To add my personal opinions. I'm currently leaning towards Proposal #2b. It 
keeps things relatively the same, but provides our users with more flexibility. 
The performance hit is something that we should make note of though. Here's a 
page that explains some of the performance issues around loading css.

http://www.stevesouders.com/blog/2009/04/09/dont-use-import/

I wonder if we should still depricate the fss-reset.css file and encourage the 
use of linking the two separate files directly instead?

Thanks
Justin

On 2011-03-31, at 4:18 PM, Justin Obara wrote:

> At the FSS meeting today, we talked about a couple of proposals for 
> addressing fss-reset, see below. Please let us know which solution you prefer 
> and why. I'm hoping that we'll be able to come to a consensus on what is the 
> best approach to take.
> 
> Thanks
> Justin
> 
> Proposal #1
> -----------------
> 
> Depricate the current fss-reset.css file and switch to using the reset, base, 
> and font files from YUI3, including both the global and in-context versions 
> of the files. These files would all be provided individually (not 
> concatenated).
> 
> Rationale #1
> ----------------
> 
> rely on YUI's expertise and resources
> it's more widely used and familiar CSS
> basically our current fss-reset is a componsition of YUI2's reset, base, and 
> font files. Including YUI directly is more explicit about this
> fss-reset.css mixes together reset, base, and font and might be too much for 
> some users.
> it both clears all styles and sets default styling
> 
> Cons #1
> ------------
> 
> confusing to be carrying around 2 reset files
> themes are likely built with the assumption that one of the resets has been 
> applied
> YUI3 namespaces their in-context version, but includes the version number in 
> the namespace. This will make upgrading difficult for our users.
> 
> Proposal #2
> -----------------
> 
> Keep fss-reset.css as is, and add a contextual/scoped version. This would be 
> similar to how YUI3 has an in-context version. In this case the user would be 
> able to only reset portions of the document, rather than having it affect the 
> entire page when included.
> 
> Rationale #2
> ------------------
> 
> By keeping fss-reset, it keeps things stable. There will be no need to worry 
> about backwards compatibility. 
> The additional contextual/scoped version allows for more flexibility in the 
> use of the fss-reset on a page.
> 
> Cons #2
> ------------
> 
> This doesn't include an update to the latest YUI 3 files. 
> Users who have found the FSS too heavy handed will continue to
> 
> Proposal #2b
> ------------------
> 
> Same as proposal 2, with the exception of breaking up the fss-reset into 
> multiple files. For example fss-reset-clear.css and fss-reset-base.css (names 
> are just examples, not necessarily what we will use). The fss-reset file 
> would remain, but would only include information to link in the other two 
> files. 
> 
> Rationale #2b
> --------------------
> 
> This adds even more flexibility on top of Proposal #2, by allowing users to 
> pick and choose what they like. 
> For example, if you want the reset, but not the default base styles. 
> Backwards compatibility is preserved by maintaining the single fss-reset.css 
> file which fetches the two parts.
> 
> Cons #2b
> -------------
> 
> Lots of extra files. reset will grow from 1 file to 6, after including the 
> contextual/scoped versions. 
> Importing CSS from another CSS file will slightly increase the overhead of 
> fetching the resources for a page. 
> 
> On 2011-03-17, at 9:33 AM, Justin Obara wrote:
> 
>> At the dev meeting yesterday we talked a bit about updating the fss-reset 
>> file. One of the things that we discussed was "At what point does updating 
>> fss-reset become a backwards compatibility breaking api change?".
>> 
>> Another option would be to provide the YUI css files (base, reset, and font) 
>> individually rather than concatenating them together. We could then update 
>> or deprecate the fss-reset file. This would provide backwards compatibility 
>> but allow for more flexibility for our users.
>> 
>> I'm wondering what the communities thoughts are on this issue.
>> 
>> Thanks
>> Justin
>> 
>> On 2011-03-15, at 12:59 PM, Justin Obara wrote:
>> 
>>> I've attached a pdf with the summary of the changes between the current 
>>> fss-reset and the fss-reset-new file that concats the YUI 3.3 css files.
>>> 
>>> Not that we'll probably want to maintain the custom changes.
>>> 
>>> Thanks
>>> Justin
>>> 
>>> <reset_changes.pdf>
>>> On 2011-03-14, at 3:34 PM, Justin Obara wrote:
>>> 
>>>> During our FSS chat there was some talk about the fss-reset file. Some 
>>>> felt that it was too heavy handed at times, others didn't mind. 
>>>> 
>>>> Here's a site that has info about several of the popular reset files, 
>>>> http://www.cssreset.com/ . 
>>>> 
>>>> The FSS-reset is based on the YUI v2.5.2 reset, base, and font files. YUI 
>>>> has since release YUI v3.3. In addition to the few changes they have made 
>>>> to these files, they have also provided contextual versions. What this 
>>>> means is that rather than having the styles apply globally, you can chose 
>>>> a version of the file that scopes changes to a class name. 
>>>> 
>>>> http://developer.yahoo.com/yui/3/cssreset/
>>>> 
>>>> I think we should consider upgrading to the latests version of the YUI 
>>>> reset, base, and font files, as well as providing an fss-reset-context 
>>>> file that is based on the respective contextual versions.
>>>> 
>>>> I've created a branch in my infusion fork on github which includes the 
>>>> updated reset file (fss-reset-new.css), the contextual reset file 
>>>> (fss-reset-context.css), and a test page (withNewReset.html).
>>>> 
>>>> https://github.com/jobara/infusion/tree/fss-reset-upgrade
>>>> 
>>>> Please let me know what you think.
>>>> 
>>>> Thanks
>>>> Justin
>>> 
>> 
> 

_______________________________________________________
fluid-work mailing list - [email protected]
To unsubscribe, change settings or access archives,
see http://fluidproject.org/mailman/listinfo/fluid-work

Reply via email to