[TYPO3-core] RFC #14490: New Backend design

[Benjamin Mack]

Hey guys,

I now arrived in Germany.

On 27.05.10 14:30, JoH asenau wrote:
>> Yes, for sure. If more people would contribute instead of only
>> yelling, we would be way better already.
>> What about modernbe as UXW result? Why is this not in core? I had to
>> do some of this now myself.
>
> First of all: You didn't have to do it yourself - you could have asked, but
> you didn't.
Let's don't do nitpicking on another topic, please! We all know that the 
outcome of "modernbe" was not really well communicated by all of us.

> Second: It was a decision I discussed with Benni as the current RM before.
> We decided to *not* integrate this stuff directly into the core but make it
> an extension (modernbe) instead so people can test it in real life *before*
> we move it to the core as a sysext.
Yup. Thomas Hempel is still interested in getting this integrated in 
4.5. Communication is key here.

As far as the rest of these discussions go: Yes, the patch is not 100% 
done yet. Someone who COMPLETELY reskins the backend just can't do 100% 
at once, I think that *everybody* should be aware of that, otherwise 
just try it for yourself to create your skin, and you'll soon know how 
much work it actually is.

This immense task *is* a community effort. Discussions are necessary, 
but it is also a community work when it comes to actually help out and 
get the last 20% fixed. So, how about someone takes a couple of hours 
and makes the whole thing look good in IE6; instead of coming up with 
"yeah -1 because it's not 100% done yet"?
We *need* everybody on this, it's so easy to write something on a 
mailing list (and yes, it takes time as well, time that could be well 
spent in improving the current solution), but why is it so difficult to 
actually help out, especially when we need all the help we can get?

Here are my comments for 4.4 and IE6: Yes, we all support IE6. I think 
that a good ECMS nowadays should fully support IE6. But what does "IE6 
support" means? Personally I don't think that the interface has to look 
exactly the same on IE6 as on modern browsers. I can live with the fact 
that IE6 does *not* have a gradient background, but just a plain grey 
one, or that some shadows are missing or whatever. However, IE6 support 
means that all the functionality is the same, the buttons are on the 
same spot, and that we don't have any crappy JS error messages. 
Everything else is just almost impossible to do for such a large system 
like TYPO3 (because it's never 100% done ;-)).

Here are some more statements about the technical implementations:

  * Conditional comments
Only IE-compatible, and thus, it's again a "browser-specific 
implementation" (that was the critic on the "body CSS classes" btw), and 
so, they are just like the CSS3-implementations - browser-specific. So 
the argument to use conditional comments does *not* count in my eyes. I 
see the benefit that certain styles are only loaded when IE is created 
in order to decrease DL time for other browsers, but I'd much rather 
have a generic solution for the BE (like the one with the body CSS 
classes, which *are* generic as long as they are supported from 
t3lib_div::browserInfo), and the skin itself can fiddle around with it 
on how to include it. This way, the css compressor doesn't have to deal 
with that either, otherwise this part of TYPO3 (additionally to the 
pageRenderer would worry about that as well).
IMHO the core should not care or worry (or even have an API) to include 
browser-specific stuff. If a skin wants to include specific stylesheets 
for a certain browser, it should deal with it in its extension. I hope 
this explains the IE-point quite well.


  * CSS: "data:..."
I also think that this should be solved in a different way, doesn't seem 
like a clean solution in my eyes. But i'm not too deep in the CSS-data 
material. If it's just for background images, I'd go with repeating 
plain images (just as we have in the current t3skin), to not confuse 
people with it.


  * CSS3: Use it or not?
Well, if all modern browsers (IE8/9, FF3+, Webkit, Opera) are going to 
use it, and IE6 degrades nicely without a big change, I don't see a 
problem with it - as long as these are minor visual things.


Just so you know: This patch is the final result the hard work of three 
guys in the last two weeks, day-and-night-shift.

So, I'd like to get a general opinion here: Are we able to put our butts 
together and our egos aside, and make this happen for 4.4 or should we 
drop the whole work and wait for 4.5, when the guys who are motivated 
now will not contribute anymore because of these "it's not 100% 
comments" like this?

All the best,
Benni.