Proposal: Basic OSS Cultural Issue: You guys really should stop being such idiots. (Was: Re: Contribution and question about getting involved in the project.)

[Jonathan Revusky-3]

Jonathan,

On Saturday 17 May 2008 00:18, Jonathan Revusky wrote:
 >> I'll explain. Look, I have been de facto project lead on FreeMarker for
 >> 6 years now. I don't know who I'm addressing, but I do not believe that
 >> you, whoever you are,  have comparable experience of running a widely
 >> used open source project. I can tell you, based on my experience, that
 >> what you are saying is... well... has no connection with reality
whatsoever.

 >This, along with the rest of your rant, is extraordinary.

<shrug>

What is extraordinary about it? I guess I could break it down into 3
possibilities:

Multiple choice question. What is extraordinary here?

(A) The willful, aggressive idiocy of people in this community...

(B) Me getting angry and frustrated by said idiocy...

(C) Me expressing my anger and frustration as I did...

Frankly, I don't see anything extraordinary about any of the three
things above.

Now, it so happens that of the three things above, there is only one of
them that you can do something about. That's A. Hence my proposal.

You guys have to have a cultural shift whereby certain kinds of idiocy
are no longer tolerated. That's not guaranteed to jump-start
development, but even if it doesn't.... it's still worthwhile, I'd say,
so it's a no-lose situation.

 >Much like similar rants by you  I have read on the webmacro and >velocity
 >lists.
 >To get 4 pages of hits  from google for "Jonathan Revusky" +troll is
 >an >achievement
 >http://www.google.com/search?hl=en&q=%22Jonathan+Revusky%22+%2Btroll

Well, there is a certain kind of aggressive, willful idiocy that I just
can't stand. But look, observe the objective technical results of this
kind of idiocy. You mention webmacro and velocity. Utterly stagnant
projects. They've been going nowhere for years, right?

And what do they have in common with JavaCC. Technical stagnation AND
agressive idiocy on the part of people involved. It's true that the
correlation does not imply causality, but it's at least a reasonable
working hypothesis: a culture of aggressive idiocy simply does not
produce technical results.

 > Sreeni was one of the original authors, so has had over ten years of
 > > leading the
 > category killer open source java parser.

Tim, JavaCC has only been an open source project for 5 years, I believe.
It was released as open source in 2003. Okay, I suppose Sreeni has been
project lead over the period, but the fact is that any objective
assessment of Sreeni's leadership of this as an open source project
would be dismal. I made a point of looking through the CVS commits from
2003 onwards. By and large, it's all fluff. I mean, maybe one or two
things happened, but the total amount of work put into this project
since 2003 is probably not more than what one motivated individual could
do in a few days.

The situation is utterly dismal. Now, after 5 years of this, that you
insult the one person who showed up who is willing and able to move
things forward substantially is just completely nutso. But it all gets
back to the basic meme of willful idiocy.


 >There is a difference in kind between the sort of tinkering that you,
I >and the others
 >are doing with Javacc and a change to the algorithms such that javacc
 >is on a
 >different mathematical footing.

Well, that's basically a red herring. There is no need to put JavaCC on
any different mathematical footing. What there is an initial need for,
and I've been working on it, is a massive code cleanup, as well as
properly separating pieces out so the codebase is parititioned in a way
that reflects a sensible SoC (separation of concerns.) Leveraging a
template engine for code output is a way of achieving SoC and also gets
a lot of the yuckiest code out of the java codebase so one can begin to
work on it.

Now, as for the history of this, let me point out that, even when JavaCC
was written in the late 1990's, the whole problem it attacks was quite
well understood and there were many implementations, some of them
available as open source that could be freely eyeballed. There was lexx
and yacc and flex and bison and so on, those are things from the 1970's.
Okay, JavaCC is largely an improvement, mainly because it targets a more
modern language and also supports the full unicode character set by
default. Still, even in the late 90's, it hardly was something
revolutionary.

I grant that, until recently, I knew nothing, for example, about the
DFA/NFA algorithms used to generate the lexers, but, you know, if you
google around, you see that this kind of stuff is part of an
undergraduate CS curriculum. (I was not a CS major.) You see it showing
up in course notes and so on that have been put up on the web.
Apparently, the NFA algorithm originates in some academic paper from
1959 and was first implemented in code by Ken Thompson (the creator of
Unix) at some point in the 1960's. You know, you talk as if Sreeni and
Sriram had actually invented this or other algorithms or something.

 >Your inability to respect that distinction means that I would have no
 >confidence in any contribution you had to make to the project.

Bingo! You see, this is the kind of willful idiocy that I'm talking
about. What are you saying in the above?

You start with the premise that I (like you and others here) am
incapable of making any fundamental contribution, only nth order
tinkering. Now, let's leave aside the question of asking what your basis
for believing this is. (I have no idea. Though, certainly, if you
believe that you are incapable of doing anything, it is bound to be a
self-fulfilling prophecy... but how this conclusion extends to me, I
really don't know.... the fact that you and others are useless gits does
not imply logically that I am.) But... let's just examine what you are
saying.

Basically, you start with a premise that I am incapable of making a
contribution, and the fact that I (delusionally, I guess) think that I
can make a contribution means that you would have no confidence in any
contribution that would be made by me....

You see, this truly is extraordinary. This isn't garden variety idiocy.
It's like a gem of supreme idiocy: You would have no confidence in any
contribution made by me precisely because I believe I am capable of
making a contribution..... (This reminds me of Groucho Marx's famous
quip that he would not sink so low as to join a club that would have him
as a member. Big difference though... that was a joke.... )

So, anyway, based on this, whose contribution would you have confidence
in then? Well, I guess, presumably, you would have confidence in the
contribution of somebody who thinks that he cannot make a contribution.

Sounds crazy, I know, but it stands to reason....

Of course, who is going to show up and contribute if they do not believe
that they are capable of making a contribution in the first place????

I mean, obviously, what you are saying here is fundamentally IDIOTIC.
And it's utterly ridiculous as well. Code is code, for crying out loud!
I put up code where you or anybody else could look at it. So, AFAIK,
nobody looks at it (with the notable exception of Paul Cager, who looked
at the code and wanted to use it.) But after not reviewing the
contribution, you now trot out the basic argument that whatever
contribution I made cannot be any good anyway, precisely because I
believe myself capable of making a contribution.

You see, my basic point here, is that you guys have to stop this kind of
IDIOCY. It just can't be tolerated. When Sreeni said that the reason not
to leverage any third party library was because the project had not done
so in the past, OBVIOUSLY he was engaging in idiocy of a high order.

But you engage in idiocy of an even higher order, second degree idiocy,
let's say, when you step in and defend him with further idiocy like
this. Still, the root problem, in my view is that, in both cases, you
guys just say idiotic things and it's considered acceptable, I guess.
Because this stuff is so transparently idiotic that other people must
see how idiotic it is. Yet nobody points it out, much less asks you to
stop. (Except me...)

Now, I really suggest that you should not further escalate to 3rd degree
idiocy, which is, after I have carefully deconstructed the idiocy of
what you're saying, to respond by linking threads where other people get
angry at me... and why.... guess what! Because I carefully deconstructed
the idiocy of what they were saying! It becomes recursive and you get
idiocy squared and idiocy cubed and so on.

The real solution has to be that you guys bite the bullet and stop being
such ******* IDIOTS!!!  And that could be hard, just like quitting
smoking or something. But I think it's possible. FWIW, I don't think the
idiocy on display by you or Sreeni or Paul Cager or Dale Anson is
congenital. If I thought that, then this would all be unfair, because
you couldn't help it. There really does seem to be something deliberate
and willful about it.

Jonathan Revusky
--
KawaDD Parser Generator http://code.google.com/p/kawadd
lead developer, FreeMarker project, http://freemarker.org/


 > TimP



Respons


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Remi Koutcherawy]

Please Jonathan, try to make short.
I can't read well English is not my native language.
And http://www.google.com/ do not give much help reading you.

I understand JavaCC is going to evolve.
Three things worth for me :
1) Do not add dependencies on a different Product
2) Keep the lexical states
3) Make JJDoc able to parse JJTree files

Because
I am trying to make evolve Eclipse plugin
http://sourceforge.net/projects/eclipse-javacc
1) dependencies are hard to handle with multi platform environment.
   Bugs report with a System a Java version  an Eclipse version I do not
have, are a pain to correct.
2) lexical states makes me switch to JavaCC l
  Terence Parr tells how to avoid them, but it means switching between
different lexers.
  http://www.antlr2.org/doc/lexer.html#Lexical_States
3) JJDoc bugs on JJTree options set in the .jjt file

For JJDoc I resort to modify JavaCC and pack it with the plugin
http://eclipse-javacc.cvs.sourceforge.net/eclipse-javacc/sf.eclipse.javacc/javacc_mod_src.zip?view=log

Rémi Koutchérawy

Le 14/06/2008 23:47, Jonathan Revusky a écrit :
>
> Jonathan,
>
> [...]
Voilà ce que Goggle traduit :

La stagnation technique et audacieux idiotie de la part de personnes
impliquées. Il est vrai que la corrélation n'implique pas causalité,
mais c'est au moins une hypothèse de travail raisonnable: une culture de
l'agressivité idiotie simplement ne pas produire les résultats techniques.

Je me demande si c'est plus compréhensible pour ceux qui le lisent dans
leur langue maternelle...
>
> Jonathan Revusky
> --
> KawaDD Parser Generator http://code.google.com/p/kawadd
> lead developer, FreeMarker project, http://freemarker.org/


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[J.Chris Findlay]

re 2: the dependency is on the build only - the build parser will not
have any extra dependency.  And that dependency is on a single extra
.jar anyway.

On 6/15/08, Remi Koutcherawy <remi.koutcherawy@...> wrote:

> Please Jonathan, try to make short.
>  I can't read well English is not my native language.
>  And http://www.google.com/ do not give much help reading you.
>
>  I understand JavaCC is going to evolve.
>  Three things worth for me :
>  1) Do not add dependencies on a different Product
>  2) Keep the lexical states
>  3) Make JJDoc able to parse JJTree files
>
>  Because
>  I am trying to make evolve Eclipse plugin
> http://sourceforge.net/projects/eclipse-javacc
>  1) dependencies are hard to handle with multi platform environment.
>   Bugs report with a System a Java version  an Eclipse version I do not
> have, are a pain to correct.
>  2) lexical states makes me switch to JavaCC l
>   Terence Parr tells how to avoid them, but it means switching between
> different lexers.
>   http://www.antlr2.org/doc/lexer.html#Lexical_States
>  3) JJDoc bugs on JJTree options set in the .jjt file
>
>  For JJDoc I resort to modify JavaCC and pack it with the plugin
> http://eclipse-javacc.cvs.sourceforge.net/eclipse-javacc/sf.eclipse.javacc/javacc_mod_src.zip?view=log
>
>  Rémi Koutchérawy
>
>  Le 14/06/2008 23:47, Jonathan Revusky a écrit :
>
> >
> > Jonathan,
> >
> > [...]
> >
>  Voilà ce que Goggle traduit :
>
>  La stagnation technique et audacieux idiotie de la part de personnes
> impliquées. Il est vrai que la corrélation n'implique pas causalité, mais
> c'est au moins une hypothèse de travail raisonnable: une culture de
> l'agressivité idiotie simplement ne pas produire les résultats techniques.
>
>  Je me demande si c'est plus compréhensible pour ceux qui le lisent dans
> leur langue maternelle...
>
> >
> > Jonathan Revusky
> > --
> > KawaDD Parser Generator http://code.google.com/p/kawadd
> > lead developer, FreeMarker project, http://freemarker.org/
> >
>
>
> ---------------------------------------------------------------------
>  To unsubscribe, e-mail:
> users-unsubscribe@...
>  For additional commands, e-mail: users-help@...
>
>

--
 - J.Chris Findlay
   (c:

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Remi Koutcherawy]

Yes
on the build, that is what bother me.
I need to make JavaCC builder (and not only generated parser)  run with
the Eclipse plugin.

Please do not use an extra tool.

Remi

Le 15/06/2008 11:35, J.Chris Findlay a écrit :
> re 2: the dependency is on the build only - the build parser will not
> have any extra dependency.  And that dependency is on a single extra
> .jar anyway



---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Jonathan Revusky-3]

Remi Koutcherawy wrote:
> Yes
> on the build, that is what bother me.
> I need to make JavaCC builder (and not only generated parser)  run with
> the Eclipse plugin.

Yes, I use the plugin. I appreciate the work you have done on that.

But really, you should try to better understand what is being discussed
before jumping in and expressing such categorical opinions.

First of all, none of these changes are going into JavaCC. I started
this work and wanted to donate it to the project, but simply could not
deal with this culture of aggressive idiocy. (I said enough about that
elsewhere.)

So I created a separate project called KawaDD which is a fork of the
JavaCC codebase. That is at http://code.google.com/p/kawadd and the
current road map is here

http://code.google.com/p/kawadd/wiki/RoadMap

Now, as a gesture for you, I just translated that page to French. It is
here:

http://code.google.com/p/kawadd/wiki/CarteDeRoute

(In case, you're wondering, I speak French like you do. Maybe better...
;--)) So you can write me any questions in French and I'll answer you in
French, but on the list, we should stick to English.

>
> Please do not use an extra tool.

Okay, look, I don't think that what you're saying is based on much
understanding of what is going on here. First of all, KawaDD is a
separate project. You don't even have to support it if you don't want
to, and since there is no official release yet, we're still talking
hypothetically.

Now, that said, I think you should plan to support it and I also think
that once you really understand what is going on here, you will want to
support it. I also don't think that there will be many technical
obstacles to supporting both JavaCC and KawaDD with your plugin, and if
there are, since I naturally want it to work, I will help you with any
problems that arise.

If you want to check whether there is a problem, just grab the latest
version and see if you have a problem getting it to work with your plugin.

svn co http://kawadd.googlecode.com/svn/trunk/kawadd
ant

Or you can see the instructions for grabbing it as a project in eclipse
that I put up here.

http://code.google.com/p/kawadd/wiki/StartHacking

(Sorry, only in English.)

I do not see offhand why you would have any problem getting this to
work. But if you do, write me on the KawaDD discussion list or in
private (and then it can be in French if you want).

Regards,

Jonathan Revusky
--
KawaDD Parser Generator, http://code.google.com/p/kawadd
lead developer, FreeMarker project, http://freemarker.org/



>
> Remi
>
> Le 15/06/2008 11:35, J.Chris Findlay a écrit :
>> re 2: the dependency is on the build only - the build parser will not
>> have any extra dependency.  And that dependency is on a single extra
>> .jar anyway


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Jonathan Revusky-3]

Remi Koutcherawy wrote:
> Please Jonathan, try to make short.
> I can't read well English is not my native language.
> And http://www.google.com/ do not give much help reading you.

I think the reason that you don't understand it is mostly because you
lack the context. You would have to read through the earlier messages
and understand what had happened. You can do that if you are interested,
but I there is little reason for me to explain it further.

>
> I understand JavaCC is going to evolve.

Well,.... no, it's not. JavaCC is going to remain exactly the same. (I
don't have a crystal ball, but I think it's a very (very very very) safe
bet. :-))

The project that is going to evolve is called KawaDD.

> Three things worth for me :
> 1) Do not add dependencies on a different Product

Why do you care about this?

Suppose that instead of leveraging an existing library for templating
(FreeMarker) one added all that same functionality as classes under
org.javacc.templating.*. Then that would be okay for you because it
wouldn't be a "different product".

But that whole thing is an ersatz discussion. Technically, there is no
real difference between one thing and the other. The whole concept of a
"different product" is basically meaningless technically. Technically,
they're all just java classes. A given set of java classes being a
separate product or not is really a marketing thing, or has to do with
intellectual property and so on, but has no technical meaning.

The only thing that occurs to me is package names, if some other plugin
uses FreeMarker, a different version.... but don't eclipse plugins all
run in their own ClassLoader? Anyway, if there is a package name
conflict, there are known solutions for this. Ever seen a tool called
Jar Jar Links? Very handy. If freemarker.template.* is conflicting with
something somewhere, just run JarJar on your jars and change
freemarker.template.* to koutcheraway.freemarker.template.* and voilà,
no package name conflicts.

> 2) Keep the lexical states

There is no issue of removing lexical states. What KawaDD does remove is:

a) static parsers

b) reusing (ReIniting) parsers (you have to throw it away and new a new one)

The reason is that, on careful consideration, I have concluded that
static and reusable parsers are.... idiotic. In a language that is
inherently multithreaded like Java.

> 3) Make JJDoc able to parse JJTree files

Interesting. I think this is already achieved. Try it out with KawaDD.
You see, I intend for JJTree to melt away and to just have one program
that generates both parser and tree-builder code, so I alredy tweaked
the KawaDD parser to accept the extra productions in JJTree. (Note that
JJDoc just uses the JavaCC parser.)

>
> Because
> I am trying to make evolve Eclipse plugin
> http://sourceforge.net/projects/eclipse-javacc
> 1) dependencies are hard to handle with multi platform environment.
>   Bugs report with a System a Java version  an Eclipse version I do not
> have, are a pain to correct.

It's not an issue in this case. Any KawaDD release will be bundled with
its own freemarker.jar that has been tested to work with that specific
release.


> 2) lexical states makes me switch to JavaCC l
>  Terence Parr tells how to avoid them, but it means switching between
> different lexers.
>  http://www.antlr2.org/doc/lexer.html#Lexical_States

THe lack of lexical states made it pretty much impossible for me to
switch to ANTLR. There is no way I would remove them from KawaDD.

> 3) JJDoc bugs on JJTree options set in the .jjt file

This is trivial. The KawaDD grammar already handles the NodeDescriptor
productions, things like #DefNode(2) or whatever. It does nothing on
them right now, but it parses them. I can't remember offhand whether I
modified it to accept the extra JJTree options, but... basically,
consider it done. :-)

>
> For JJDoc I resort to modify JavaCC and pack it with the plugin
> http://eclipse-javacc.cvs.sourceforge.net/eclipse-javacc/sf.eclipse.javacc/javacc_mod_src.zip?view=log 

If you have read http://code.google.com/p/kawadd/wiki/CarteDeRoute you
will see that one near-term goal is simply to unify the three programs
and just treat them as one. Offhand, that may make things simpler for a
tool author like you. In general, I'll be interested in making changes
so that KawaDD is more suitable for tooling than JavaCC has been.

Of course, I admit that it will be easier to maintain your support for
JavaCC. Nothing is going to change out from under you. But that, of
course, is because it's a dead project.

Regards,

Jonathan Revusky
--
KawaDD Parser Generator, http://code.google.com/p/kawadd
lead developer, FreeMarker project, http://freemarker.org/

>
>
> Rémi Koutchérawy
>
> Le 14/06/2008 23:47, Jonathan Revusky a écrit :
>>
>> Jonathan,
>>
>> [...]
> Voilà ce que Goggle traduit :
>
> La stagnation technique et audacieux idiotie de la part de personnes
> impliquées. Il est vrai que la corrélation n'implique pas causalité,
> mais c'est au moins une hypothèse de travail raisonnable: une culture de
> l'agressivité idiotie simplement ne pas produire les résultats techniques.
>
> Je me demande si c'est plus compréhensible pour ceux qui le lisent dans
> leur langue maternelle...
>>
>> Jonathan Revusky
>> --
>> KawaDD Parser Generator http://code.google.com/p/kawadd
>> lead developer, FreeMarker project, http://freemarker.org/

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Stuart A. Yeates]

Extra dependencies are a problem, as pointed out previously they
multiply issues of versions, porting and debugging.

Potentially they also allow us access to great functionality without
having to write it ourselves.

In this balancing act and it's pretty much up to the challenger to
show that the benefits outweigh the problems. Since the new version is
still young, maybe we can reserve judgment until it's mature and judge
the benefits and downsides once it's done?

cheers
stuart

On Mon, Jun 16, 2008 at 12:07 AM, Remi Koutcherawy
<remi.koutcherawy@...> wrote:

> Yes
> on the build, that is what bother me.
> I need to make JavaCC builder (and not only generated parser)  run with the
> Eclipse plugin.
>
> Please do not use an extra tool.
>
> Remi
>
> Le 15/06/2008 11:35, J.Chris Findlay a écrit :
>>
>> re 2: the dependency is on the build only - the build parser will not
>> have any extra dependency.  And that dependency is on a single extra
>> .jar anyway
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@...
> For additional commands, e-mail: users-help@...
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[J.Chris Findlay]

On 6/16/08, Stuart A. Yeates <syeates@...> wrote:
> Extra dependencies are a problem, as pointed out previously they
>  multiply issues of versions, porting and debugging.

Porting is no problem as this is all written in Java.. :-p
--
 - J.Chris Findlay
   (c:

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Stuart A. Yeates]

On Mon, Jun 16, 2008 at 9:20 AM, J.Chris Findlay
<j.chris.findlay@...> wrote:
> On 6/16/08, Stuart A. Yeates <syeates@...> wrote:
>> Extra dependencies are a problem, as pointed out previously they
>>  multiply issues of versions, porting and debugging.
>
> Porting is no problem as this is all written in Java.. :-p

The number of people in various forums seeking help getting their java
software to work in a particular browser or a particular version of
the JVM suggests this is not the case.

Indeed, one of the things I'm personally hoping to come out of this
whole mess is a way to generate output targeted at a particular
version of java.

cheers
stuart

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[J.Chris Findlay]

On 6/16/08, Stuart A. Yeates <syeates@...> wrote:

> On Mon, Jun 16, 2008 at 9:20 AM, J.Chris Findlay
>  <j.chris.findlay@...> wrote:
>  > On 6/16/08, Stuart A. Yeates <syeates@...> wrote:
>  >> Extra dependencies are a problem, as pointed out previously they
>  >>  multiply issues of versions, porting and debugging.
>  >
>  > Porting is no problem as this is all written in Java.. :-p
>
>
> The number of people in various forums seeking help getting their java
>  software to work in a particular browser or a particular version of
>  the JVM suggests this is not the case.
>
>  Indeed, one of the things I'm personally hoping to come out of this
>  whole mess is a way to generate output targeted at a particular
>  version of java.

That's one of the things using a templating mechanism like FreeMarker
will make a hang of a lot easier.

>  cheers
>
> stuart
>
>
>  ---------------------------------------------------------------------
>  To unsubscribe, e-mail: users-unsubscribe@...
>  For additional commands, e-mail: users-help@...
>
>

--
 - J.Chris Findlay
   (c:

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Julian Hyde-2]

> > Much like similar rants by you  I have read on the webmacro
> > velocity lists.
> > To get 4 pages of hits  from google for "Jonathan Revusky" +troll is
> > an achievement
> > http://www.google.com/search?hl=en&q=%22Jonathan+Revusky%22+%2Btroll
>
> Well, there is a certain kind of aggressive, willful idiocy
> that I just
> can't stand. But look, observe the objective technical
> results of this
> kind of idiocy. You mention webmacro and velocity. Utterly stagnant
> projects. They've been going nowhere for years, right?

I've been reading this flame war with growing amusement for several weeks.
I'm not particularly concerned at the outcome, because JavaCC already does
pretty much everything I want it to.

Two possibilities:

1. Revusky is a completely reasonable, nice guy that you'd love to have as
part of your open source project. There is a vast conspiracy of dozens of
idiots on multiple projects trying to blacken his name because they are
idiots and they hate to see anyone writing useful code.

2. Revusky is a troll. His technical abilities and energy are undeniable,
but he can be so divisive, and reading/responding to his rants so time
consuming, that if he wanted to become a member of one of my projects, I
would politely refuse.

Which, dear reader, do you believe to be the case? As they say on radio
phone-ins, I'll take my answer off the air.

Julian Hyde


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Laughing Man]

Some parts of this message have been removed. Learn more about Nabble's security policy.

> Two possibilities:

>
> 1. Revusky is a completely reasonable, nice guy that you'd love to have as
> part of your open source project. There is a vast conspiracy of dozens of
> idiots on multiple projects trying to blacken his name because they are
> idiots and they hate to see anyone writing useful code.
>
> 2. Revusky is a troll. His technical abilities and energy are undeniable,
> but he can be so divisive, and reading/responding to his rants so time
> consuming, that if he wanted to become a member of one of my projects, I
> would politely refuse.
>
> Which, dear reader, do you believe to be the case? As they say on radio
> phone-ins, I'll take my answer off the air.
>
> Julian Hyde
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@...
> For additional commands, e-mail: users-help@...
>

I don't think he's a troll. A troll implies a deliberate attempt to infuriate people, for the schadenfreude. My own opinion is that he honestly believe what it is he's saying.

---

Discover the new Windows Vista Learn more!

[J.Chris Findlay]

I'd pick the middle-ground:
He's a disciplined coder and a generally nice guy, except for a
certain thing he's allergic to - the particular sub-brand of insanity
demonstrated by the javaCC higherups.

From where I sit I can see both sides and the reasons for both, and I
tend to generally fall on his side of the fence, probably in large
part to being similarly allergic to unsupportable but unchanged
positions.  This is the same reason why I hate M$ and HP/Compaq and
their fanbois.

On 6/16/08, Julian Hyde <julian@...> wrote:

> > > Much like similar rants by you  I have read on the webmacro
>
> > > velocity lists.
>  > > To get 4 pages of hits  from google for "Jonathan Revusky" +troll is
>  > > an achievement
>  > > http://www.google.com/search?hl=en&q=%22Jonathan+Revusky%22+%2Btroll
>  >
>  > Well, there is a certain kind of aggressive, willful idiocy
>  > that I just
>  > can't stand. But look, observe the objective technical
>  > results of this
>  > kind of idiocy. You mention webmacro and velocity. Utterly stagnant
>  > projects. They've been going nowhere for years, right?
>
>
> I've been reading this flame war with growing amusement for several weeks.
>  I'm not particularly concerned at the outcome, because JavaCC already does
>  pretty much everything I want it to.
>
>  Two possibilities:
>
>  1. Revusky is a completely reasonable, nice guy that you'd love to have as
>  part of your open source project. There is a vast conspiracy of dozens of
>  idiots on multiple projects trying to blacken his name because they are
>  idiots and they hate to see anyone writing useful code.
>
>  2. Revusky is a troll. His technical abilities and energy are undeniable,
>  but he can be so divisive, and reading/responding to his rants so time
>  consuming, that if he wanted to become a member of one of my projects, I
>  would politely refuse.
>
>  Which, dear reader, do you believe to be the case? As they say on radio
>  phone-ins, I'll take my answer off the air.
>
>
>  Julian Hyde
>
>
>
>  ---------------------------------------------------------------------
>  To unsubscribe, e-mail: users-unsubscribe@...
>  For additional commands, e-mail: users-help@...
>
>

--
 - J.Chris Findlay
   (c:

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[J.Chris Findlay]

Also, as I have discovered all too frequently (to my dismay), binary
logic doesn't work so well when applied to human interactions..

On 6/16/08, Laughing Man <xuincherguixe@...> wrote:

>
>  > Two possibilities:
> >
> > 1. Revusky is a completely reasonable, nice guy that you'd love to have as
> > part of your open source project. There is a vast conspiracy of dozens of
> > idiots on multiple projects trying to blacken his name because they are
> > idiots and they hate to see anyone writing useful code.
> >
> > 2. Revusky is a troll. His technical abilities and energy are undeniable,
> > but he can be so divisive, and reading/responding to his rants so time
> > consuming, that if he wanted to become a member of one of my projects, I
> > would politely refuse.
> >
> > Which, dear reader, do you believe to be the case? As they say on radio
> > phone-ins, I'll take my answer off the air.
> >
> > Julian Hyde
> >
> >
> >
> ---------------------------------------------------------------------
> > To unsubscribe, e-mail:
> users-unsubscribe@...
> > For additional commands, e-mail: users-help@...
> >
>
> I don't think he's a troll. A troll implies a deliberate attempt to
> infuriate people, for the schadenfreude. My own opinion is that he honestly
> believe what it is he's saying.
>
> ________________________________
> Discover the new Windows Vista Learn more!

--
 - J.Chris Findlay
   (c:

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...