forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Shaun Evans" <sh...@shaunevans.fsnet.co.uk>
Subject RE: [Skins] Leather-dev vs. scale-dev
Date Tue, 02 Nov 2004 21:20:19 GMT
> I just had a look at your skin. Before I start didn't we agree on the 
> specific containers that have been already implemented in leather?
> 
> I have choosen this container to make live in the future better. ;-)
> 
> You are using different containers then in leather. Your 
> DOM-tree looks 
> like this:
> container
> +-header
> +-left-nav
> +-content
> +-footer
> 
> But leather have (by purpose)
> container
>    +-branding
>    +-search
>    +-nav
>    +-content
>    +-siteinfo

I am aware of this now, thanks for the link. Scale-dev was simply a test
(hence the dev): now I understand entirely I will tell you what I
thought you meant originally. I thought that leather-dev was a skin
where everything was controlled from the CSS, where people at the CSS
end could 'build their own skin' without even touching XSL, built using
divs with IDs referenced in the CSS. Until now I was not aware that
there were standard naming conventions that leather was entirely based
around.

The search is intentionally inside branding, and I can't think of where
it would go if not there. As it is visually inside it in leather-dev
anyway, I see it as more CSS-efficient to have search inside branding. I
have done the same with the tabs.

I have already (since receiving your email, I type slowly) altered the
structure to make it like this:
 container
    +-branding
      +-search
      +-tabs
    +-nav
    +-content
    +-siteinfo

> Maybe I have to explain a wee bit more about the approach.
>  From my experience we should not introduce more skins that 
> are changing just the design. I think we have to have a 
> common base of elements that can be changed just with the 
> css. In the future we then have a wide range of css-styles 
> for the leather skin. We can then decide to just support one 
> skin. I said one skin because Corium is not really a skin but 
> a skinBot.
> 

+1 - I agree, and I believe both leather-dev and scale-dev are currently
capable of doing this. However, leather-dev's CSS is much more difficult
to work with than scale-dev (primarily because it's huge).

I think that we should work to implement this codeblock into
forrest.properties (snipped from
http://marc.theaimsgroup.com/?l=forrest-dev&m=109699727312801&w=2):

# If you want to use other css-files then the default ones than
# override it here. Point to the location of the dir containg the files
#forrest.skins.CSS.dir = /home/thorsten/style/zenCssGarden/someSkin
#forrest.skins.CSS.dir = http://provider/style/someSkin

Also, this could be added:

#project.skin=leather-dev
#project.skin.CSS.pref=basic.css

Where basic.css can be replaced with a number of different
next-generation 'skins' (which are actually css files) e.g. scale.css
(scale-dev), forrest.css (which is the next-gen forrest.apache.org
website), and new.css (which is a blank CSS except for headings for each
component of leather, used for creating new 'skins').

> I would like to dedicate one page for leather. I will write a 
> background section, why we have choosen the approach of 
> semantic container, and a section about how we want to use it 
> in the future.

+1, willing to help with documentation on editing the CSS.

> P.S.: Please, Shaun, try to rewrite the css that I can just 
> exchange the css-files. If you have problems with some 
> positioning than write to this list and we can see how we can 
> solve it. BTW it do not have to be the 
> same outcome as the original Mozilla one.

It's not just the positioning that I was having trouble with in
leather-dev. Rather than having a huge skin that I can't manage, I
prefer to work from the ground up. If you want me to, I will post my
latest scale-dev (which comes closer to the leather-dev) in JIRA.
IMHO, and I mean no offense by saying this, I would like to see
scale-dev used because it _works_, whereas leather-dev doesn't (from the
viewers' point of view). leather-dev, if you see my point, is a great
example of xsl and could (IMO, should) be used as a source of code to
put into scale-dev. However, as I said, it is better to get something
that works before you try to 'improve' it all the way up to a
releaseable standard.

Trying not to sound pedantic, I am quite disappointed with the way you
have reacted to my contribution of scale-dev. I am not new to Forrest,
but I am new to it's development and I am unfamiliar with the way that
you do things. Being a web designer, not a developer, at heart, I have
my own way of working with things, and it's not the same as the way you
do. If you said that I was trying to make changes to the way you do
things, you're right. Hopefully, in the future, I will be able to work
with the way you do things, and you will take my opinion into account.
(I distinctly remember something similar happening last week with Sean
Wheller)

Thorsten: Excuse me if what I have said isn't to your taste. Our end
views with regard to scale and leather are much the same, but you
believe that it should be done using leather, and I believe the same for
scale. Please consider my point of using scale as a base, and leather as
a source of advanced features that I haven't implemented in scale-dev
yet, as it makes life easy and simple, and I am willing to help you with
it.

Thanks
#shaun


Mime
View raw message