alternate Language Engine
Posted by Gonzalo Diethelm <go...@diethelm.org>.
I once faced a similar project, where we needed some PHP-
and Perl-backed pages. I thought the simplicity and power
of Velocity were just perfect for my needs. At one point,
I charted the following plan:
* Write a Velocity-compatible template engine in C, with
hooks for everything that could be environment dependent,
like how to access a context, how to call an object's
method, how to access an object's properties, etc.
* Write "flavors" for this engine in each desired language.
The first one would be Java, as a proof of concept.
Alas, I never found the time to do this. I still believe
it could be a good way to spread Velocity to other languages
and/or environments. Imagine that! One set of template pages,
many possible "back-ends".
Regards,
--
Gonzalo A. Diethelm
gonzalo.diethelm@aditiva.com
> -----Original Message-----
> From: Anthony Eden [mailto:aeden@signaturedomains.com]
> Sent: Thursday, June 26, 2003 17:18
> To: Velocity Users List
> Subject: Re: Alternate Language Engine
>
>
> You might want to have a look at http://freshmeat.net/articles/view/447/
> for an article which discusses creating a basic template engine in
> Ruby. Should be pretty easy to build one from there, especially with
> the regex support.
>
> Sincerely,
> Anthony Eden
>
> Brian McCallister wrote:
>
> > Hello all,
> >
> > This may be a bit way off-scope (particularly as Jakarta is Java
> > centric) but I have a problem I would like to solve via Velocity which
> > I cannot quite yet.
> >
> > Basically I need to integrate look and feel of several web
> > applications that are spread across Java and Ruby. The Java side of
> > the application is planning to use Velocity for page generation, and
> > the Ruby side is... planning to use eruby (think ASP/PHP but with
> > Ruby) unless I can come up with something better. I am attempting to
> > do this so that we don't wind up maintaining twice the templates for
> > common things (top/bottom/left etc).
> >
> > I would dearly love to use a common set of templates between all of
> > the components (be they ruby, Java, or... whatever else winds up being
> > grafted on down the line).
> >
> > So, does anyone know of any effort to port a VTL engine to Ruby? As I
> > am pretty confident that no one has done this yet - is anyone familiar
> > with the inner workings of VTL interested in helping to build one?
> >
> > Thanks,
> > Brian
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: velocity-user-help@jakarta.apache.org
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: velocity-user-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: Alternate Language Engine
Posted by Anthony Eden <ae...@signaturedomains.com>.
You might want to have a look at http://freshmeat.net/articles/view/447/
for an article which discusses creating a basic template engine in
Ruby. Should be pretty easy to build one from there, especially with
the regex support.
Sincerely,
Anthony Eden
Brian McCallister wrote:
> Hello all,
>
> This may be a bit way off-scope (particularly as Jakarta is Java
> centric) but I have a problem I would like to solve via Velocity which
> I cannot quite yet.
>
> Basically I need to integrate look and feel of several web
> applications that are spread across Java and Ruby. The Java side of
> the application is planning to use Velocity for page generation, and
> the Ruby side is... planning to use eruby (think ASP/PHP but with
> Ruby) unless I can come up with something better. I am attempting to
> do this so that we don't wind up maintaining twice the templates for
> common things (top/bottom/left etc).
>
> I would dearly love to use a common set of templates between all of
> the components (be they ruby, Java, or... whatever else winds up being
> grafted on down the line).
>
> So, does anyone know of any effort to port a VTL engine to Ruby? As I
> am pretty confident that no one has done this yet - is anyone familiar
> with the inner workings of VTL interested in helping to build one?
>
> Thanks,
> Brian
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: velocity-user-help@jakarta.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Alternate Language Engine
Posted by Brian McCallister <mc...@forthillcompany.com>.
Hello all,
This may be a bit way off-scope (particularly as Jakarta is Java
centric) but I have a problem I would like to solve via Velocity which
I cannot quite yet.
Basically I need to integrate look and feel of several web applications
that are spread across Java and Ruby. The Java side of the application
is planning to use Velocity for page generation, and the Ruby side
is... planning to use eruby (think ASP/PHP but with Ruby) unless I can
come up with something better. I am attempting to do this so that we
don't wind up maintaining twice the templates for common things
(top/bottom/left etc).
I would dearly love to use a common set of templates between all of the
components (be they ruby, Java, or... whatever else winds up being
grafted on down the line).
So, does anyone know of any effort to port a VTL engine to Ruby? As I
am pretty confident that no one has done this yet - is anyone familiar
with the inner workings of VTL interested in helping to build one?
Thanks,
Brian
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
news: Velocity used in Intellij IDEA
Posted by Tim Colson <tc...@cisco.com>.
Howdy folks -
I just received some code generation templates for IntelliJ IDEA from
one of my colleagues and noticed that they use Velocity. I didn't see
this in the "Powered By" page, perhaps worth adding.
-Timo
>From the IntelliJ IDEA 3.x help system:
IntelliJ IDEA provides several built-in templates for creating new
files, classes and interfaces, as well as some supplementary templates
to be automatically used when generating particular blocks of code.
Along with using the built-in templates, you can easily create your own.
To view/edit the existing templates, or to define your own, select the
Options | File Templates menu item. The File Templates dialog will
appear, containing the following three tabs:
Templates - for file templates Includes - for code fragment
templates to be dynamically included into the file templates Code -
for code templates
...
When writing or editing templates, you can use built-in and custom
variables, conditional directives, etc. which are briefly described in
the Description field of the File Template dialog. For more details on
template syntax, please, refer to
http://jakarta.apache.org/velocity/user-guide.html.
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
RE: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Bill Chmura <Bi...@Explosivo.com>.
This is still going?
I'm starting to see infomercials on Freemarker in my sleep, or did I
actually see it?
-----Original Message-----
From: Dave Newton [mailto:dave@solaraccess.com]
Sent: Thursday, June 26, 2003 9:51 AM
To: Velocity Users List
Subject: Re: [OT]Re: DVSL status? (or got another XSL replacement?)
(Another funny aside is that when I agreed with Jon a few months ago he
was more than happy to take it off-line and send me a personal email,
but for some reason he needs to display his debate prowess when it's
somebody that _doesn't_ agree. Whatever. Everybody has their own
personal needs, I suppose. Jonathon, I'm _more_ than happy to continue
this drivel off-line if you can stand my incoherence, juggling of topics
I know nothing about, and seething hostility. I'd be even MORE excited
to continue the conversation in person, and my July vacation is a
PERFECT time for that. Lemme know.)
On Thu, 2003-06-26 at 02:47, Jonathan Revusky wrote:
> I just reread my various messages, and I don't know what you're
> talking
> about. I've been quite calm throughout -- even despite Dave Newton's
> seething incoherent hostility towards me. I have even answered Dave
> pretty calmly.
What I've been doing is "seething hostility?"
Dude, you have NO idea what hostility is, and you're definitely testing
my Buddha nature. Seething hostility. Indeed.
Grow up and recognize there are a LOT more important things to be
fretting about other than whether or not somebody's using FreeMarker and
their reasons for doing so.
> As for your statement that I was preaching to the wrong choir,
> etcetera,
> you were *clearly* suggesting that you would use Velocity simply
because
> it was an apache project -- independently of technical considerations.
I
> could not and still cannot parse it differently.
HOLY CRAP, HE DIDN'T! I EVEN QUOTED HIS EXACT WORDS!
> What is funny about this is that in a recent exchange, one Henning P.
> Schmiedehausen, one of the Turbine developers, ridiculed Daniel Dekany
> for suggesting that Jakarta/ASF projects and Jakarta/ASF developers
have
> an extreme bias towards using other Jakarta/ASF projects. This was
> supposedly paranoid conspiracy theory stuff. <shrug>
Nobody disagrees with you that bias towards a given project might not be
for solely technical reasons. For instance of me using FreeMarker at
this point are slim, due _solely_ to your attitude. Not technical at
all. And I feel pretty damn good about it.
Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Dave Newton <da...@solaraccess.com>.
(Another funny aside is that when I agreed with Jon a few months ago he
was more than happy to take it off-line and send me a personal email,
but for some reason he needs to display his debate prowess when it's
somebody that _doesn't_ agree. Whatever. Everybody has their own
personal needs, I suppose. Jonathon, I'm _more_ than happy to continue
this drivel off-line if you can stand my incoherence, juggling of topics
I know nothing about, and seething hostility. I'd be even MORE excited
to continue the conversation in person, and my July vacation is a
PERFECT time for that. Lemme know.)
On Thu, 2003-06-26 at 02:47, Jonathan Revusky wrote:
> I just reread my various messages, and I don't know what you're talking
> about. I've been quite calm throughout -- even despite Dave Newton's
> seething incoherent hostility towards me. I have even answered Dave
> pretty calmly.
What I've been doing is "seething hostility?"
Dude, you have NO idea what hostility is, and you're definitely testing
my Buddha nature. Seething hostility. Indeed.
Grow up and recognize there are a LOT more important things to be
fretting about other than whether or not somebody's using FreeMarker and
their reasons for doing so.
> As for your statement that I was preaching to the wrong choir, etcetera,
> you were *clearly* suggesting that you would use Velocity simply because
> it was an apache project -- independently of technical considerations. I
> could not and still cannot parse it differently.
HOLY CRAP, HE DIDN'T! I EVEN QUOTED HIS EXACT WORDS!
> What is funny about this is that in a recent exchange, one Henning P.
> Schmiedehausen, one of the Turbine developers, ridiculed Daniel Dekany
> for suggesting that Jakarta/ASF projects and Jakarta/ASF developers have
> an extreme bias towards using other Jakarta/ASF projects. This was
> supposedly paranoid conspiracy theory stuff. <shrug>
Nobody disagrees with you that bias towards a given project might not be
for solely technical reasons. For instance of me using FreeMarker at
this point are slim, due _solely_ to your attitude. Not technical at
all. And I feel pretty damn good about it.
Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Serge Knystautas wrote:
> Holy crap guys...
>
> I've tried to think of something funny to say about this thread, but
> just too flabergasted. John, I'm sorry if I've touched a red button.
> Please reread my email as I'm sure once my comment is in context again,
> it will [hopefully] calm you.
>
I just reread my various messages, and I don't know what you're talking
about. I've been quite calm throughout -- even despite Dave Newton's
seething incoherent hostility towards me. I have even answered Dave
pretty calmly.
As for your statement that I was preaching to the wrong choir, etcetera,
you were *clearly* suggesting that you would use Velocity simply because
it was an apache project -- independently of technical considerations. I
could not and still cannot parse it differently.
What is funny about this is that in a recent exchange, one Henning P.
Schmiedehausen, one of the Turbine developers, ridiculed Daniel Dekany
for suggesting that Jakarta/ASF projects and Jakarta/ASF developers have
an extreme bias towards using other Jakarta/ASF projects. This was
supposedly paranoid conspiracy theory stuff. <shrug>
Best Regards,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
FreeMarker-Velocity comparison page, http://freemarker.org/fmVsVel.html
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Serge Knystautas <se...@lokitech.com>.
Holy crap guys...
I've tried to think of something funny to say about this thread, but
just too flabergasted. John, I'm sorry if I've touched a red button.
Please reread my email as I'm sure once my comment is in context again,
it will [hopefully] calm you.
--
Serge Knystautas
President
Lokitech >>> software . strategy . design >> http://www.lokitech.com
p. 301.656.5501
e. sergek@lokitech.com
>
> Serge can use whatever software he wants. That much is clear.
>
>
> Serge specifically said he was using it because it was an apache
> project. That is quite clear and there's no getting around that.
>
>
> I'll let you in on a little secret, Dave. I actually knew perfectly well
> that Serge was never going to use FreeMarker and I knew why already.
>
> I was just quite surprised that he openly said why. :-)
>
>
> It is true that I don't like somebody saying that he is using Velocity
> simply because it's an Apache project. However, I have long stopped
> being shocked by this. Though I was surprised that he openly stated it.
>
>
> Well, you seem to think that you can speak for Serge. Now you seem to
> think that you can speak in the name of the collective. I don't
> understand why.
>
>
> Serge Knystautas was making noises along these lines -- though he didn't
> express it in terms as crass.
>>
>> As you state, of course, I don't speak for Serge.
>
>
> You were trying to, but that was not legitimate discourse.
>
>
> Yes, but that still doesn't explain why you think it appropriate for you
> take it upon yourself to answer for Serge.
>
>>
>> I never claimed it was a _technically_ valid reason, I just claimed it's
>> a _valid_ reason. For Serge.
>
>
> He was asking about various XML-transformation stuff. There is anakia
> and dvsl that are both Velocity-based. As far as I could tell, Serge has
> no existing systems based on those things.
>
>
> Suppose Serge starts using FreeMarker tomorrow. What do I gain?
>
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Nathan Bubna wrote:
> Jonathan Revusky said:
> ...
>
>>I have also contrasted Velocity unfavorably to FreeMarker in various
>>respects --
>
> ...
>
>>lack of any notion of namespaces or local
>>variable scope in macros.
>
>
> you know i think you've said this before, and at the time, i thought, "huh? i
> don't think that's true," but didn't have time to follow up on it. today i
> did. i don't use it and don't often here it requested as a feature, but it
> turns out my vague recollection was right. the Developer's guide mentions
> this:
>
> <quote>
> velocimacro.context.localscope = false
> Controls whether reference access (set/get) within a Velocimacro will change
> the context, or be of local scope in that Velocimacro.
> </quote>
I was unaware of that, I have to admit.
OTOH, this feature, as described, makes me wonder what the people
involved were smoking. In fact, it is such a screwball setup that I am
surprised you want to publicize it! :-)
I mean, the whole idea that a configuration switch is what determines
whether changes to a variable in a macro persist outside the macro
invocation!!!??? And the switch then applies to all variables in all
macros!!!??? In *all* templates???!!!
Is that right????!!! It can't be set on anything but a global
configuration level??? Nathan, do you think that, as currently set up,
this is a useful feature?
I mean, shouldn't the person writing the macro be able to declare some
variables as local and others as global as needed? You know, even if I
concede that Velocity does have some notion of local namespaces, it does
not appear that it is set up in a particularly useful way... it might as
well not exist frankly.
The other namespace thing I was referring to is the notion of being able
to import macro libraries in different namespaces. And Velocity has no
notion of this. See:
http://freemarker.org/docs/dgui_misc_namespace.html
Basically, in FreeMarker, you can go:
<#import "myMacros.ftl" as myMacros>
<#import "yourMacros.ftl" as yourMacros>
and then myMacros can have a macro called displayFoo and yourMacros can
have a macro called displayFoo as well, but they don't clash.
<@myMacros.foo/>
<@yourMacros.foo/>
This is the namespace feature I was referring to in the discussion of
the XML processing stuff. It's really a key feature in terms of being
able to do powerful XML transformations. It allows you to do
customization layers and so on, via a sequence of namespaces. So, you
can go:
<#visit document using [myMacros, yourMacros]>
and then if myMacros does not contain a macro for handling a given node,
it falls back on looking in yourMacros.
The lack of this kind of namespaces functionality (though not just that)
is what prevents Velocity (in its current state) from being the basis
of a really serious XML-transformation tool.
So, what I was saying was broadly true, I think. And if I made a
mistake, it was in good faith.
>
> ...
>
>>However, all of my statements have been factual. This should be obvious.
>
>
> if you say so.
Well, if something is not factual, do point it out. I am satisfied that
my statement Velocity's lack of namespaces was basically true.
>
>
>>Given the hostility towards me, if I had said anything that anybody
>>could challenge on a factual basis, they would have done so.
>
>
> actually, i rarely read your emails.
Nathan, I am suspicious of your claim that you don't read my emails. I
mean, it could be true, but then how come you read this last one? Okay,
you rarely read my emails. What caused you to read this last one and not
read other ones?
I will never know for sure, but if I had to bet money, I would bet that
you read all my emails and that in the above, you are dissembling. Just
a feeling I get....
> a quick scan is about all they typically
> get from me. they're too long most of the time.
You're right about that. My messages tend to be longer than they should
be. It's because it's more time-consuming to write shorter messages, so
I dash off longer ones.
> the noise/signal ratio just
> doesn't make it worth the effort. i have better things to do. also, there
> are other developers and users that have just kill-filed you since at least a
> third of the time your emails are propaganda (however factual they might be
> doesn't change that) and about half of the time they just turn into big, fat,
> waste-of-time flame wars like this.
Ah yes, these flame-wars that I "single-handedly" perpetuate... ;-)
You know, really, I should tell you guys something. If somebody shows up
on your mailing list and points out that a competing product has
features that your product lacks, here is how you respond:
"Thank you, Mr. X. for bringing this to our attention. We are always
very interested in information that can help us make Y better."
It's a very bad mistake IMO to flame somebody in that spot. You lose
face. It gives the impression that you are afraid that people will
compare your product with the competitor and find it lacking. (You don't
want to give that impression -- even if it's true. ;-))
Ciao,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
FreeMarker-Velocity comparison page, http://freemarker.org/fmVsVel.html
>
> Nathan Bubna
> nathan@esha.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Nathan Bubna <na...@esha.com>.
Jonathan Revusky said:
...
> I have also contrasted Velocity unfavorably to FreeMarker in various
> respects --
...
> lack of any notion of namespaces or local
> variable scope in macros.
you know i think you've said this before, and at the time, i thought, "huh? i
don't think that's true," but didn't have time to follow up on it. today i
did. i don't use it and don't often here it requested as a feature, but it
turns out my vague recollection was right. the Developer's guide mentions
this:
<quote>
velocimacro.context.localscope = false
Controls whether reference access (set/get) within a Velocimacro will change
the context, or be of local scope in that Velocimacro.
</quote>
...
> However, all of my statements have been factual. This should be obvious.
if you say so.
> Given the hostility towards me, if I had said anything that anybody
> could challenge on a factual basis, they would have done so.
actually, i rarely read your emails. a quick scan is about all they typically
get from me. they're too long most of the time. the noise/signal ratio just
doesn't make it worth the effort. i have better things to do. also, there
are other developers and users that have just kill-filed you since at least a
third of the time your emails are propaganda (however factual they might be
doesn't change that) and about half of the time they just turn into big, fat,
waste-of-time flame wars like this.
Nathan Bubna
nathan@esha.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by "Henning P. Schmiedehausen" <hp...@intermeta.de>.
Daniel Dekany <dd...@freemail.hu> writes:
>Note that Jonathan is a heavily sporting man... do not be rash.
>Hey, really... we should arrange a FM devels VS Velocity devels
>full-contact meeting... the resulting MPEG would be a top download for
>sure.
http://www.brainyquote.com/quotes/quotes/i/q133913.html
Regards
Henning
--
Dipl.-Inf. (Univ.) Henning P. Schmiedehausen INTERMETA GmbH
hps@intermeta.de +49 9131 50 654 0 http://www.intermeta.de/
Java, perl, Solaris, Linux, xSP Consulting, Web Services
freelance consultant -- Jakarta Turbine Development -- hero for hire
--- Quote of the week: "It is pointless to tell people anything when
you know that they won't process the message." --- Jonathan Revusky
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Daniel Dekany <dd...@freemail.hu>.
Saturday, June 28, 2003, 4:34:48 PM, Dave Newton wrote:
> On Sat, 2003-06-28 at 09:59, Dave Newton wrote:
>> It's just so sad that you only see your side.
>
> Let me clarify a bit.
Great... one would believe that thread ends, Jonathan will not replay,
and then you stat to talk with yourself. :)
> I believe that most people here agree with your technical points. Hard
> to argue them, and not much fun besides. You really do make
> communication with you much more difficult than it needs to be, and
> that's a bummer. World's too small to be like that.
>
> The problem isn't the message, it's the location, mode, and quantity of
> delivery. You may believe you're doing everybody a favor by stirring up
> the Velocity camp--but what's actually happened?
>
> As a last ditch effort:
>
> Serge said:
>> Honestly, you're preaching to the choir of a different church... I've
>> been an apache committer for 4+ years and am unlikely to adopt a
>> non-apache tool when there is a very well known, if comfortably
>> mature, project within Apache already.
>
> Actual facts (Serge, feel free to annotate if I'm wrong):
> 1) Serge has been a committer for 4+ years
> 2) Serge will use an existing Apache tool to solve a problem if it fits
> his needs
3) The fist name of Serge is Serge
> Probable facts:
> 1) Serge is a pretty smart guy
Excuse me... what does it to be certain? ;)
> 2) Serge won't use something it doesn't get the job done
5) Serge doesn't stick his finders into the 220/260V wall socket.
> 3) Velocity gets (certain) jobs done acceptably
> 4) Serge will use a different tool when necessary/desirable
(Incredible...)
BTW, bonus probable fact:
6) ASF and VSS are religious foundations (they has church)
> Things Jonathon added:
Or,
Probable facts + some extremists you have added:
> 1) Serge won't use anything but ASF code
> 2) Serge is using Velocity ___only___ because it's ASF
>
> Now, somebody else already pointed this out, but you still seem to think
> that those last things you added on are a) true and b) supported by
> previous posts. I don't think too many other people see it that way,
> including Serge, who (again, as you correctly point out) is the only
> person that can officially speak to this.
>
> Serge, however, is a much nicer, much more patient human than I am.
>
> I guess my bottom line is this: All through this BS I have addressed
> every point of yours I find valid (and there are many) and been very
> forthright when I agree with what you say and when I don't.
>
> In your posts I see __NO__ give-and-take. You skip over perfectly valid
> points/observations to get to something you can find an issue with, no
> matter how insignificant or unrelated. You honestly seem to believe that
> everybody here is against you, disagrees with everything you say, and
> want nothing to do with you.
I'm pretty sure that Jonathan knows that there are people who wants to
do *something* with him... :)
<note>
Actually, we have found an IRC log where two guys talks about that they
would beat the shit out of him... Honestly. :)))
</note>
> What _I_ want is for you to be constructive and far less verbose and
> maybe, just maybe, if possible, perhaps on occasion, take the time to
> recognize other people's opinions as being valid for them, and try to
> stop coming across as a sour-grapes jackass developer.
>
> We KNOW you're a good developer--and we'd rather deal with you on _that_
> level rather than all this other stuff. I'd even hazard a guess that
> most people here have looked at FreeMarker and think it's a pretty solid
> product. I do, anyway.
>
> But...
[snip]
BTW, are you and Jonathan stenographer champions? If I had to type this
much, I wouldn't have time to sleep for sure...
I wonder that I will create:
freemarker-vs-velocity-iq-wrestling@sourceforge.net
or something like that...
> WE GET IT. VELOCITY DEVELOPMENT IS STALE. GET OVER IT. FREEMARKER IS
> BETTER (in some ways, anyway). SOME OF US ARE USING VELOCITY ANYWAY FOR
> REASONS YOU MIGHT NOT LIKE OR AGREE WITH. GET OVER IT. IT'S OUR CHOICE,
> AND IT'S ALL GOING TO BE OKAY!
>
> Seriously--I'm on vacation all July.
(It was time to... :))
> I'd be happy to continue this off-line over a beer.
Note that Jonathan is a heavily sporting man... do not be rash.
Hey, really... we should arrange a FM devels VS Velocity devels
full-contact meeting... the resulting MPEG would be a top download for
sure.
> I've found an awful lot of stupid internet behavior disappears once
> the internet part is taken away and real life rules take over.
[snip]
BTW, I should do something instead of reading this and writing
bullshit... :) OK...
--
Best regards,
Daniel Dekany
a.k.a.
Mel Gibson
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Dave Newton <da...@solaraccess.com>.
On Sat, 2003-06-28 at 09:59, Dave Newton wrote:
> It's just so sad that you only see your side.
Let me clarify a bit.
I believe that most people here agree with your technical points. Hard
to argue them, and not much fun besides. You really do make
communication with you much more difficult than it needs to be, and
that's a bummer. World's too small to be like that.
The problem isn't the message, it's the location, mode, and quantity of
delivery. You may believe you're doing everybody a favor by stirring up
the Velocity camp--but what's actually happened?
As a last ditch effort:
Serge said:
> Honestly, you're preaching to the choir of a different church... I've
> been an apache committer for 4+ years and am unlikely to adopt a
> non-apache tool when there is a very well known, if comfortably
> mature, project within Apache already.
Actual facts (Serge, feel free to annotate if I'm wrong):
1) Serge has been a committer for 4+ years
2) Serge will use an existing Apache tool to solve a problem if it fits
his needs
Probable facts:
1) Serge is a pretty smart guy
2) Serge won't use something it doesn't get the job done
3) Velocity gets (certain) jobs done acceptably
4) Serge will use a different tool when necessary/desirable
Things Jonathon added:
1) Serge won't use anything but ASF code
2) Serge is using Velocity ___only___ because it's ASF
Now, somebody else already pointed this out, but you still seem to think
that those last things you added on are a) true and b) supported by
previous posts. I don't think too many other people see it that way,
including Serge, who (again, as you correctly point out) is the only
person that can officially speak to this.
Serge, however, is a much nicer, much more patient human than I am.
I guess my bottom line is this: All through this BS I have addressed
every point of yours I find valid (and there are many) and been very
forthright when I agree with what you say and when I don't.
In your posts I see __NO__ give-and-take. You skip over perfectly valid
points/observations to get to something you can find an issue with, no
matter how insignificant or unrelated. You honestly seem to believe that
everybody here is against you, disagrees with everything you say, and
want nothing to do with you.
What _I_ want is for you to be constructive and far less verbose and
maybe, just maybe, if possible, perhaps on occasion, take the time to
recognize other people's opinions as being valid for them, and try to
stop coming across as a sour-grapes jackass developer.
We KNOW you're a good developer--and we'd rather deal with you on _that_
level rather than all this other stuff. I'd even hazard a guess that
most people here have looked at FreeMarker and think it's a pretty solid
product. I do, anyway.
But...
WE GET IT. VELOCITY DEVELOPMENT IS STALE. GET OVER IT. FREEMARKER IS
BETTER (in some ways, anyway). SOME OF US ARE USING VELOCITY ANYWAY FOR
REASONS YOU MIGHT NOT LIKE OR AGREE WITH. GET OVER IT. IT'S OUR CHOICE,
AND IT'S ALL GOING TO BE OKAY!
Seriously--I'm on vacation all July. I'd be happy to continue this
off-line over a beer. I've found an awful lot of stupid internet
behavior disappears once the internet part is taken away and real life
rules take over.
Maybe you really do believe this is all me and I'm just a guy struggling
with concepts I don't understand, I dunno. That'd be pretty wild and a
pretty big leap-of-faith on your part, but okay. I don't think you will
find much lacking in my intelligence department except the common sense
to shut my mouth after spending a week or so going around in circles.
But apparently I'm not the only stupid one when it comes to that, hey?
So let's take it off-line like you did last time when I agreed with you
back a couple weeks ago. Where you said something I now find somewhat
interesting... hmm:
"Well, it's pretty obvious that I am not terribly well-intentioned
towards the Velocity people. In fact, there is definitely a certain
antipathy."
I don't want to see FreeMarker being adopted just to shut you up
(although if anybody want to publicly do so, please, PLEASE do so he
feels better! :)
You're a smart guy, good developer, irritating instigator. Let's hack
the last one off that list and move on.
Smooches,
Dave
(And yes, "you are irritating" isn't a valid debating topic. Nothing to
debate. If life were one big debate I'm sure you'd do swimmingly. But it
ain't, and there's more to it than picking apart only the items you take
issue with.)
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Dave Newton <da...@solaraccess.com>.
On Sat, 2003-06-28 at 08:18, Jonathan Revusky wrote:
> Well, when somebody's big debating point is "you're a really irritating
> guy" the guy has given up; he's conceded the debate.
Given up, yes. Nothing to concede, though.
It's just so sad that you only see your side.
Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Dirk Hoffmann <dh...@web.de>.
Jonathan Revusky wrote:
> [...]
> I think some of these mailing lists would be a gold mine for people
> who like to deconstruct deceptive uses of language...
>
Yeah, I like your flame wars for they help me to improve my English :-)
but nothing more :-(
Regards
Dirk
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Daniel Dekany wrote:
> Saturday, June 28, 2003, 11:15:59 AM, Jonathan Revusky wrote:
>
>
>>Dave Newton wrote:
>
> [snip]
>
>>>I think it's because I can quote emails where people are annoyed with
>>>your attitude, I guess.
>>
>><sigh>
>>
>>I was not born yesterday and I am quite aware that some people are very
>>pissed off at me and I know why.
>
> [snip]
>
> Sure, but it seems that you know only a subset of the reasons...
I dunno. There could be a reason or two that I am not aware of. I kind
of doubt it. My sense of things is that I know all the reasons that some
of the people on this list are infuriated at me.
It's a funny thing, though. There is a group here who really wants to
get the project back on track, and my blunt posts about the sad state of
Velocity development have surely been quite helpful to their cause. I
have mercilessly taken apart certain (invalid) arguments that people
come up. I also expressed openly something that those people probably
also believe but would never say: all this stuff about MVC
purism/minimalism/elegance is mostly a fancy intellectual
rationalization for the fact that no development has been taking place.
I would say that, if there is a new Velocity release with decimal number
support fairly soon, the people who want that feature (a large number of
people probably) ought to write me a thank-you note! :-)
>
> BTW, what about working on FM? :)
Well, it's true that I haven't committed any code for a while -- like
since 10 pm last night.... :-)
> Also we have some interesting pretty
> OT threads on the FM devel, such as Java memory management, or even
> house moving, just in the case you have been fed to the gills with FM
> development. :) OK, OK, I know you just can't stop debating,
Well, to be fair, I'm not debating alone here. Okay, it's true that some
of the other people involved are such weak debaters that it might seem
that way. But if you look closely, you'll see that there is always at
least one intervening post from someone else sandwiched between any two
of mine.
Once, on some such forum, someone accused me of "single-handedly
perpetuating a flame-war". This struck me as an amazing accusation. How
on earth does one "single-handedly" perpetuate a flame-war? Surely you
need at least one other person to be responding tit-for-tat. People's
choice of words is interesting sometimes. Recently, I have noticed that
people refer to posts of mine that bother them as "rants" or
"diatribes", i.e. monologues. Either consciously or not, they choose
terminology that implies that nobody else was involved in these threads,...
I think some of these mailing lists would be a gold mine for people who
like to deconstruct deceptive uses of language...
> but,
> believe me, there is a point when it is better to give it up.
Well, when somebody's big debating point is "you're a really irritating
guy" the guy has given up; he's conceded the debate.
So, it's not like there's really an ongoing debate here per se....
Regards,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
New XML transformation capabilities: http://freemarker.org/docs/xgui.html
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Daniel Dekany <dd...@freemail.hu>.
Saturday, June 28, 2003, 11:15:59 AM, Jonathan Revusky wrote:
> Dave Newton wrote:
[snip]
>> I think it's because I can quote emails where people are annoyed with
>> your attitude, I guess.
>
> <sigh>
>
> I was not born yesterday and I am quite aware that some people are very
> pissed off at me and I know why.
[snip]
Sure, but it seems that you know only a subset of the reasons...
BTW, what about working on FM? :) Also we have some interesting pretty
OT threads on the FM devel, such as Java memory management, or even
house moving, just in the case you have been fed to the gills with FM
development. :) OK, OK, I know you just can't stop debating, but,
believe me, there is a point when it is better to give it up.
--
Best regards,
Daniel Dekany
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Dave Newton wrote:
> On Wed, 2003-06-25 at 22:05, Jonathan Revusky wrote:
>
>>>Okay, this is my last response, for which I'm sure we're all grateful.
>
>
> Arg, I suck.
No comment.
>
>
>>I'll let you in on a little secret, Dave. I actually knew perfectly well
>>that Serge was never going to use FreeMarker and I knew why already.
>
>
> So this isn't even a _necessary_ conversation, let alone _relevant_.
>
> Nice.
>
>
>>Well, you may not think you're being hostile, but I definitely perceive
>>you as hostile.
>
>
> Okey :)
Yeah, well, you may not know the effect you are producing on others, but
I do put it to you that, if someone addressed you in the tone you've
been using with me over the last few messages.... you would perceive
that person as being pretty darned hostile.
I think that's safe to say.
>
>
>>>I'm not sure if you just don't see it or are ignoring it or what, but
>>>it's _your_ attitude here that is putting people off _and_ putting
>>>people off FreeMarker (as was mentioned at least once in this thread and
>>>in a couple off-line threads).
>>
>>Well, you seem to think that you can speak for Serge. Now you seem to
>>think that you can speak in the name of the collective. I don't
>>understand why.
>
>
> I think it's because I can quote emails where people are annoyed with
> your attitude, I guess.
<sigh>
I was not born yesterday and I am quite aware that some people are very
pissed off at me and I know why.
However, I will point out something about this. Over the last few weeks,
I have made off-and-on various comments about the sad state of Velocity
development: the fact that there have been no CVS commits to speak of
for a year, no roadmap, no bugfixes, no improvements to the docs... no
nothing.
I have also contrasted Velocity unfavorably to FreeMarker in various
respects -- lack of decimal number support, lack of control over
superfluous whitespace, lack of any notion of namespaces or local
variable scope in macros. And many other things.
People don't like this.
However, all of my statements have been factual. This should be obvious.
Given the hostility towards me, if I had said anything that anybody
could challenge on a factual basis, they would have done so.
But they can't, so they are reduced to this kind of unseemly whining
about how "irritating" I am and so on, and also "threatening" not to use
FreeMarker because I'm such a bad guy and stuff like that.
>
>
>>In short, given your marketplace paradigm, it's not clear what your
>>objection to my behavior is.
>
>
> You're irritating?
But of course! I also have terrible table manners, and am as ugly as sin.
But those aren't legitimate debating points either, you see....
>
>
>>You're the one who alluded to a marketplace. Competition is in the
>>nature of the marketplace. It's necessarily competitive, unless you have
>>a monopoly, but that's surely not what you meant when you suggested
>>letting the marketplace decide.
>
>
> In your world-view perhaps the marketplace is necessarily competitive,
> but not in mine. It _can_ be and often _is_, but it doesn't HAVE to be
Well, this is an interesting theory, but I think that most people think
of the marketplace as a highly competitive situation. And that surely
goes even more strongly for a marketplace in computer technology. Very
competitive, very fast-moving. That's just how it is.
> (_Natural Capitalism_ talks about ethical capitalism and touches on some
> of these issues.)
Is there some theory of "ethical" capitalism whereby the system is not
highly competitive in nature?
(And if so, is it remotely relevant to what we're talking about
here???!!! :-´))
>
>
>>I mean, there's a lot of very incoherent discourse emanating from you,
>>where you don't seem to understand the concepts that you're juggling with.
>
>
> Ah. Gotta hate that, huh?
Especially when people point it out...
>
> (I'm struggling to recall a time when I was _ever_ incoherent, but
> having problems. You ain't the only studly IQ boy out there, yanno, and
> before making claims about my knowledge I'd rethink. At least I'm on
> solid ground thinking you're annoying.)
>
>
>>>I don't know anybody that thinks along the "if it ain't ASF it ain't
>>>shit" lines that you bring up.
>>
>>Serge Knystautas was making noises along these lines -- though he didn't
>>express it in terms as crass.
>
>
> He never said ANYTHING like that. He said something more along the lines
> of (along with the standard "but I'm already using it" which you
> consistently ignore) "it's not ASF, I'm less likely to use it."
He did say precisely that.
>
> Big difference.
Look, I really think you should let Serge (and anybody else) speak for
himself.
>
>
>>>Hey, if I'm a long-standing committer to
>>>a code base you're darn _right_ I'm going to try to stay within that
>>>code base as long as it's practical. I just don't think Serge thinks
>>>that it's important enough of an issue to switch.
>>
>>The above is not a factual representation. Serge is no more a committer
>>to the Velocity codebase than he is to the FreeMarker codebase. He is
>>simply opting for Velocity because it's part of ASF.
>
>
> Yeah. The ASF codebase. I know. You just seem to have a problem because
> you don't like his reasons.
There is no "ASF codebase". It's a fictitious construct.
Okay, I know that there is a desire for people to think in those terms,
like for them to make a mental association between the Apache web server
(written in C) and all this Jakarta stuff (written in Java.) But they
have pretty much nothing to do with one another. In fact, Jakarta
projects largely have nothing to do with one another. Velocity is not
any more part of the same unified codebase with Tomcat than it is, say,
with Jetty (a non-apache Tomcat competitor.)
There is no technical basis behind the idea of an "ASF codebase". It's a
marketing concept at best. Thus, if somebody refers to that, they are
referencing non-technical reasons.
>
>
>>>As you state, of course, I don't speak for Serge.
>>
>>You were trying to, but that was not legitimate discourse.
>
>
> *yawn*
>
>
>>Yes, but that still doesn't explain why you think it appropriate for you
>>take it upon yourself to answer for Serge.
>
>
> What on earth makes you think I would answer for anybody except me?!
My observation that you took it on yourself to answer for Serge.
>
> Do you really spend the majority of your time arguing semantics with
> people? Is it fulfilling?
Dave, I was not talking to you in this dialogue. I was addressing Serge
Knystautas. You jumped into a discussion with me, not vice versa.
<snip>
>>If somebody deliberately goes to a given shop where things are more
>>expensive and of lower quality and does not consider the range of
>>options available, other places where one can buy, who does he damage
>>really?
>
>
> Himself?
>
> Which makes it his business?
Well, in the general case, somebody could be acting as an agent for
others. If I am charged with buying computers for the local school
district, say, I am under some kind of ethical obligation to spend the
money responsibly. If I go to one store where everything is more
expensive and don't shop around.... it clearly becomes the taxpayer's
business, i.e. a question of general public interest.
whether it's anybody else's business, in the general case, depends on
the exact situation.
To knowingly opt for an inferior technology really is a bit like
knowingnly paying too much for something. If I invested in a technical
venture and the CTO bought some computer equipment, and goes to a store
owned by his buddies where everything is more expensive.....
now, suppose he opts for an inferior because it came from ASF (his
buddies) in the full knowledge that a competing tool was superior and
more actively developed and supported, would I not potentially have
grounds for being pissed off in both cases? And for broadly the same
reasons?
>
>
>>Basically just himself, right?
>
>
>>Suppose Serge starts using FreeMarker tomorrow. What do I gain?
>
>
> Nothing?
>
> Which makes this a curious exercise?
I was exploring the various behavior and thought patterns involved here.
I am discussing this topic calmly and in good faith. That people get
this bothered by what I'm saying, the implications of it.... well,
that's really kind of beyond my control, isn't it?
Regards,
Jonthan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
FreeMarker-Velocity comparison page, http://freemarker.org/fmVsVel.html
>
> But somehow I'm bettin' in your head y'all one this one, hey?
>
> *hee*
>
>
>>Nothing, right?
>>On that thought....
>
>
> Woohooo!
>
> Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Dave Newton <da...@solaraccess.com>.
On Wed, 2003-06-25 at 22:05, Jonathan Revusky wrote:
> > Okay, this is my last response, for which I'm sure we're all grateful.
Arg, I suck.
> I'll let you in on a little secret, Dave. I actually knew perfectly well
> that Serge was never going to use FreeMarker and I knew why already.
So this isn't even a _necessary_ conversation, let alone _relevant_.
Nice.
> Well, you may not think you're being hostile, but I definitely perceive
> you as hostile.
Okey :)
> > I'm not sure if you just don't see it or are ignoring it or what, but
> > it's _your_ attitude here that is putting people off _and_ putting
> > people off FreeMarker (as was mentioned at least once in this thread and
> > in a couple off-line threads).
> Well, you seem to think that you can speak for Serge. Now you seem to
> think that you can speak in the name of the collective. I don't
> understand why.
I think it's because I can quote emails where people are annoyed with
your attitude, I guess.
> In short, given your marketplace paradigm, it's not clear what your
> objection to my behavior is.
You're irritating?
> You're the one who alluded to a marketplace. Competition is in the
> nature of the marketplace. It's necessarily competitive, unless you have
> a monopoly, but that's surely not what you meant when you suggested
> letting the marketplace decide.
In your world-view perhaps the marketplace is necessarily competitive,
but not in mine. It _can_ be and often _is_, but it doesn't HAVE to be
(_Natural Capitalism_ talks about ethical capitalism and touches on some
of these issues.)
> I mean, there's a lot of very incoherent discourse emanating from you,
> where you don't seem to understand the concepts that you're juggling with.
Ah. Gotta hate that, huh?
(I'm struggling to recall a time when I was _ever_ incoherent, but
having problems. You ain't the only studly IQ boy out there, yanno, and
before making claims about my knowledge I'd rethink. At least I'm on
solid ground thinking you're annoying.)
> > I don't know anybody that thinks along the "if it ain't ASF it ain't
> > shit" lines that you bring up.
> Serge Knystautas was making noises along these lines -- though he didn't
> express it in terms as crass.
He never said ANYTHING like that. He said something more along the lines
of (along with the standard "but I'm already using it" which you
consistently ignore) "it's not ASF, I'm less likely to use it."
Big difference.
> > Hey, if I'm a long-standing committer to
> > a code base you're darn _right_ I'm going to try to stay within that
> > code base as long as it's practical. I just don't think Serge thinks
> > that it's important enough of an issue to switch.
> The above is not a factual representation. Serge is no more a committer
> to the Velocity codebase than he is to the FreeMarker codebase. He is
> simply opting for Velocity because it's part of ASF.
Yeah. The ASF codebase. I know. You just seem to have a problem because
you don't like his reasons.
> > As you state, of course, I don't speak for Serge.
> You were trying to, but that was not legitimate discourse.
*yawn*
> Yes, but that still doesn't explain why you think it appropriate for you
> take it upon yourself to answer for Serge.
What on earth makes you think I would answer for anybody except me?!
Do you really spend the majority of your time arguing semantics with
people? Is it fulfilling?
> > Secondly, I didn't really answer any questions, I just pointed out some
> > ideas in your post I didn't care for, that's all. No biggie.
See, you do what you accuse other people of all the time--simply
skipping over chunks that just make too much sense for you to pick
apart. Gotta hate that too, hey?
> Well, to use knowingly an inferior tool because it comes from ASF is
> suspect by various common-sense criteria.
Various but not all.
Give it a rest!
> > (Just as a minor aside, there are some Microsoft products I consider
> > superior to some products I use now. But I don't use 'em, 'cuz they're
> > from Microsoft. I don't think that makes me a bad person.)
> I think there are rational reasons to avoid Microsoft products actually.
> In any case, I never said anything about anybody being a bad person.
There are rational reasons to continue using Velocity, too... reasons
that includes Serge's: "Honestly, you're preaching to the choir of a
different church... I've been an apache committer for 4+ years and ___am
unlikely to adopt a non-apache tool when there is a very well known, if
comfortably mature, project within Apache already.___"
I bet if Velocity sucked (which maybe it does already, who cares) he
wouldn't use it. Serge, if I'm wrong, I'll buy you and Jonathan a round
:) But only if you promise not to talk about this any more!
> If somebody deliberately goes to a given shop where things are more
> expensive and of lower quality and does not consider the range of
> options available, other places where one can buy, who does he damage
> really?
Himself?
Which makes it his business?
> Basically just himself, right?
> Suppose Serge starts using FreeMarker tomorrow. What do I gain?
Nothing?
Which makes this a curious exercise?
But somehow I'm bettin' in your head y'all one this one, hey?
*hee*
> Nothing, right?
> On that thought....
Woohooo!
Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Dave Newton wrote:
> Okay, this is my last response, for which I'm sure we're all grateful.
>
> On Wed, 2003-06-25 at 19:53, Jonathan Revusky wrote:
>
>>>2) He already told you; he _wants_ to give preferential treatment to ASF
>>>projects. No biggie, it's okay.
>>
>>I really dunno. Is it okay?
>
>
> Yeah. Unless you're going to be applying for assistance from the
> God/dess of Software in getting the Final Answer, yeah, it's okay.
Serge can use whatever software he wants. That much is clear.
>
>
>>Let's say I'm Irish and I want to hire people of Irish extraction. I
>>won't consider a Chinese or Indian candidate who is more highly
>>qualified and will work for less. This could be illegal.
>
>
> Are you insinuating that your software is being discriminated against
> and that something should be done about it?!
I was simply inviting comparisons. I was pointing out that favoritism in
certain contexts is in fact illegal.
>
>
>>If somebody working for me was going to evaluate the various
>>technologies available, and simply didn't look at the full range, but
>>only a subset, for completely arbitrary reasons.... "I've been an apache
>>committer for 4+ years so I only look at apache projects. In fact, I'll
>>use inferior technology (knowingly) because it's from ASF."
>
>
> If that had been the only reason given I might buy that. I also buy that
> Velocity _works_, so he's not choosing between FreeMarker and
> something-that-doesn't-work. We know he knows about FreeMarker; he's
> still using Velocity.
Serge specifically said he was using it because it was an apache
project. That is quite clear and there's no getting around that.
>
> Except that you're sad he's not using FreeMarker I'm not sure what your
> deal is.
I'll let you in on a little secret, Dave. I actually knew perfectly well
that Serge was never going to use FreeMarker and I knew why already.
I was just quite surprised that he openly said why. :-)
>
>
>>>>Second question: Consider my position. I know how much work I've put
>>>>into the FreeMarker XML stuff. I also know how technically inferior
>>>>things like DVSL and Anakia are. In fact, I think that you yourself do,
>>>>because you seem to have looked at FreeMarker a bit. And we both surely
>>>>know that Anakia and DVSL are largely abandoned and unsupported.
>>>
>>>
>>><emphasis-added>
>>>
>>>>How do you think I react to your telling me that you won't look at our
>>>>community's work simply because it's not part of ASF?
>>>
>>></emphasis-added>
>>>
>>>By endlessly trundling around on this great wheel called
>>>Velocity-vs-FreeMarker, I guess, which is becoming a major bummer.
>>
>>Your attempts to flame me are really quite silly. If, one day in the
>>near or medium-term future, Velocity development really is resumed, and
>>you end up being able to add 2 and 2.2 in a Velocity template, you will
>>have me to thank. It is only the pressure of continued FreeMarker
>>development, and the knowledge of how uncompetitive Velocity currently
>>is, that seems to be causing people to get off their asses a little bit.
>
>
> All I said was that (to me) you're coming across as whiny that somebody
> isn't using FreeMarker and/or that they're using something else for
> reasons you don't like.
It is true that I don't like somebody saying that he is using Velocity
simply because it's an Apache project. However, I have long stopped
being shocked by this. Though I was surprised that he openly stated it.
>
>
>>Given this, your hostility towards me is quite misplaced. You can only
>>reap benefits from my behavior.
>
>
> I'm not hostile, I'm irritated. And that last sentence is more
> irritating.
Well, you may not think you're being hostile, but I definitely perceive
you as hostile.
>
> I'm not sure if you just don't see it or are ignoring it or what, but
> it's _your_ attitude here that is putting people off _and_ putting
> people off FreeMarker (as was mentioned at least once in this thread and
> in a couple off-line threads).
Well, you seem to think that you can speak for Serge. Now you seem to
think that you can speak in the name of the collective. I don't
understand why.
>
> I would think letting the marketplace decide and letting go would be the
> healthy course to take.
Dave, economic theory says that the marketplace leads to an efficient
solution when the participants have perfect information. As for my
various posts about FreeMarker (and the state of Velocity development)
they are invariably factual, and as such, they provide information, and
hence would tend to improve the efficiency of the "marketplace".
In short, given your marketplace paradigm, it's not clear what your
objection to my behavior is.
>
>
>>As things stand, it really does seem that the people involved in
>>Velocity just don't believe that they really have to compete on
>>technical merit with other things in the same space. And there is this
>>whole mentality, where things that aren't apache projects don't really
>>exist, that seemed exemplified by Serge's post. (And of course, if
>>something doesn't really exist, you don't have to compete with it, I
>>suppose.)
>
>
> I don't see _either_ of these attitudes here.
>
> Or, clarified, maybe the Velocity people _don't_ feel the need to
> compete. Indeed, I'm not sure why you _do_.
<shrug>
You're the one who alluded to a marketplace. Competition is in the
nature of the marketplace. It's necessarily competitive, unless you have
a monopoly, but that's surely not what you meant when you suggested
letting the marketplace decide.
I mean, there's a lot of very incoherent discourse emanating from you,
where you don't seem to understand the concepts that you're juggling with.
>
> I don't know anybody that thinks along the "if it ain't ASF it ain't
> shit" lines that you bring up.
Serge Knystautas was making noises along these lines -- though he didn't
express it in terms as crass.
> Hey, if I'm a long-standing committer to
> a code base you're darn _right_ I'm going to try to stay within that
> code base as long as it's practical. I just don't think Serge thinks
> that it's important enough of an issue to switch.
The above is not a factual representation. Serge is no more a committer
to the Velocity codebase than he is to the FreeMarker codebase. He is
simply opting for Velocity because it's part of ASF.
>
> As you state, of course, I don't speak for Serge.
You were trying to, but that was not legitimate discourse.
>
>>Dave, I was asking Serge these questions. What is your reason for
>>jumping in and trying to answer on his behalf?
>
>
> Firstly, it's a public forum, and anything you post publicly can and
> will be read and responded to, whether quietly or out loud.
Yes, but that still doesn't explain why you think it appropriate for you
take it upon yourself to answer for Serge.
>
> Secondly, I didn't really answer any questions, I just pointed out some
> ideas in your post I didn't care for, that's all. No biggie.
>
>
>>>Jon, if you're going to get all irritated because somebody doesn't want
>>>to use FreeMarker for reasons _you_ consider suspect
>>
>>Well, c'mon. The stated reasons clearly *are* suspect! :-)
>
>
> _Some_ may be. To you.
Well, to use knowingly an inferior tool because it comes from ASF is
suspect by various common-sense criteria.
>
>
>>Well, actually the above is already a valid technical reason IMO -- that
>>you have your app running with Velocity and don't have any reason to
>>switch. That's valid! To say: "I've been an apache committer for 4+
>>years so I will only use apache projects...." That does *not* seem like
>>a technically valid reason to me!
>
>
> I never claimed it was a _technically_ valid reason, I just claimed it's
> a _valid_ reason. For Serge.
He was asking about various XML-transformation stuff. There is anakia
and dvsl that are both Velocity-based. As far as I could tell, Serge has
no existing systems based on those things.
>
> (Just as a minor aside, there are some Microsoft products I consider
> superior to some products I use now. But I don't use 'em, 'cuz they're
> from Microsoft. I don't think that makes me a bad person.)
I think there are rational reasons to avoid Microsoft products actually.
In any case, I never said anything about anybody being a bad person.
If somebody deliberately goes to a given shop where things are more
expensive and of lower quality and does not consider the range of
options available, other places where one can buy, who does he damage
really?
Basically just himself, right?
Suppose Serge starts using FreeMarker tomorrow. What do I gain?
Nothing, right?
On that thought....
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
FreeMarker-Velocity comparison page, http://freemarker.org/fmVsVel.html
>
> Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Dave Newton <da...@solaraccess.com>.
Okay, this is my last response, for which I'm sure we're all grateful.
On Wed, 2003-06-25 at 19:53, Jonathan Revusky wrote:
> > 2) He already told you; he _wants_ to give preferential treatment to ASF
> > projects. No biggie, it's okay.
> I really dunno. Is it okay?
Yeah. Unless you're going to be applying for assistance from the
God/dess of Software in getting the Final Answer, yeah, it's okay.
> Let's say I'm Irish and I want to hire people of Irish extraction. I
> won't consider a Chinese or Indian candidate who is more highly
> qualified and will work for less. This could be illegal.
Are you insinuating that your software is being discriminated against
and that something should be done about it?!
> If somebody working for me was going to evaluate the various
> technologies available, and simply didn't look at the full range, but
> only a subset, for completely arbitrary reasons.... "I've been an apache
> committer for 4+ years so I only look at apache projects. In fact, I'll
> use inferior technology (knowingly) because it's from ASF."
If that had been the only reason given I might buy that. I also buy that
Velocity _works_, so he's not choosing between FreeMarker and
something-that-doesn't-work. We know he knows about FreeMarker; he's
still using Velocity.
Except that you're sad he's not using FreeMarker I'm not sure what your
deal is.
> >>Second question: Consider my position. I know how much work I've put
> >>into the FreeMarker XML stuff. I also know how technically inferior
> >>things like DVSL and Anakia are. In fact, I think that you yourself do,
> >>because you seem to have looked at FreeMarker a bit. And we both surely
> >>know that Anakia and DVSL are largely abandoned and unsupported.
> >
> >
> > <emphasis-added>
> >
> >>How do you think I react to your telling me that you won't look at our
> >>community's work simply because it's not part of ASF?
> >
> > </emphasis-added>
> >
> > By endlessly trundling around on this great wheel called
> > Velocity-vs-FreeMarker, I guess, which is becoming a major bummer.
>
> Your attempts to flame me are really quite silly. If, one day in the
> near or medium-term future, Velocity development really is resumed, and
> you end up being able to add 2 and 2.2 in a Velocity template, you will
> have me to thank. It is only the pressure of continued FreeMarker
> development, and the knowledge of how uncompetitive Velocity currently
> is, that seems to be causing people to get off their asses a little bit.
All I said was that (to me) you're coming across as whiny that somebody
isn't using FreeMarker and/or that they're using something else for
reasons you don't like.
> Given this, your hostility towards me is quite misplaced. You can only
> reap benefits from my behavior.
I'm not hostile, I'm irritated. And that last sentence is more
irritating.
I'm not sure if you just don't see it or are ignoring it or what, but
it's _your_ attitude here that is putting people off _and_ putting
people off FreeMarker (as was mentioned at least once in this thread and
in a couple off-line threads).
I would think letting the marketplace decide and letting go would be the
healthy course to take.
> As things stand, it really does seem that the people involved in
> Velocity just don't believe that they really have to compete on
> technical merit with other things in the same space. And there is this
> whole mentality, where things that aren't apache projects don't really
> exist, that seemed exemplified by Serge's post. (And of course, if
> something doesn't really exist, you don't have to compete with it, I
> suppose.)
I don't see _either_ of these attitudes here.
Or, clarified, maybe the Velocity people _don't_ feel the need to
compete. Indeed, I'm not sure why you _do_.
I don't know anybody that thinks along the "if it ain't ASF it ain't
shit" lines that you bring up. Hey, if I'm a long-standing committer to
a code base you're darn _right_ I'm going to try to stay within that
code base as long as it's practical. I just don't think Serge thinks
that it's important enough of an issue to switch.
As you state, of course, I don't speak for Serge.
> Dave, I was asking Serge these questions. What is your reason for
> jumping in and trying to answer on his behalf?
Firstly, it's a public forum, and anything you post publicly can and
will be read and responded to, whether quietly or out loud.
Secondly, I didn't really answer any questions, I just pointed out some
ideas in your post I didn't care for, that's all. No biggie.
> > Jon, if you're going to get all irritated because somebody doesn't want
> > to use FreeMarker for reasons _you_ consider suspect
> Well, c'mon. The stated reasons clearly *are* suspect! :-)
_Some_ may be. To you.
> Well, actually the above is already a valid technical reason IMO -- that
> you have your app running with Velocity and don't have any reason to
> switch. That's valid! To say: "I've been an apache committer for 4+
> years so I will only use apache projects...." That does *not* seem like
> a technically valid reason to me!
I never claimed it was a _technically_ valid reason, I just claimed it's
a _valid_ reason. For Serge.
(Just as a minor aside, there are some Microsoft products I consider
superior to some products I use now. But I don't use 'em, 'cuz they're
from Microsoft. I don't think that makes me a bad person.)
Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: [OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Dave Newton wrote:
> (And hopefully soon OL (Off-List))
>
> On Wed, 2003-06-25 at 17:43, Jonathan Revusky wrote:
>
>>Serge Knystautas wrote:
>>
>>>I appreciate your enthusiasm, but I've already committed to Velocity on
>>>this project and am looking to standardize on it rather than migrate to
>>>a different templating tool. The lack of new development does give me
>>>pause to think, but it's not stop or rewind.
>
>
> The remainder of Jonathon's reply is superfluous at this point as the
> "Why Velocity?" question was answered.
I didn't find the answer very satisfying. I posted a couple of follow-up
questions. I cannot oblige Serge or anyone else to answer.
>
>
>>Well, it's good in a way that you're honest about this, Serge, that you
>>say you're using something because it comes from ASF -- independently of
>>issues like quality or the lack of ongoing development -- that you're
>>honest, and don't invent other, phony reasons. However, I have 2
>>questions to ask in this regard.
>
>
> You greatly over-simplified his response and ignored other reasons.
If I somehow misrepresented what Serge was saying, it was not
intentional. However, it's up to Serge to clarify. Not you.
>
>
>>First question: Consider that you are looking for a tool to help you
>>with a job. In the first instance, you look at the entire universe --
>>apache and non-apache -- and in the second instance, you look at a
>>subset, only apache projects. On the whole, which strategy is likely to
>>dominate the other? Okay, the answer is obvious... so the real question
>>is: given that, on the whole you will do better looking at the entire
>>universe of applicable tools (not just ASF stuff) how do you justify
>>restricting your space of choices to just ASF stuff?
>
>
> 1) He doesn't _have_ to justify it to you or anybody else
Straw-man fallacy. I never said he did.
> 2) He already told you; he _wants_ to give preferential treatment to ASF
> projects. No biggie, it's okay.
I really dunno. Is it okay?
If I am charged with the task of buying widgets for a company, is it
okay if I only look at acme widgets? Should I not be covering the entire
market and trying to get the best deal? Is it okay if I say I want to
give preferential treatment to acme widgets simply because they're from
acme even if the widgets are inferior quality and more expensive?
Let's say I'm Irish and I want to hire people of Irish extraction. I
won't consider a Chinese or Indian candidate who is more highly
qualified and will work for less. This could be illegal.
If somebody working for me was going to evaluate the various
technologies available, and simply didn't look at the full range, but
only a subset, for completely arbitrary reasons.... "I've been an apache
committer for 4+ years so I only look at apache projects. In fact, I'll
use inferior technology (knowingly) because it's from ASF."
Truth told, in that position, I'd likely fire somebody who dropped that
kind of line on me. If I charged somebody with evaluating the various
options for XML transformation or for anything else, I'd want them to
look at the full range of what was available. It's a competitive,
rough-and-tumble world. Why should I tie one hand behind my back by only
considering a small subset of what is available out there? Doesn't seem
like a formula for success to me. Nosirrr.
>
>>Second question: Consider my position. I know how much work I've put
>>into the FreeMarker XML stuff. I also know how technically inferior
>>things like DVSL and Anakia are. In fact, I think that you yourself do,
>>because you seem to have looked at FreeMarker a bit. And we both surely
>>know that Anakia and DVSL are largely abandoned and unsupported.
>
>
> <emphasis-added>
>
>>How do you think I react to your telling me that you won't look at our
>>community's work simply because it's not part of ASF?
>
> </emphasis-added>
>
> By endlessly trundling around on this great wheel called
> Velocity-vs-FreeMarker, I guess, which is becoming a major bummer.
Your attempts to flame me are really quite silly. If, one day in the
near or medium-term future, Velocity development really is resumed, and
you end up being able to add 2 and 2.2 in a Velocity template, you will
have me to thank. It is only the pressure of continued FreeMarker
development, and the knowledge of how uncompetitive Velocity currently
is, that seems to be causing people to get off their asses a little bit.
Given this, your hostility towards me is quite misplaced. You can only
reap benefits from my behavior.
As things stand, it really does seem that the people involved in
Velocity just don't believe that they really have to compete on
technical merit with other things in the same space. And there is this
whole mentality, where things that aren't apache projects don't really
exist, that seemed exemplified by Serge's post. (And of course, if
something doesn't really exist, you don't have to compete with it, I
suppose.)
>
>
>>These are pointed sorts of questions, but I don't think they are rude or
>>off-topic to the discussion at hand, or flames for that matter. I am
>>curious as to how you answer them.
>
>
> It's not off-topic to the discussion at hand. Unfortunately the
> discussion at hand IS, so maybe take it to the dev-list and ponder the
> dev-lessness of Velocity there, or take it off-line.
Dave, I was asking Serge these questions. What is your reason for
jumping in and trying to answer on his behalf?
>
> Jon, if you're going to get all irritated because somebody doesn't want
> to use FreeMarker for reasons _you_ consider suspect
Well, c'mon. The stated reasons clearly *are* suspect! :-)
> then you don't need
> to continue the thread, you need to suck it up and get on with your
> life.
>
> (For the record, I'm currently using both Velocity and FreeMarker and am
> also sticking with Velocity 'cuz it's already running and working in our
> applications and I simply don't have any sort of time to deal with
> anything like switching right now.)
Well, actually the above is already a valid technical reason IMO -- that
you have your app running with Velocity and don't have any reason to
switch. That's valid! To say: "I've been an apache committer for 4+
years so I will only use apache projects...." That does *not* seem like
a technically valid reason to me!
Regards,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
FreeMarker-Velocity comparison page, http://freemarker.org/fmVsVel.html
>
> Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
[OT]Re: DVSL status? (or got another XSL replacement?)
Posted by Dave Newton <da...@solaraccess.com>.
(And hopefully soon OL (Off-List))
On Wed, 2003-06-25 at 17:43, Jonathan Revusky wrote:
> Serge Knystautas wrote:
> > I appreciate your enthusiasm, but I've already committed to Velocity on
> > this project and am looking to standardize on it rather than migrate to
> > a different templating tool. The lack of new development does give me
> > pause to think, but it's not stop or rewind.
The remainder of Jonathon's reply is superfluous at this point as the
"Why Velocity?" question was answered.
> Well, it's good in a way that you're honest about this, Serge, that you
> say you're using something because it comes from ASF -- independently of
> issues like quality or the lack of ongoing development -- that you're
> honest, and don't invent other, phony reasons. However, I have 2
> questions to ask in this regard.
You greatly over-simplified his response and ignored other reasons.
> First question: Consider that you are looking for a tool to help you
> with a job. In the first instance, you look at the entire universe --
> apache and non-apache -- and in the second instance, you look at a
> subset, only apache projects. On the whole, which strategy is likely to
> dominate the other? Okay, the answer is obvious... so the real question
> is: given that, on the whole you will do better looking at the entire
> universe of applicable tools (not just ASF stuff) how do you justify
> restricting your space of choices to just ASF stuff?
1) He doesn't _have_ to justify it to you or anybody else
2) He already told you; he _wants_ to give preferential treatment to ASF
projects. No biggie, it's okay.
> Second question: Consider my position. I know how much work I've put
> into the FreeMarker XML stuff. I also know how technically inferior
> things like DVSL and Anakia are. In fact, I think that you yourself do,
> because you seem to have looked at FreeMarker a bit. And we both surely
> know that Anakia and DVSL are largely abandoned and unsupported.
<emphasis-added>
> How do you think I react to your telling me that you won't look at our
> community's work simply because it's not part of ASF?
</emphasis-added>
By endlessly trundling around on this great wheel called
Velocity-vs-FreeMarker, I guess, which is becoming a major bummer.
> These are pointed sorts of questions, but I don't think they are rude or
> off-topic to the discussion at hand, or flames for that matter. I am
> curious as to how you answer them.
It's not off-topic to the discussion at hand. Unfortunately the
discussion at hand IS, so maybe take it to the dev-list and ponder the
dev-lessness of Velocity there, or take it off-line.
Jon, if you're going to get all irritated because somebody doesn't want
to use FreeMarker for reasons _you_ consider suspect then you don't need
to continue the thread, you need to suck it up and get on with your
life.
(For the record, I'm currently using both Velocity and FreeMarker and am
also sticking with Velocity 'cuz it's already running and working in our
applications and I simply don't have any sort of time to deal with
anything like switching right now.)
Dave
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Daniel Dekany wrote:
> Saturday, June 28, 2003, 10:44:38 AM, Jonathan Revusky wrote:
>
>
>>Serge Knystautas wrote:
>>
>>>Jonathan Revusky wrote:
>
> [snip]
>
>>Well, I am rather surprised that you think that DVSL is something that
>>can be legitimately compared with FM's XML transformation capabilities.
>
>
> Well, if we are talking about "capabilities", nobody would think that
> seriously. Things starts there that DVSL uses VTL, so whatever smart the
> tool author is... It's something like VPP vs FMPP... unfair race.
>
>>DVSL is basically a quick hack written by one guy on a weekend at least
>>a year ago, and that has since been abandoned. The XML transformation
>>capabilities in FreeMarker were hammered out via a long process of
>>give-and-take by various people in our community.
>>
>>As I pointed out, the FreeMarker manual is now generated by this XML
>>transformation and this is serious documentation -- we have automatic
>>TOC generation, cross-references, auto-generated index and so on. This
>>represents some real proof that we have a tool that can get this kind of
>>job done.
>
>
> To be precise and 100% fair: The documentation generation process is
> helped, or even partially driven by a Java program written specifically
> for this task (i.e. to generate the FM Manual). Well, I didn't actually
> looked into the process in details, but I'm quite certain that my above
> statements stands.
There is java driver code, about 300 lines, that sets up certain data
structures that the FTL templates need -- for example, a hash lookup of
ID->Element that is used when generating cross-references. Also, some
data structures are built up for index cross-references and generation.
So that we can autogenerate the index file here for example:
http://freemarker.org/docs/alphaidx.html
But anyway, you're right that, strictly speaking, the FreeMarker docs
are generated by a mixed Java/FTL program -- about 300 lines of Java and
800 lines of FTL. It's rolled up as an ant task. The whole thing can be
checked out from our CVS. It's the docgen module. I want to get this
into shape so that we can actually have docgen be a separate
downloadable distro -- a fairly complete example of building complex
documentation using FreeMarker.
>
> Also, regarding to an earlier mail: note that the documentation uses a
> restricted subset of XDocBook schema. It's not a general purpose
> XDocBook style-sheet -- that would be a separate project to develop such
> a tool.
Well, Docbook is a big DTD. I think it has something like 300 elements.
We don't use all those elements, so we don't have handlers for all of
them. But, in principle, there's no problem in adding them. It's just
work. :-)
But the other thing is that we don't currently have things set up with
the clarity of what pieces are generically reusable Docbook FTL
transformation and what pieces are our own customized layer for
generating *our* docs. We need to get this into a more polished state
before making docgen a separate downloadable distro.
And anyway, the FreeMarker release with the new XML transformation stuff
is still officially a preview. This is very new stuff. But our new
docgen stuff is valid proof of concept, it shows that you can undertake
very complex documentation generation stuff using FreeMarker as your
tool. I doubt that anybody has undertaken something similar using DVSL
or anakia.
So, like I said, I think that, if somebody really wants a template
engine that has those capabilities -- i.e. can be used as an XSLT
subsitute for docs generation, FM is currently the only game in town.
And that is with the proviso that the newer XML transformation features
are only available in a preview release currently.
Ciao,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
New XML transformation capabilities: http://freemarker.org/docs/xgui.html
>
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: DVSL status? (or got another XSL replacement?)
Posted by Daniel Dekany <dd...@freemail.hu>.
Saturday, June 28, 2003, 10:44:38 AM, Jonathan Revusky wrote:
> Serge Knystautas wrote:
>> Jonathan Revusky wrote:
[snip]
> Well, I am rather surprised that you think that DVSL is something that
> can be legitimately compared with FM's XML transformation capabilities.
Well, if we are talking about "capabilities", nobody would think that
seriously. Things starts there that DVSL uses VTL, so whatever smart the
tool author is... It's something like VPP vs FMPP... unfair race.
> DVSL is basically a quick hack written by one guy on a weekend at least
> a year ago, and that has since been abandoned. The XML transformation
> capabilities in FreeMarker were hammered out via a long process of
> give-and-take by various people in our community.
>
> As I pointed out, the FreeMarker manual is now generated by this XML
> transformation and this is serious documentation -- we have automatic
> TOC generation, cross-references, auto-generated index and so on. This
> represents some real proof that we have a tool that can get this kind of
> job done.
To be precise and 100% fair: The documentation generation process is
helped, or even partially driven by a Java program written specifically
for this task (i.e. to generate the FM Manual). Well, I didn't actually
looked into the process in details, but I'm quite certain that my above
statements stands.
Also, regarding to an earlier mail: note that the documentation uses a
restricted subset of XDocBook schema. It's not a general purpose
XDocBook style-sheet -- that would be a separate project to develop such
a tool.
--
Best regards,
Daniel Dekany
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Serge Knystautas wrote:
> Jonathan Revusky wrote:
>
>> Well, Serge, I can tell that you know perfectly well that Velocity is
>> not any more or less flexible in this regard (template loaders).
>
>
> Yes, I had reviewed the FreeMarker docs and API so I could respond
> accurately to your email.
>
>>> As for your XML transformation stuff, it looks like you've gone
>>> towards DOM first, with XPath secondary. I think XPath is the key to
>>> long term usage of XML transformations... as part of the JSTL expert
>>> group, we made the xml taglib rely greatly on XPath. It makes
>>> freemarker easier to use if you're familiar with DOM, but then makes
>>> the XPath syntax a bit tougher to use. I think DVSL's core adoption
>>> of XPath makes it simpler, once you get your head around how XPath
>>> works.
>>
>>
>>
>> The FreeMarker DOM wrapper allows you to use XPath queries too. But it
>> has its own simpler navigation model that you can use for 90%+ of what
>> you want to do though. It's designed to be very easy and intuitive,
>> far simpler than XPath. The only way to know whether that design goal
>> was achieved is to actually try using it.
>>
>> But again, XPath is fully supported, so I think it's a non-issue.
>
>
> I understand. You indicated you wanted feedback on your approach for
> handling XML, so I looked at what you did and wrote something up.
Well, I'm interested in feedback. We take feedback quite seriously.
However, you should understand that I'm mostly interested in feedback
that is based on the person actually trying to use the tool.
>
> I know you have XPath support, but it is secondary to the DOM structure.
> For my money, you should ditch DOM and just force users to get
> comfortable with XPath.
I find the above comment confusing. XPath is based on DOM after all. As
for forcing users to get comfortable with XPath -- presumably "for their
own good" I guess -- I don't really think in those terms. I think in
terms of providing a tool that is as simple as it can be for people's
needs, but no simpler. I don't think in terms of "forcing" people to use
XPath or any such thing.
Quite frankly, your comments there are (to me, at least) suggestive of
rigid, dogmatic thought patterns: "XPath is *the* thing, so everyone
should have to get comfortable with it..." I think our approach to this
problem space has been more pragmatic -- lighter and more flexible.
And actually, the idea of using an XPath query engine for everything
throughout was proposed and rejected. The problem is that the semantics
clash significantly with regular FreeMarker template language semantics.
Meanwhile, most such queries I think are quite simple in practice:
getting the direct children of a given name and so on. For simpler
usage, it makes sense to have semantics that don't run against the
regular FTL semantics.
The other reason not to delegate to an XPath engine every time you want
to get some very simple bit of information is performance-related. Our
new FM-based docs generation stuff is 15x faster than the XSLT-based
thing it replaces. I suspect (I haven't done any profiling) that the
lack of reliance on XPath for everything is a big part of that speedup.
But primarily, it's based on the idea that simple things should be simple.
> Again, that's what we did in the JSTL xml
> taglib, and that's part of what I like about DVSL.
Well, I am rather surprised that you think that DVSL is something that
can be legitimately compared with FM's XML transformation capabilities.
DVSL is basically a quick hack written by one guy on a weekend at least
a year ago, and that has since been abandoned. The XML transformation
capabilities in FreeMarker were hammered out via a long process of
give-and-take by various people in our community.
As I pointed out, the FreeMarker manual is now generated by this XML
transformation and this is serious documentation -- we have automatic
TOC generation, cross-references, auto-generated index and so on. This
represents some real proof that we have a tool that can get this kind of
job done. Meanwhile, I would be surprised if anybody has ever undertaken
a comparable XML transformation project using DVSL.
Also, I have a suspicion that there is a further problem, which is that,
in its current state, Velocity is not powerful enough underlying
technology to undertake this task. The lack of any namespace-related
features just means that it wouldn't scale to a task of that scale.
Also, the whitespace-control issue is going to give you endless
headaches, I would say.
Really, quite frankly, I don't think that Velocity is a player in the
XML transformation space. And FreeMarker has not been a credible player
in that space either -- it only has become so in the current, 2.3
release cycle. I actually suspect that if you want a template engine
that can also double as an XSLT replacement, FM is probably the only
game in town right now.
>
>>> Honestly, you're preaching to the choir of a different church... I've
>>> been an apache committer for 4+ years and am unlikely to adopt a
>>> non-apache tool when there is a very well known, if comfortably
>>> mature, project within Apache already.
>>
>>
>>
>> Well, it's good in a way that you're honest about this, Serge, that
>> you say you're using something because it comes from ASF --
>> independently of issues like quality or the lack of ongoing
>> development -- that you're honest, and don't invent other, phony
>> reasons. However, I have 2 questions to ask in this regard.
>
> >
>
>> First question: Consider that you are looking for a tool to help you
>> with a job. In the first instance, you look at the entire universe --
>> apache and non-apache -- and in the second instance, you look at a
>> subset, only apache projects. On the whole, which strategy is likely
>> to dominate the other? Okay, the answer is obvious... so the real
>> question is: given that, on the whole you will do better looking at
>> the entire universe of applicable tools (not just ASF stuff) how do
>> you justify restricting your space of choices to just ASF stuff?
>
> >
>
>> Second question: Consider my position. I know how much work I've put
>> into the FreeMarker XML stuff. I also know how technically inferior
>> things like DVSL and Anakia are. In fact, I think that you yourself
>> do, because you seem to have looked at FreeMarker a bit. And we both
>> surely know that Anakia and DVSL are largely abandoned and
>> unsupported. How do you think I react to your telling me that you
>> won't look at our community's work simply because it's not part of
>> ASF? (Now, you can use whatever you want, and if you want to use an
>> inferior tool with no ongoing devleopment or support specifically
>> because it is under the Apache umbrella, that's your business. But how
>> do you think I react to this?)
>
>
> I think my comments were misconstrued.
I honestly don't think so. In what way were they misconstrued?
>
> Before responding, I went to your website and read your docs. I then
> wrote the bulk of my email with a technical comparison between Velocity
> and FreeMarker (specific to my criteria and needs). I indicated some
> rather minor points of what I like about Velocity and had earlier
> concluded they are pretty similar products. I also wrote technical
> feedback on your XML usage, and also referenced another non-apache
> product (JSTL) I had worked on.
>
> I said you were "preaching to the wrong choir" because your acrimonious
> non-technical diatribes criticizing the Velocity community for the
> benefit of FreeMarker was going to have little effect on someone who has
> worked so long with the ASF.
I don't know what "non-technical diatribes" you're referring to. Any
comments I made about the state of Velocity as a project were factually
accurate -- that there had been no CVS commits for a year, that
bug-fixes and patches had simply been ignored. Also, technical comments
where I contrasted Velocity's lack of certain basic features such as
decimal numbers and namespaces and so on with FM's capabilities, these
were also technically factual.
In any case, one does not really need to take my word for it that my
comments were accurate. It is quite obvious really. If I had said
anything that was false, it is pretty clear that there are people here
who would jump on it. The fact that nobody really challenges any of my
statements on a factual level pretty much indicates that my various
comments about the state of Velocity development have been truthful.
Now, it is understandable that, given your self-confessed partisan
political sympathies, you don't *like* me saying these things. However,
if my statements are truthful, then it is not clear to me what your
grievance is. Essentially, you are representing that you are soooo very
turned off FreeMarker by my making a series of factual statements here.
<shrug>
>
> As an interesting aside and somewhat ironic in light of your inferences,
> this weekend I had just wrote the powered-by page for my product that
> prompted this question about XSL replacements... the core pieces include
> Hibernate, XJavaDoc (support lib of XDoclet), Sitemesh, Jaxen, JavaCVS,
> Java2HTML, Checkstyle, PMD, JavaNCSS... sure there's Ant, standard
> taglibs, Velocity, Lucene, and lots of Commons libs, but still. It was
> quite interesting to see on a list.
I had no idea what libraries you're using. I was reacting to your
statement where you clearly stated that you had committed to using
Velocity over alternatives because it was from ASF.
>
>> These are pointed sorts of questions, but I don't think they are rude
>> or off-topic to the discussion at hand, or flames for that matter. I
>> am curious as to how you answer them.
>
>
> I appreciate your curiosity, and I hope the feedback in my earlier email
> (ditching DOM to go straight for XPath) will get tallied for a possible
> future recount.
There has been a month of very heavy discussion on the FreeMarker-devel
list in terms of pounding our XML transformation stuff into its current
state. There are some very intelligent people on our list. In fact,
people here will be familiar with some of them -- Daniel Dekany, Attila
Szegedi for example.
It is unlikely that the comments of a person (I mean you) who has never
even used the tool could carry much weight. IOW, I don't think that, via
a superficial glance, without any attempt to use the tool, you are
likely to be able to alert us to any issues that we have not already
considered. Anyway, we are interested in feedback, but... from users,
really... :-)
>
> I hear some criticism of Apache holding so many templating and scripting
> products, but as a macro-economist at heart, I believe strongly in
> proliferation and competition, so I wish you all the best with
> FreeMarker. Kudos to you for having done so well with FreeMarker, that
> it is a matured and a widely-known project.
>
> As I said originally, I'm looking for an XSL-replacement that will allow
> me to standardize on Velocity. So beyond giving you feedback on
> FreeMarker (that would be better served on its listserv so all the
> FreeMarker committers can hear and/or disregard), I don't think this
> thread is very productive on velocity-user.
Well, I dunno. The thread is certainly broadly on-topic: using template
engines for transforming XML, i.e. as a substitute for XSLT.
Some people might be interested in what we're talking about and other
people surely aren't. But that's typical of a technically oriented
mailing list. There are usually various threads going on, and some of
them one finds of interest and thus, one follows them. Other threads are
of lesser interest. <shrug>
Regards,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
Check out FreeMarker's new XML transformation capabilities:
http://freemarker.org/docs/xgui.html
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: DVSL status? (or got another XSL replacement?)
Posted by Serge Knystautas <se...@lokitech.com>.
Jonathan Revusky wrote:
> Well, Serge, I can tell that you know perfectly well that Velocity is
> not any more or less flexible in this regard (template loaders).
Yes, I had reviewed the FreeMarker docs and API so I could respond
accurately to your email.
>> As for your XML transformation stuff, it looks like you've gone
>> towards DOM first, with XPath secondary. I think XPath is the key to
>> long term usage of XML transformations... as part of the JSTL expert
>> group, we made the xml taglib rely greatly on XPath. It makes
>> freemarker easier to use if you're familiar with DOM, but then makes
>> the XPath syntax a bit tougher to use. I think DVSL's core adoption
>> of XPath makes it simpler, once you get your head around how XPath works.
>
>
> The FreeMarker DOM wrapper allows you to use XPath queries too. But it
> has its own simpler navigation model that you can use for 90%+ of what
> you want to do though. It's designed to be very easy and intuitive, far
> simpler than XPath. The only way to know whether that design goal was
> achieved is to actually try using it.
>
> But again, XPath is fully supported, so I think it's a non-issue.
I understand. You indicated you wanted feedback on your approach for
handling XML, so I looked at what you did and wrote something up.
I know you have XPath support, but it is secondary to the DOM structure.
For my money, you should ditch DOM and just force users to get
comfortable with XPath. Again, that's what we did in the JSTL xml
taglib, and that's part of what I like about DVSL.
>> Honestly, you're preaching to the choir of a different church... I've
>> been an apache committer for 4+ years and am unlikely to adopt a
>> non-apache tool when there is a very well known, if comfortably
>> mature, project within Apache already.
>
>
> Well, it's good in a way that you're honest about this, Serge, that you
> say you're using something because it comes from ASF -- independently of
> issues like quality or the lack of ongoing development -- that you're
> honest, and don't invent other, phony reasons. However, I have 2
> questions to ask in this regard.
>
> First question: Consider that you are looking for a tool to help you
> with a job. In the first instance, you look at the entire universe --
> apache and non-apache -- and in the second instance, you look at a
> subset, only apache projects. On the whole, which strategy is likely to
> dominate the other? Okay, the answer is obvious... so the real question
> is: given that, on the whole you will do better looking at the entire
> universe of applicable tools (not just ASF stuff) how do you justify
> restricting your space of choices to just ASF stuff?
>
> Second question: Consider my position. I know how much work I've put
> into the FreeMarker XML stuff. I also know how technically inferior
> things like DVSL and Anakia are. In fact, I think that you yourself do,
> because you seem to have looked at FreeMarker a bit. And we both surely
> know that Anakia and DVSL are largely abandoned and unsupported. How do
> you think I react to your telling me that you won't look at our
> community's work simply because it's not part of ASF? (Now, you can use
> whatever you want, and if you want to use an inferior tool with no
> ongoing devleopment or support specifically because it is under the
> Apache umbrella, that's your business. But how do you think I react to
> this?)
I think my comments were misconstrued.
Before responding, I went to your website and read your docs. I then
wrote the bulk of my email with a technical comparison between Velocity
and FreeMarker (specific to my criteria and needs). I indicated some
rather minor points of what I like about Velocity and had earlier
concluded they are pretty similar products. I also wrote technical
feedback on your XML usage, and also referenced another non-apache
product (JSTL) I had worked on.
I said you were "preaching to the wrong choir" because your acrimonious
non-technical diatribes criticizing the Velocity community for the
benefit of FreeMarker was going to have little effect on someone who has
worked so long with the ASF.
As an interesting aside and somewhat ironic in light of your inferences,
this weekend I had just wrote the powered-by page for my product that
prompted this question about XSL replacements... the core pieces include
Hibernate, XJavaDoc (support lib of XDoclet), Sitemesh, Jaxen, JavaCVS,
Java2HTML, Checkstyle, PMD, JavaNCSS... sure there's Ant, standard
taglibs, Velocity, Lucene, and lots of Commons libs, but still. It was
quite interesting to see on a list.
> These are pointed sorts of questions, but I don't think they are rude or
> off-topic to the discussion at hand, or flames for that matter. I am
> curious as to how you answer them.
I appreciate your curiosity, and I hope the feedback in my earlier email
(ditching DOM to go straight for XPath) will get tallied for a possible
future recount.
I hear some criticism of Apache holding so many templating and scripting
products, but as a macro-economist at heart, I believe strongly in
proliferation and competition, so I wish you all the best with
FreeMarker. Kudos to you for having done so well with FreeMarker, that
it is a matured and a widely-known project.
As I said originally, I'm looking for an XSL-replacement that will allow
me to standardize on Velocity. So beyond giving you feedback on
FreeMarker (that would be better served on its listserv so all the
FreeMarker committers can hear and/or disregard), I don't think this
thread is very productive on velocity-user.
--
Serge Knystautas
President
Lokitech >>> software . strategy . design >> http://www.lokitech.com
p. 301.656.5501
e. sergek@lokitech.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Serge Knystautas wrote:
> Jonathan Revusky wrote:
>
>> Serge, I think you're in luck. You see, the main emphasis of the
>> current 2.3 release cycle has been to make FreeMarker a feasible
>> substitute for XSLT in most cases. If you look at our mailing list
>> archives, that's what we've been talking about for the past month --
>> making FM into an feasible XSLT replacement. We're really eager to get
>> some feedback on this. What is coincidental about this is that, with
>> the 2.3pre5 release, I felt confident enough in the state of our XML
>> transformation stuff to announce it on XML-related forums. For
>> example, here is my post to comp.text.xml from today:
>
>
> Jonathan,
>
> I appreciate your enthusiasm, but I've already committed to Velocity on
> this project and am looking to standardize on it rather than migrate to
> a different templating tool. The lack of new development does give me
> pause to think, but it's not stop or rewind.
Well, you were the one expressing concern that there had been no commits
to velocity-dvsl in 9+ months. But surely you know that there have been
no commits to the main velocity repository for about that period of time
too.
>
> I like Velocity's flexibility,
Well, Serge, I can tell that you know perfectly well that Velocity is
not any more or less flexible in this regard (template loaders). There
is a built-in RDBMS template loader in the velocity distro, I think, but
it's quite trivial. In FM, you simply implement this interface:
http://freemarker.org/docs/api/freemarker/cache/TemplateLoader.html
You could load templates from a database, or from LDAP or wherever.
> as in one case I'm using layered virtual
> database template loaders (in FreeMarker parlance) and another just a
> static basic eval without any real template. With FreeMarker, it looks
> like I would have to write the database template loader. It does look
> like the multitemplateloader would work nicely for the layering, and I'm
> sure there's a way to do non-template based freemarker evaluation (just
> pass a String to use as the template text). All in all, not dissimilar
> products.
>
>> The current FreeMarker manual was generated from a canonical docbook
>> XML format using FreeMarker itself for the transformation. Before this
>> version, we were using the Norman Walsh XSLT stylesheets. The FM-based
>> transformation is about 15 times faster than the XSLT stuff using
>> Saxon. The output is equivalent. Here it is, in fact:
>
>
> As for your XML transformation stuff, it looks like you've gone towards
> DOM first, with XPath secondary. I think XPath is the key to long term
> usage of XML transformations... as part of the JSTL expert group, we
> made the xml taglib rely greatly on XPath. It makes freemarker easier
> to use if you're familiar with DOM, but then makes the XPath syntax a
> bit tougher to use. I think DVSL's core adoption of XPath makes it
> simpler, once you get your head around how XPath works.
The FreeMarker DOM wrapper allows you to use XPath queries too. But it
has its own simpler navigation model that you can use for 90%+ of what
you want to do though. It's designed to be very easy and intuitive, far
simpler than XPath. The only way to know whether that design goal was
achieved is to actually try using it.
But again, XPath is fully supported, so I think it's a non-issue.
>
>> Anyway, I hope that's helpful to you. You could be doing yourself a
>> favor... :-)
>
>
> Thanks, yes I believe I was doing myself a favor to rely on the velocity
> community for feedback on my issue, and they did so very quickly. :)
>
> Honestly, you're preaching to the choir of a different church... I've
> been an apache committer for 4+ years and am unlikely to adopt a
> non-apache tool when there is a very well known, if comfortably mature,
> project within Apache already.
Well, it's good in a way that you're honest about this, Serge, that you
say you're using something because it comes from ASF -- independently of
issues like quality or the lack of ongoing development -- that you're
honest, and don't invent other, phony reasons. However, I have 2
questions to ask in this regard.
First question: Consider that you are looking for a tool to help you
with a job. In the first instance, you look at the entire universe --
apache and non-apache -- and in the second instance, you look at a
subset, only apache projects. On the whole, which strategy is likely to
dominate the other? Okay, the answer is obvious... so the real question
is: given that, on the whole you will do better looking at the entire
universe of applicable tools (not just ASF stuff) how do you justify
restricting your space of choices to just ASF stuff?
Second question: Consider my position. I know how much work I've put
into the FreeMarker XML stuff. I also know how technically inferior
things like DVSL and Anakia are. In fact, I think that you yourself do,
because you seem to have looked at FreeMarker a bit. And we both surely
know that Anakia and DVSL are largely abandoned and unsupported. How do
you think I react to your telling me that you won't look at our
community's work simply because it's not part of ASF? (Now, you can use
whatever you want, and if you want to use an inferior tool with no
ongoing devleopment or support specifically because it is under the
Apache umbrella, that's your business. But how do you think I react to
this?)
These are pointed sorts of questions, but I don't think they are rude or
off-topic to the discussion at hand, or flames for that matter. I am
curious as to how you answer them.
Best Regards,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
FreeMarker for XML transformations, http://freemarker.org/docs/xgui.html
FreeMarker 2.3pre5 is out!
>
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: DVSL status? (or got another XSL replacement?)
Posted by Serge Knystautas <se...@lokitech.com>.
Jonathan Revusky wrote:
> Serge, I think you're in luck. You see, the main emphasis of the current
> 2.3 release cycle has been to make FreeMarker a feasible substitute for
> XSLT in most cases. If you look at our mailing list archives, that's
> what we've been talking about for the past month -- making FM into an
> feasible XSLT replacement. We're really eager to get some feedback on
> this. What is coincidental about this is that, with the 2.3pre5 release,
> I felt confident enough in the state of our XML transformation stuff to
> announce it on XML-related forums. For example, here is my post to
> comp.text.xml from today:
Jonathan,
I appreciate your enthusiasm, but I've already committed to Velocity on
this project and am looking to standardize on it rather than migrate to
a different templating tool. The lack of new development does give me
pause to think, but it's not stop or rewind.
I like Velocity's flexibility, as in one case I'm using layered virtual
database template loaders (in FreeMarker parlance) and another just a
static basic eval without any real template. With FreeMarker, it looks
like I would have to write the database template loader. It does look
like the multitemplateloader would work nicely for the layering, and I'm
sure there's a way to do non-template based freemarker evaluation (just
pass a String to use as the template text). All in all, not dissimilar
products.
> The current FreeMarker manual was generated from a canonical docbook XML
> format using FreeMarker itself for the transformation. Before this
> version, we were using the Norman Walsh XSLT stylesheets. The FM-based
> transformation is about 15 times faster than the XSLT stuff using Saxon.
> The output is equivalent. Here it is, in fact:
As for your XML transformation stuff, it looks like you've gone towards
DOM first, with XPath secondary. I think XPath is the key to long term
usage of XML transformations... as part of the JSTL expert group, we
made the xml taglib rely greatly on XPath. It makes freemarker easier
to use if you're familiar with DOM, but then makes the XPath syntax a
bit tougher to use. I think DVSL's core adoption of XPath makes it
simpler, once you get your head around how XPath works.
> Anyway, I hope that's helpful to you. You could be doing yourself a
> favor... :-)
Thanks, yes I believe I was doing myself a favor to rely on the velocity
community for feedback on my issue, and they did so very quickly. :)
Honestly, you're preaching to the choir of a different church... I've
been an apache committer for 4+ years and am unlikely to adopt a
non-apache tool when there is a very well known, if comfortably mature,
project within Apache already.
--
Serge Knystautas
President
Lokitech >> software . strategy . design >> http://www.lokitech.com
p. 301.656.5501
e. sergek@lokitech.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: DVSL status? (or got another XSL replacement?)
Posted by Jonathan Revusky <jo...@revusky.com>.
Serge Knystautas wrote:
> I've got a documentation system that's using Velocity for much of the
> HTML output. A few of the HTML documents I want to output are coming
> directly from an XML file, and so initially I'm using XSL, but I would
> really like to use Velocity templating to expose some extra info via
> Java objects.
Oooh boy, I simply cannot resist replying to this.
Serge, I think you're in luck. You see, the main emphasis of the current
2.3 release cycle has been to make FreeMarker a feasible substitute for
XSLT in most cases. If you look at our mailing list archives, that's
what we've been talking about for the past month -- making FM into an
feasible XSLT replacement. We're really eager to get some feedback on
this. What is coincidental about this is that, with the 2.3pre5 release,
I felt confident enough in the state of our XML transformation stuff to
announce it on XML-related forums. For example, here is my post to
comp.text.xml from today:
http://groups.google.com/groups?dq=&hl=en&lr=&ie=UTF-8&oe=UTF-8&selm=5d083b27.0306250207.3d670938%40posting.google.com
So I don't need to repeat all that here. :-)
>
> Anakia looks somewhat abandoned, and DVSL seems to fit the bill pretty
> well. But it looks somewhat abandoned as well with no commits in 9+
> months.
Weeelll, yaknow, that basically reflects the state of the main Velocity
project. Frankly, if the lack of ongoing development gives you pause, I
don't know why you are so eager to use Velocity for this task.
> In fact the most recent commits were by Jason van Zyl to
> "demonstrate memory leaks," not the most encouraging sign.
>
> For using Velocity to transform XML documents, is DVSL the best thing
> available, or are there other tools I should be considering? Thanks in
> advance for any suggestions.
>
The current FreeMarker manual was generated from a canonical docbook XML
format using FreeMarker itself for the transformation. Before this
version, we were using the Norman Walsh XSLT stylesheets. The FM-based
transformation is about 15 times faster than the XSLT stuff using Saxon.
The output is equivalent. Here it is, in fact:
http://freemarker.org/docs/index.html
Anyway, I hope that's helpful to you. You could be doing yourself a
favor... :-)
Cheers,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
XML transformations with FreeMarker: http://freemarker.org/docs/xgui.html
FreeMarker 2.3pre5 is out!
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
Re: DVSL status? (or got another XSL replacement?)
Posted by Serge Knystautas <se...@lokitech.com>.
Tim Colson wrote:
>>Anakia looks somewhat abandoned, and DVSL seems to fit the
>>bill pretty well. But it looks somewhat abandoned as well
>
>
> IIRC Geir whipped up DVSL over a weekend in Dec 2001 because he was
> bored. <grin>
> http://www.mail-archive.com/velocity-user@jakarta.apache.org/msg04788.ht
> ml
>
Nice.
> Like many open source projects, if nobody has an itch, then a specific
> project area won't get scratched. At least, for me that was the case
> back then - it sounded neat, but I had no need to transform XML.
It worked really great! I was able to migrate my XSLT files to DVSL
templates and change the Java code that runs it in under an hour. Now
I've got all my velocity utilities and content attributes easily working
in those transformations!
Thanks Geir and all!
--
Serge Knystautas
President
Lokitech >>> software . strategy . design >> http://www.lokitech.com
p. 301.656.5501
e. sergek@lokitech.com
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org
RE: DVSL status? (or got another XSL replacement?)
Posted by Tim Colson <tc...@cisco.com>.
> Anakia looks somewhat abandoned, and DVSL seems to fit the
> bill pretty well. But it looks somewhat abandoned as well
IIRC Geir whipped up DVSL over a weekend in Dec 2001 because he was
bored. <grin>
http://www.mail-archive.com/velocity-user@jakarta.apache.org/msg04788.ht
ml
Over the next few months a lot of folks gave it a try and reported good
results and were happier with the syntax versus XSLT. I do not recall
any show-stopping issues.
Like many open source projects, if nobody has an itch, then a specific
project area won't get scratched. At least, for me that was the case
back then - it sounded neat, but I had no need to transform XML.
More recently though, I've helped out a tiny bit with the Velocity Tools
documentation which generates HTML from XML files using DVSL. I modified
the DVSL a bit and found it rather easy going. For what I was doing, it
works. The docs were good enough for me.
I used it again to transform the JavaOne XML schedule into something
prettier with about 10 lines of template code and the ANT task. Spiffy.
I suggest digging in and giving it a try, see if it fits for your needs.
:-)
Cheers,
Timo
---------------------------------------------------------------------
To unsubscribe, e-mail: velocity-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: velocity-user-help@jakarta.apache.org