USe variables for site scheme colours, allowing users to change their colour scheme

[matgb]

Title:
USe variables for site scheme colours, allowing users to change their colour scheme

Area:
Site Scheme

Summary:
Allow users to pick the colours used within their preferred site scheme, probably by setting the existing colourschemes as default variables, allowing people to skin the site as they wish without the need for many many site scheme colour scheme requests and suggestions.

Description:
Me, I love Gradation, and can handle Tropo Purple, but the rest of the colours do little for me, and I actually find Tropo Red a little offputting. I do like the paler off white background colour though, means I don't have to worry about getting a headache if I have to login from the homepage.

If the extant schemes are kept as main supported colour schemes, but users are able to change the colours to whatever they wish, this would prevent the need for extra schemes for those with different preferences, and allow people that prefer one menu structure with a different colourscheme to adapt.

I'm told a block on having different schemes is logo design, but if all logos are there as a checkbox option on that page, we can accept extra submissions of diffferent coloured logos, within reason, but people can pick shecmes that complement the logo, I'd have no problem with having the red logo on Gradation TBH, might even prefer it.

There have been at least two suggestions I've seen recently for different types of site scheme preferences, this would pretty much eliminate the need for extra schemes (and the work that entails updating each), and allow much more flexibility.

Poll #4512 USe variables for site scheme colours, allowing users to change their colour scheme

Open to: Registered Users, detailed results viewable to: All, participants: 60

View Respondents

This suggestion:

View Answers

Should be implemented as-is.
48 (80.0%)

Should be implemented with changes. (please comment)
2 (3.3%)

Shouldn't be implemented.
1 (1.7%)

(I have no opinion)
9 (15.0%)

(Other: please comment)
0 (0.0%)

Comments

[azurelunatic]

This would be lovely! I can handle the existing siteschemes, but none of them quite say "home". (I'm a blue person.)

[branchandroot]

"Blueshift" is technically in the works, but the templating system for site styles is so appallingly clunky there are just no words for what a PITA making and maintaining different site styles is.

Honestly, if the templating system could just be reformed to something sane, so the site always uses the same html, just arranged and/or colored in different ways as per style or user desires, it seems like that would fix so very many downstream problems. This one included!

[matgb]

if the templating system could just be reformed to something sane, so the site always uses the same html, just arranged and/or colored in different ways as per style or user desires

I've not looked at it, as the code'll be beyond me anyway, but this would suit me completely, as a long term solution it'd be grand.

I knew having multipe schemes is a PITA, hence the desire to negate the need with this, if a much bigger solution is needed and doable then that'd suit me, but if this could be done fairly easily as a stop-gap, or it could be done as an early step in the much larger project, that'd be useful.

So many people find Trop Red very offputting it bothers me, and teaching people how to login, stay logged in, and change their styles is at times difficult. I find it easy, but I'm nowhere close to a typical use case.

[branchandroot]

*nodnod* I'd be all about some kind of user-cookie that kept track of a handful of color settings. Then we could have the current three (or three and a half...) basic layouts for form and just slap on an extra style sheet to define colors.

Actually, that might work as both stopgap and something that will lend itself to eventual overhaul of the templates--the user-selected colors could just go into the last site style sheet in line with !important tacked on to each color, and that would do it fine as things stand.

...except with Celerity, which uses images to produce the trailing lines. Hmm. Maybe any user-selected colors applied to Celerity could trigger the inclusion of a few lines of background-image:none to take the images out.

[azurelunatic]

Cannot one rebuild images in the correct color, server-side, or would that be too much of a general strain on the system? I am thinking that at the time the colors are set, the system would check the place to see if there is already an image generated in that color, if there is not one, build one, and if there is one, use that one.

[branchandroot]

I've never really explored any of the software that can do that kind of thing, so I'm not sure. I'd be afraid of that being a really massive hit to the system when this is first implemented.

[denise]

We could quite easily do an image swap, yeah.

[matgb]

An attached style sheet would work, yes, and if that's fairly easy to do, nice.

There's a little bit of me tempted to install Stylish and skin Tropo, but this netbook is good, but not good enough to overload the browser with billions of addons. Plus, if a normal user can't do it and it's something several want, it's soemthing we should do anyway.

[azurelunatic]

interest: $lj::will_not_work_without_steam_radiator

That is to say, I understand this is a problem with many of the legacy systems.

[kyrielle]

Oh. My. God.

How have I missed noticing that interest?

LOVE it. *dying*

[azurelunatic]

http://qdb.dreamwidth.net/dw/73

[kyrielle]

<3 <3 <3

So dead of laughter here.

[denise]

And then I needed an interest that had some non-alphanumeric characters for testing something, and the rest was history. *g*

[kyrielle]

You. ROCK.

Not that I've worked on systems that seem to have similar modules, but, uh, yeah.

(Another favorite smart remark around our office involves getting water in the eye from the plumbing for the kitchen sink that's somewhere in the code. Because 'everything but the kitchen sink' clearly isn't descriptive enough.)

[denise]

And let me guess: actual kitchen sink support is scheduled for version 2.0, but somebody built all the plumbing for it already. ;)

[matgb]

I hadn't seen it either, yet I regularly look at denise's profile, and it's one of two interests there...

[denise]

This is in the works. It's just really realy fucking hard. :)

[branchandroot]

I believe it! Every time I've looked at those files I back away slowly and then run screaming for the horizon.