You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@forrest.apache.org by Thorsten Scherler <th...@apache.org> on 2005/09/14 02:53:29 UTC

[RT] Split cocoon from forrest

Hi all, 

the update try experience made me think why are we not splitting cocoon
apart from forrest. 

I mean you would need them both to make forrest running. That would
overcome the whole procedure. We can set certain revision numbers of
cocoon-2.2 till it is not released. Kind of we having in lenya.

Any thoughts?
-- 
thorsten

"Together we stand, divided we fall!" 
Hey you (Pink Floyd)


Re: [RT] Split cocoon from forrest

Posted by Joachim Breitsprecher <jb...@dserv.net>.
David Crossley wrote:
> I don't like it.

Neither do I.


> I can visualise us having to support a stack of Cocoon user
> questions and we don't have enough people here to help.

It is not only that but also a whole bunch of new problems could be
introduced. Right now, all users of a particular Forrest revision use
the same Cocoon version. That makes it easier to track down problems and
I think of it as one of the main features of Forrest. Users and
developers don't have to worry about checking out and building the
correct revision of Cocoon to build Forrest. They just check out and
build Forrest and it works. Updating Forrest also only involves invoking
"svn up" and "./build.sh". You don't have to check after every update
whether you also have to update Cocoon.


Thorsten Scherler wrote:
> I mean you would need them both to make forrest running. That would
> overcome the whole procedure. We can set certain revision numbers of
> cocoon-2.2 till it is not released. Kind of we having in lenya.

I really like Lenya. It is a great project. However, the one thing I
*don't* like about Lenya is having to download both Lenya and Cocoon and
build them just to have a working basis. I think they should also bundle
Cocoon :-)


David Crossley wrote:
> I would rather than we Forrest committers did a little bit
> of extra work, to make it easier for our developers/users.

I am not a Forrest committer, so it is easy for me to say "+1" here. But
even if I was, I would still be "+1" :-)


-- Joachim

Re: [RT] Split cocoon from forrest

Posted by Ross Gardler <rg...@apache.org>.
Nicola Ken Barozzi wrote:
> Ross Gardler wrote:
> ...
> 
>>I've been experimenting with turning Forrest into a block. There are
>>some problem with this related to some of the stuff we do in
>>forrest.xconf, however, I have a proposal for how this can be resolved
>>(as part of the redefinition of the config files).
>>
>>When I have time, probably a week or so away, I will write this up.
> 
> 
> Seems good, as Forrest will become a set of blocks on top of Cocoon.
> Now that I think of it, we should divide Forrest into separate blocks
> too, and then have an uber-block that binds them together.
> 
> IOW we could have OTOMH these blocks:
> 
>  - linkrewriter (already there)
>  - locationmap
>  - "views"
>  - themeing
>  - etc

+1

> It would not be bad if our "plugins" became blocks too, even now.

+1 (I'm learning all about blocks now)

>>We will then be able to include the relevant parts of Cocoon from SVN
>>(use SVN external) and compile it when we run build.sh
> 
> 
> Hmmm, this is not that good ;-P
> 
> I'd prefer that we used a fixed precompiled version of Cocoon, so not to
> be stuck with an an eventually broken Cocoon, and also that we do not
> have to compile it ourselves.

I'm thinking along the lines of release early, release often. Together 
with nightly binary snapshots. However, this certainly involves more 
discussion, perhaps we should postpone that until we've made Forrest a 
block.

Ross

Re: [RT] Split cocoon from forrest

Posted by Nicola Ken Barozzi <ni...@apache.org>.
Ross Gardler wrote:
...
> I've been experimenting with turning Forrest into a block. There are
> some problem with this related to some of the stuff we do in
> forrest.xconf, however, I have a proposal for how this can be resolved
> (as part of the redefinition of the config files).
> 
> When I have time, probably a week or so away, I will write this up.

Seems good, as Forrest will become a set of blocks on top of Cocoon.
Now that I think of it, we should divide Forrest into separate blocks
too, and then have an uber-block that binds them together.

IOW we could have OTOMH these blocks:

 - linkrewriter (already there)
 - locationmap
 - "views"
 - themeing
 - etc

It would not be bad if our "plugins" became blocks too, even now.

> We will then be able to include the relevant parts of Cocoon from SVN
> (use SVN external) and compile it when we run build.sh

Hmmm, this is not that good ;-P

I'd prefer that we used a fixed precompiled version of Cocoon, so not to
be stuck with an an eventually broken Cocoon, and also that we do not
have to compile it ourselves.

> More on this soon(ish).

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
            - verba volant, scripta manent -
   (discussions get forgotten, just code remains)
---------------------------------------------------------------------


Re: [RT] Split cocoon from forrest

Posted by Ross Gardler <rg...@apache.org>.
Nicola Ken Barozzi wrote:
> David Crossley wrote:
> 
>>Thorsten Scherler wrote:
>>
>>
>>>the update try experience made me think why are we not splitting cocoon
>>>apart from forrest. 
>>>
>>>I mean you would need them both to make forrest running. That would
>>>overcome the whole procedure. We can set certain revision numbers of
>>>cocoon-2.2 till it is not released. Kind of we having in lenya.
>>>
>>>Any thoughts?
>>
>>I don't like it.
> 
> 
> Seconded
> 
> 
>>I can visualise us having to support a stack of Cocoon user
>>questions and we don't have enough people here to help.
>>
>>Also we use cocoon trunk so that makes it even harder
>>for our users.
>>
>>I would rather than we Forrest committers did a little bit
>>of extra work, to make it easier for our developers/users.
> 
> 
> Or simply rearrange and separate the Cocoon bits from the Forrest bits
> to make update easier.

I've been experimenting with turning Forrest into a block. There are 
some problem with this related to some of the stuff we do in 
forrest.xconf, however, I have a proposal for how this can be resolved 
(as part of the redefinition of the config files).

When I have time, probably a week or so away, I will write this up.

We will then be able to include the relevant parts of Cocoon from SVN 
(use SVN external) and compile it when we run build.sh

More on this soon(ish).

Ross

Re: [RT] Split cocoon from forrest

Posted by Nicola Ken Barozzi <ni...@apache.org>.
David Crossley wrote:
> Thorsten Scherler wrote:
> 
>>the update try experience made me think why are we not splitting cocoon
>>apart from forrest. 
>>
>>I mean you would need them both to make forrest running. That would
>>overcome the whole procedure. We can set certain revision numbers of
>>cocoon-2.2 till it is not released. Kind of we having in lenya.
>>
>>Any thoughts?
> 
> I don't like it.

Seconded

> I can visualise us having to support a stack of Cocoon user
> questions and we don't have enough people here to help.
> 
> Also we use cocoon trunk so that makes it even harder
> for our users.
> 
> I would rather than we Forrest committers did a little bit
> of extra work, to make it easier for our developers/users.

Or simply rearrange and separate the Cocoon bits from the Forrest bits
to make update easier.

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
            - verba volant, scripta manent -
   (discussions get forgotten, just code remains)
---------------------------------------------------------------------


Re: [RT] Split cocoon from forrest

Posted by David Crossley <cr...@apache.org>.
Thorsten Scherler wrote:
> 
> the update try experience made me think why are we not splitting cocoon
> apart from forrest. 
> 
> I mean you would need them both to make forrest running. That would
> overcome the whole procedure. We can set certain revision numbers of
> cocoon-2.2 till it is not released. Kind of we having in lenya.
> 
> Any thoughts?

I don't like it.

I can visualise us having to support a stack of Cocoon user
questions and we don't have enough people here to help.

Also we use cocoon trunk so that makes it even harder
for our users.

I would rather than we Forrest committers did a little bit
of extra work, to make it easier for our developers/users.

-David

Future target user (was Re: [RT] Split cocoon from forrest)

Posted by Ross Gardler <rg...@apache.org>.
addi wrote:
> On Tuesday September 13 2005 8:53 pm, Thorsten Scherler wrote:
> 

...

> So anyway I am -1 on the idea but I am also honestly curious about the 
> community's concept of the future target user of Forrest. And if I'm off 
> the mark or misunderstanding the intent, please let me know.

This is an interesting question and one that has come up a few times. 
There is no definitive answer since each dev has their own view on the 
*future* target user.

I think we are all in agreement that the *current* target user is a 
fairly tech savvy beast.

Personally, I see Forrest being at the core of a set of tools for end 
users who are computer literate, but little more. We have some work in 
this direction (and Eclipse based application for example). However, 
these are a long way from being usable by non technical end users.



Ross

Re: [RT] Split cocoon from forrest

Posted by addi <ad...@rocktreesky.com>.
On Tuesday September 13 2005 8:53 pm, Thorsten Scherler wrote:
> Hi all,
>
> the update try experience made me think why are we not splitting cocoon
> apart from forrest.
>
> I mean you would need them both to make forrest running. That would
> overcome the whole procedure. We can set certain revision numbers of
> cocoon-2.2 till it is not released. Kind of we having in lenya.
>
> Any thoughts?

Hm, well, coming from a user perspective I think that is a big step.  The more pieces someone has to track down and check off, the less likely they are to experiment.  Generally if I am looking for a solution and I come across one that has a list of things I need, I will keep looking and see if I can find one that doesn't have as many pieces, especially if I am looking at deadlines.  Not only is there the tracking down and installing but what if a prerequiste doesn't install right?  Even if it is my mistake with a typo or something, it stalls the whole thing.  The more pieces, the more points of potential issues.

I know that currently Forrest is in dev and rough and you need to sort of know something to use it really so installing cocoon is no big deal to most us, but I guess the question is what is the long-term goal in terms low-barrier entry for an average puter user?  I don't mean that as a rhetorical question either, because if the idea is that it will mainly be used by computer proficient people rather than "just someone who would like a cool app to make a web site", then that really sort weighs in here in my opinion.  

Cocoon is not geared toward the average puter user  They have a lot of documentation (which is great) but in their install instructions it says things like "Your mileage may vary, but you know how to setup environments, right?" while Forrest strongly promotes on the front page: " Forrest is designed with the new user in mind. Much effort has gone into making the process of generating a new site easy and simple."

So anyway I am -1 on the idea but I am also honestly curious about the community's concept of the future target user of Forrest.  And if I'm off the mark or misunderstanding the intent, please let me know.

Addi