Tapestry rant and suggestions

[John Meredith <ps...@t-online.de>]

1. Rant

> Ooh yeah baby, great subject line, that's the way to get some help.

I'm sorry Howard, but this particular comment annoys the hell out of me.

It seems from this above comment that some shock element (ie. indicating
that Tapestry fails in some aspect) is required to get an answer from
the people who can navigate Tapestry with their eyes closed ie. the
developers.

I've noticed with increasing regularity, that questions posted to this
list either go unanswered or the answers are minimal at best, and I
don't just mean my questions.

On more than one occasion, I've had people ask me privately if I'd
received an answer or was able solve a particular problem I'd posted to
this list. If I have an answer, I'm more than happy to help but
unfortunately this is not always the case.

This atmosphere is not particularly conducive to converting people from
other "web-app-persuasions".

Please don't get me wrong. I love using Tapestry, and I won't stop
tinkering around with it - it been invaluable to me personally, in that
using it has taught me many things about Java you just don't find in
reference books. I also appreciate the huge amount of of time, effort
and personal sacrifices that have gone into bringing Tapestry to where
it is today.

Allow me to quote something from the webpage: "The many new features of
release 3.0 mean that Tapestry is not only the most powerful web
application framework available, it is also the fastest and easiest to
adopt, regardless of whether your background is Java, Perl, XML or PHP!"

Whilst I'm aware that statement is optimistic at best (with regards to
the background bit - the rest I generally agree with), it's still tries
to make the point that any webapp development background is welcome,
which I'm also assuming means being able to at least get a semblance of
an answer - even if it be a "RTFM you moron" (which, by the way, is much
better than no answer at all).

Before a question gets answered, does one have to demonstrate that one
can "code with the best of them" or is there some peculiar way a post a
question that has eluded myself and numerous others?

If this is a private party, please let me and others know.


2. Suggestions (I realise some of these deficiencies are well known, I
only list them here for completeness). I base these suggestions on a
completely unscientific browse through questions/discussions on this
list.

- Documentation
  a) Full examples for the component reference guide
  b) Section/tutorial dedicated to HTML forms with particular emphasis
on component interaction ie. when to use a Foreach or ListEdit, how to
customise your ValidationDelegate etc.
  c) Section/tutorial dedicated to the Table family of components.
  d) The FAQ needs to be expanded considerably - possibly integrated
into the Wiki.
  e) Complete overhaul of the Wiki with regards to organisation - even
better a Tapestry based Wiki. Yes, I realise the whole premise behind a
Wiki allows me to do this myself - but some sort of minimal organisation
is required outside of the Wiki before that can happen.

- Releases
  a) Announce releases to this list as well (I don't remember seeing one
for beta-2 here, although it was announced on the jakarta site)
  b) Some sort of roadmap with a list of proposed changes in future
releases, and the implications (positive/negative) of these changes to
developing with Tapestry e.g. What benefits will integrating Hivemind
into Tapestry (which I believe is happening in 3.1) actually provide?

- The user list (I can't comment on the dev list - out of my league)
  a) Getting an answer to a question, even if it is a simple "RTFM you
idiot" (possibly with a link to the page in question) if not considered
an affront.

I do realise that the time people have available is limited, but most
people can spare a little time - and a lot of people with a little time
would in my opinion make a huge difference. I imagine there are plenty
of people willing to invest a few minutes to do something.

For those unaware of Hibernate, can I direct you to
http://hibernate.bluemars.net/ for a very good example of how a project
(commercial or otherwise) can be presented - it does have it's
drawbacks, but on the whole it is a well designed resource for anyone
new to or already familiar with Hibernate (BTW, Hibernate and Tapestry
work incredibly well together for anyone interested in an ORM solution
for their Tapestry app).

[Actually, now that I think about it Hibernate is a great example of how
one could avoid "1. Rant" (see above). I started using both Tapestry and
Hibernate around the same time, as well as teaching myself Java. I've
found it far easier to get to grips with Hibernate, based purely on the
basis of the information available on their site and the fact you are
virtually guaranteed to get an answer to even the most stupid of
questions (the answer generally being a link to some document/FAQ). See
http://hibernate.bluemars.net/38.html for some hints.]

Anyway, rather than throw some suggestions in the air and then hide
whilst hoping that someone else do them, I'm more than willing to offer
any assistance I can, be it organisational, contributions of some sort,
or even just proof reading a FAQ, Tutorial etc.

I also hope I haven't ruffled any feathers.

Best regards,

  John

-- 
John Meredith <ps...@t-online.de>

RE: Tapestry rant and suggestions

[John Meredith <ps...@t-online.de>]

[Taking this to dev, as suggested by Howard - full post included for
completeness sake]

> > > Ooh yeah baby, great subject line, that's the way to get some help.
> > 
> > I'm sorry Howard, but this particular comment annoys the hell 
> > out of me.
> 
> If it shocks people into searching from problems a little harder, then good. Tapestry is my *third*
> job, after full-time at WebCT and writing the book.

> In other forums, when *I* need help, I don't start with "I'm a newbie, but this is broken." I start
> with, "I'm having a problem and seem to be missing something."

Both of those are still asking for help though - and not from you
personally. If you're marketing Tapestry to those familiar with other
web frameworks, then you can't just assume they are going to approach
Tapestry as you would.

> > It seems from this above comment that some shock element (ie. 
> > indicating that Tapestry fails in some aspect) is required to 
> > get an answer from the people who can navigate Tapestry with 
> > their eyes closed ie. the developers.
> 
> Dan Wells would have gotten the copylefted Tapestry Rant ... except he followed up with a pretty
> detailed explanation of his situtation.

No argument from me there.

I agree, the more information is posted the easier it is to answer.

Maybe there should be some posting guidelines or something?

> > I've noticed with increasing regularity, that questions 
> > posted to this list either go unanswered or the answers are 
> > minimal at best, and I don't just mean my questions.
> 
> A certain number of questions do go unanswered, such as "I'm new to this, but how do I something
> unnecessarily complex?".

Again, you are what you eat. If you've been eating PHP, JSP, or
god-knows how many frameworks are out there for a while, then you're
thinking is going to be a result of eating PHP, JSP etc. Some things are
*not* intuitive with Tapestry when coming from another background -
trust me :-)

I would be very surprised if anyone giving Tapestry a try had
*absolutely* no web application development background at all.

If something is unnecessarily complex, a quick pointer by those more
experienced in Tapestry as to why is more helpful than to be ignored.

> In fact, I'm always pleased by the number of questions that do get answered, and not by me.

So am I - very helpful :-)

My gripe is not that others answer questions (heck, even I have a shot
if I think I know the answer), but that some questions, which on the
face of it seem like good/interesting questions/problem descriptions,
never receive an answer ... and I'm not talking about holding someone's
hand throughout a project either - people can pay for that if that's
what they need.

> > On more than one occasion, I've had people ask me privately 
> > if I'd received an answer or was able solve a particular 
> > problem I'd posted to this list. If I have an answer, I'm 
> > more than happy to help but unfortunately this is not always the case.
> > 
> > This atmosphere is not particularly conducive to converting 
> > people from other "web-app-persuasions".
> > 
> > Please don't get me wrong. I love using Tapestry, and I won't 
> > stop tinkering around with it - it been invaluable to me 
> > personally, in that using it has taught me many things about 
> > Java you just don't find in reference books. I also 
> > appreciate the huge amount of of time, effort and personal 
> > sacrifices that have gone into bringing Tapestry to where it is today.
> > 
> > Allow me to quote something from the webpage: "The many new 
> > features of release 3.0 mean that Tapestry is not only the 
> > most powerful web application framework available, it is also 
> > the fastest and easiest to adopt, regardless of whether your 
> > background is Java, Perl, XML or PHP!"
> > 
> > Whilst I'm aware that statement is optimistic at best (with 
> > regards to the background bit - the rest I generally agree 
> > with), it's still tries to make the point that any webapp 
> > development background is welcome, which I'm also assuming 
> > means being able to at least get a semblance of an answer - 
> > even if it be a "RTFM you moron" (which, by the way, is much 
> > better than no answer at all).
> 
> I'll be glad to send people to the manual, as soon as I've written it.  All my writing energies have
> gone into the book.
> Others have stepped in to help answer the flow of questions, but nobody else has yet done any work
> on documentation (not since
> Malcom and Neil did the component reference --- and David Solis has done some additional work there
> as well, count to think of it).

As mentioned below, I'm more than happy to contribute if I can.

> > Before a question gets answered, does one have to demonstrate 
> > that one can "code with the best of them" or is there some 
> > peculiar way a post a question that has eluded myself and 
> > numerous others?
> > 
> > If this is a private party, please let me and others know.
> 
> Not in the least; in fact, what pleases me most are people just *talking* about the apps they've
> created; mostly I see an endless stream of questions.

... the endless stream of questions is due to the lack of docs/faqs etc.
Whilst you can't avoid the questions altogether, you can minimise them
considerably.

> > 2. Suggestions (I realise some of these deficiencies are well 
> > known, I only list them here for completeness). I base these 
> > suggestions on a completely unscientific browse through 
> > questions/discussions on this list.
> > 
> > - Documentation
> >   a) Full examples for the component reference guide
> >   b) Section/tutorial dedicated to HTML forms with particular 
> > emphasis on component interaction ie. when to use a Foreach 
> > or ListEdit, how to customise your ValidationDelegate etc.
> >   c) Section/tutorial dedicated to the Table family of components.
> >   d) The FAQ needs to be expanded considerably - possibly 
> > integrated into the Wiki.
> >   e) Complete overhaul of the Wiki with regards to 
> > organisation - even better a Tapestry based Wiki. Yes, I 
> > realise the whole premise behind a Wiki allows me to do this 
> > myself - but some sort of minimal organisation is required 
> > outside of the Wiki before that can happen.
> 
> Tapestry has basically moved with me as I've switched from job to job; that means Tapestry doesn't
> have a "corporate sponsor", and Apache doesn't run live web apps at apache.org.
> 
> We're still stuck with the slow, awkward, unreliable, limited PHP Wiki at sf.net.
>
> Perhaps once I launch my consulting career, I can work out a permanent home for a live Tapestry home
> that could run live demos, live FAQ, Wiki in Tapestry (or Moin Moin, or SnipSnap), a component
> exchange, and so forth.

Again, I'd be happy to pitch in here as well. I'm a sys-admin (*nix only
though :-) by trade who's fallen into webapp development, must be
something for me to do.

I can arrange some server space (in Germany) as well if that's what's
required to get the ball rolling.

> > - Releases
> >   a) Announce releases to this list as well (I don't remember 
> > seeing one for beta-2 here, although it was announced on the 
> > jakarta site)
> 
> That's an ommision if it's true.
> 
> >   b) Some sort of roadmap with a list of proposed changes in 
> > future releases, and the implications (positive/negative) of 
> > these changes to developing with Tapestry e.g. What benefits 
> > will integrating Hivemind into Tapestry (which I believe is 
> > happening in 3.1) actually provide?
> 
> A discussion for the dev list.
> 
> > 
> > - The user list (I can't comment on the dev list - out of my league)
> >   a) Getting an answer to a question, even if it is a simple 
> > "RTFM you idiot" (possibly with a link to the page in 
> > question) if not considered an affront.
> > 
> > I do realise that the time people have available is limited, 
> > but most people can spare a little time - and a lot of people 
> > with a little time would in my opinion make a huge 
> > difference. I imagine there are plenty of people willing to 
> > invest a few minutes to do something.
> > 
> > For those unaware of Hibernate, can I direct you to 
> > http://hibernate.bluemars.net/ for a very good example of how a project (commercial or otherwise)
> > can be presented - it does have it's drawbacks, but on the whole it is a well designed resource for
> > anyone new to or already familiar with Hibernate (BTW, Hibernate and Tapestry work incredibly well
> > together for anyone interested in an ORM solution for their Tapestry app).
> 
> I rememeber Hibernate from before it had all of that; I wasn't an active user, so I don't know how
> Gavin and co. dealt with questions back then, but what they have now is what they evolved to.

Be that as it may, it works and I don't think it excludes Tapestry from
following the same path. It can only benefit everyone: newbie, experts
and anyone in between.

> Anyway, rather than throw some suggestions in the air and then hide whilst hoping that someone else
> do them, I'm more than willing to offer any assistance I can, be it organisational, contributions of
> some sort, or even just proof reading a FAQ, Tutorial etc.

My little rant was not meant as a personal attack at all, but rather
what I perceive to be a very minor elitist attitude, that seems to come
across in some of the posts.

[Can I ruffle some feathers now? :-)]

  - John

-- 
John Meredith <ps...@t-online.de>

RE: Tapestry rant and suggestions

["Howard M. Lewis Ship" <hl...@comcast.net>]

> -----Original Message-----
> From: John Meredith [mailto:psynix@t-online.de] 
> Sent: Saturday, August 16, 2003 5:43 AM
> To: Tapestry users
> Subject: Tapestry rant and suggestions
> 
> 
> 1. Rant
> 
> > Ooh yeah baby, great subject line, that's the way to get some help.
> 
> I'm sorry Howard, but this particular comment annoys the hell 
> out of me.

If it shocks people into searching from problems a little harder, then good. Tapestry is my *third*
job, after full-time at WebCT and writing the book.

In other forums, when *I* need help, I don't start with "I'm a newbie, but this is broken." I start
with, "I'm having a problem and seem to be missing something."


> 
> It seems from this above comment that some shock element (ie. 
> indicating that Tapestry fails in some aspect) is required to 
> get an answer from the people who can navigate Tapestry with 
> their eyes closed ie. the developers.


Dan Wells would have gotten the copylefted Tapestry Rant ... except he followed up with a pretty
detailed explanation of his situtation.



> 
> I've noticed with increasing regularity, that questions 
> posted to this list either go unanswered or the answers are 
> minimal at best, and I don't just mean my questions.

A certain number of questions do go unanswered, such as "I'm new to this, but how do I something
unnecessarily complex?".

In fact, I'm always pleased by the number of questions that do get answered, and not by me.

> 
> On more than one occasion, I've had people ask me privately 
> if I'd received an answer or was able solve a particular 
> problem I'd posted to this list. If I have an answer, I'm 
> more than happy to help but unfortunately this is not always the case.
> 
> This atmosphere is not particularly conducive to converting 
> people from other "web-app-persuasions".
> 
> Please don't get me wrong. I love using Tapestry, and I won't 
> stop tinkering around with it - it been invaluable to me 
> personally, in that using it has taught me many things about 
> Java you just don't find in reference books. I also 
> appreciate the huge amount of of time, effort and personal 
> sacrifices that have gone into bringing Tapestry to where it is today.
> 
> Allow me to quote something from the webpage: "The many new 
> features of release 3.0 mean that Tapestry is not only the 
> most powerful web application framework available, it is also 
> the fastest and easiest to adopt, regardless of whether your 
> background is Java, Perl, XML or PHP!"
> 
> Whilst I'm aware that statement is optimistic at best (with 
> regards to the background bit - the rest I generally agree 
> with), it's still tries to make the point that any webapp 
> development background is welcome, which I'm also assuming 
> means being able to at least get a semblance of an answer - 
> even if it be a "RTFM you moron" (which, by the way, is much 
> better than no answer at all).

I'll be glad to send people to the manual, as soon as I've written it.  All my writing energies have
gone into the book.
Others have stepped in to help answer the flow of questions, but nobody else has yet done any work
on documentation (not since
Malcom and Neil did the component reference --- and David Solis has done some additional work there
as well, count to think of it).

> 
> Before a question gets answered, does one have to demonstrate 
> that one can "code with the best of them" or is there some 
> peculiar way a post a question that has eluded myself and 
> numerous others?
> 
> If this is a private party, please let me and others know.

Not in the least; in fact, what pleases me most are people just *talking* about the apps they've
created; mostly I see an endless stream of questions.


> 
> 
> 2. Suggestions (I realise some of these deficiencies are well 
> known, I only list them here for completeness). I base these 
> suggestions on a completely unscientific browse through 
> questions/discussions on this list.
> 
> - Documentation
>   a) Full examples for the component reference guide
>   b) Section/tutorial dedicated to HTML forms with particular 
> emphasis on component interaction ie. when to use a Foreach 
> or ListEdit, how to customise your ValidationDelegate etc.
>   c) Section/tutorial dedicated to the Table family of components.
>   d) The FAQ needs to be expanded considerably - possibly 
> integrated into the Wiki.
>   e) Complete overhaul of the Wiki with regards to 
> organisation - even better a Tapestry based Wiki. Yes, I 
> realise the whole premise behind a Wiki allows me to do this 
> myself - but some sort of minimal organisation is required 
> outside of the Wiki before that can happen.

Tapestry has basically moved with me as I've switched from job to job; that means Tapestry doesn't
have a "corporate sponsor", and Apache doesn't run live web apps at apache.org.

We're still stuck with the slow, awkward, unreliable, limited PHP Wiki at sf.net.

Perhaps once I launch my consulting career, I can work out a permanent home for a live Tapestry home
that could run live demos, live FAQ, Wiki in Tapestry (or Moin Moin, or SnipSnap), a component
exchange, and so forth.


> 
> - Releases
>   a) Announce releases to this list as well (I don't remember 
> seeing one for beta-2 here, although it was announced on the 
> jakarta site)

That's an ommision if it's true.

>   b) Some sort of roadmap with a list of proposed changes in 
> future releases, and the implications (positive/negative) of 
> these changes to developing with Tapestry e.g. What benefits 
> will integrating Hivemind into Tapestry (which I believe is 
> happening in 3.1) actually provide?

A discussion for the dev list.

> 
> - The user list (I can't comment on the dev list - out of my league)
>   a) Getting an answer to a question, even if it is a simple 
> "RTFM you idiot" (possibly with a link to the page in 
> question) if not considered an affront.
> 
> I do realise that the time people have available is limited, 
> but most people can spare a little time - and a lot of people 
> with a little time would in my opinion make a huge 
> difference. I imagine there are plenty of people willing to 
> invest a few minutes to do something.
> 
> For those unaware of Hibernate, can I direct you to 
http://hibernate.bluemars.net/ for a very good example of how a project (commercial or otherwise)
can be presented - it does have it's drawbacks, but on the whole it is a well designed resource for
anyone new to or already familiar with Hibernate (BTW, Hibernate and Tapestry work incredibly well
together for anyone interested in an ORM solution for their Tapestry app).

I rememeber Hibernate from before it had all of that; I wasn't an active user, so I don't know how
Gavin and co. dealt with questions back then, but what they have now is what they evolved to.

[Actually, now that I think about it Hibernate is a great example of how one could avoid "1. Rant"
(see above). I started using both Tapestry and Hibernate around the same time, as well as teaching
myself Java. I've found it far easier to get to grips with Hibernate, based purely on the basis of
the information available on their site and the fact you are virtually guaranteed to get an answer
to even the most stupid of questions (the answer generally being a link to some document/FAQ). See
http://hibernate.bluemars.net/38.html for some hints.]





Anyway, rather than throw some suggestions in the air and then hide whilst hoping that someone else
do them, I'm more than willing to offer any assistance I can, be it organisational, contributions of
some sort, or even just proof reading a FAQ, Tutorial etc.

I also hope I haven't ruffled any feathers.

Take this to the dev list and ruffle a few feathers there.


Best regards,

  John

-- 
John Meredith <ps...@t-online.de>

Re: Tapestry rant and suggestions

[Vince Marco <vm...@mac.com>]

John,

I couldn't agree with you more.  IMHO, the Apache Group totally 
underestimates the value of an online community.  The notion that a 
mailing list and a Wiki is sufficient indicates a mismanagement of 
topical knowledge.  Both JBoss and Hibernate do have web-based forums, 
which I which Tapestry had.  Wiki is for loosely collecting web-based 
content, not the specific Q&A that forums support.

I too have been frustrated in that some times my questions are answered 
and sometimes they are not.  I would like to believe that perhaps 
Tapestry is trying to expand beyond the creators.  I don't agree with 
all the design decisions put into Tapestry, but I support most of them. 
  This framework is the most productive I've used yet.  The weakness is 
that it really needs more components, such as Portal components or some 
of the OpenSymphony functionality as Tapestry components.  Since it 
does not support the standard JSP approach to webapps, I think the next 
year will be very crucial to Tapestry, as to whether it can continue 
gaining momentum.  Perhaps Howard's book will help, but I would 
certainly like to see contributions beyond Howard (I'm sure you agree, 
Howard).

I am currently involved in migrating several related webapps from 
Struts/Hibernate to Tapestry/Hibernate and the shift is nothing less 
than incredible.  This shows that you don't have to move to a 
proprietary product (ie. .NET or WebObjects) to achieve a rapid webapp 
development environment.

Great post!!

Vince

On Saturday, August 16, 2003, at 03:42  am, John Meredith wrote:

> 1. Rant
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Vince Marco
Enterprise Frameworks, Inc.
vmarco@mac.com