[suggestion] Remove theme capability for SMF 3 (or earlier)

[Norv]

Yes, we should look at compiled templates, too, BlocMan. Though not for 2.x line. If you have suggestions or examples, please do tell.

I'm not sure there is much I can do about the fact that hooks changes should have come sooner.

As it stands, hooks as they've become today are an easy enough way SMF can offer modders now, for widely modded areas of SMF (where hooks exist). They allow customizing without risking edit failures, meaning support issues, whether because of custom theme code or other mod code. That will be true for 2.0, and we are looking at increasing the extension points for 2.1, along with templates and core code cleaning. Sure, we should also make them known better, along with documentation and example usage, but that will come.
Now, a way exists, and that is what 2.0 has. It is up to the modding community to take it up, if they find it useful and easy enough.

Arantor,
Yes, SMF code and architecture has its bad parts, as well as its good parts. The reasons why it seems to require heavy workarounds to make simple visual changes with less maintenance issues in the long run, are among those that should be improved.
There is a lot to discuss here, please allow me to mention again one simple possible route: analyze the mods' needs, and establish a list, as complete as possible, of what mods need to change in templates, specially what they need more often. Some are obvious, or often needed, like mashby's example, add an image or text to an existing element on the page. Others, less so. That's why we ask the modding community to let us know their needs, their requests, their suggestions for future hooks.
We can look then at improving, or, as needed, refactoring SMF, such that mods can use additional hooks, or an API, to make those changes.
If it helps modding with less maintenance issues, it can be done, and it's one of the few refactorings that can probably be made for 2.x. (where 2.x > 2.0)

[RustyBarnacle]

I'm a heavy theme user myself.  Almost all of the forums I've built over the years have had Bloc's themes, free and paid, because they are unique.  I don't really want my forums to look like everyone else's.

I also use a lot of mods.  Some work out of the box, some I need to tinker with, some I need to really think to figure out, but I usually get there eventually.  I know I'm in the minority, just showing appreciation.

[vbgamer45]

I would like to see a new section called skins/variations. Which are just .css and image replacement changes that overwrite the default theme. I do this on my free hosting sites to allow users to customize their boards.

[redone]

Either way you slice it modifications should not require theme edits. The good news is that with future versions of SMF you have the opportunity to start from a blank page.

The package manager back in the day is what brought me to use SMF rather than other boards. I still think that feature alone is a huge reason people use SMF today.

I have used Drupal to a great degree over the last year, and yeah I will here well thats a CMS and thats different. But the object of the exercise is the same - to modify a base install. You should be able to install any theme you like without issue.  Future versions of SMF give everyone the opportunity to offer customization above and beyond what has been done before.

~RedOne

[Antechinus]

Norv says:
we need a theme for smcore
we're setting up smcore.org, with a forum
Antechinus says:
ok well people want to ditch themes
so just echo the basic html
that'll f**k 'em

[KensonPlays]

Agreed. Easier mod installs!

[Arantor]

Easier mod installs and every forum looking basically the same, just with different colours. Good plan, there.

[Antechinus]

Yeah, it sucks IMO.

[Norv]

I have used Drupal to a great degree over the last year, and yeah I will here well thats a CMS and thats different. But the object of the exercise is the same - to modify a base install. You should be able to install any theme you like without issue.  Future versions of SMF give everyone the opportunity to offer customization above and beyond what has been done before.

~RedOne

As you say, Red One. However, I'd mention that actually, there is/was a difference in the exercise... if I may nitpick the term you used.
SMF up to today (RC4/RC5) has mostly treated customization as modifying the base install, in its literal sense, in almost anything. (with few exceptions like the subscriptions component). That's not the same actually, with extending the base install, overriding and changing behavior or looks, whether by hooking into it, by using an API, and not only.

[Angelina Belle]

A lot can be accomplished with CSS, if the HTML is designed with that in mind.
http://www.mezzoblue.com/zengarden/alldesigns/

[Antechinus]

Yes we all know about the zen garden thingy.

[bloc]

A lot can be accomplished with CSS, if the HTML is designed with that in mind.
http://www.mezzoblue.com/zengarden/alldesigns/

I have known CSS Zengarden for years, and its very inspiring. But there are several things that make using that model difficult in SMF:

- all list items in Zen are a set number, the same in every design. SMF is dynamic, thus any pixel-perfect area can't be made from CSS as easily.
- Zen has several markup placeholders that do nothing than act as hooks for the CSS elements. SMF markup was written to use as less markup as possible.
- Zen has one single layout, SMF has arguably 15-20, all of which are dynamic in nature. Mods also add columns and whole sections that can't be  predicted by the CSS.
- The creators of Zen knew EXACTLY what went into the markup, in SMF you will not. That made optimisations easy for Zen, while hard for SMF.

So yes, its inspiring, but its not immediately transferable. In a way SMF 2 already did in part what you suggest: it has markup that should only have to be changed by the CSS. For example the headers or the framing CSS, these can be changed only by CSS to new innovative designs(with perhaps a few more placeholders). Trouble is that the Curve CSS is really made for ONE specific design in mind, that of the round layout of Curve. Time has shown that very few actually harness the CSS power to tweak it at all.

So what would make you think those same users would handle JUST the CSS to create such inspiring designs as Zengarden spawned? IMO: nothing. My suggestion in this topic is purely to remove something which isn't used anyway - not because its "worse".

[Angelina Belle]

You make a good point.
It is easy for me to move <div class=middletext, but not easy to move
<div id=topsection (it just doesn't behave well when I try to position:fix it at the bottom of the page)

I think that, if it were easier, I'd fiddle with that to get things where I'd like to see them.

But I decided, after a couple of hours of playing one day, that I lacked several skills required to produce anything like Manuscript. Except, you know, with a nautical theme/framing images, a racing trophy and coffee-stained bylaws and a club snapshot...

Producing a nice theme simply not very easy. That's one reason "nobody" does it.

[bloc]

Yup. I've spent countless hours testing various variations on CSS Zengarden, because i *really* like the aspect of that. But in the end its an impossible task, simply because so much must be rewritten...

But nonetheless I am once again striving for a solution lmao, although its not so much Zengarden(change only CSS for new designs) as "change every HTML in every theme" type. So a different direction, unfortunelately.

I know a member once did a Zengarden theme for SMF, in that he took the designs from Zengarden and wrapped them around SMF subtemplates(that is, the boardindex, messageindex etc) so index.template was essentially a wrapper for Zengarden. But I am not sure how far he got on it.

Yes, its hard to create good themes, and even harder to produce something new and exciting in terms of layout and style. Yes, its possible to create nice variations of the Curve theme - but making Curve itself took me a few weeks trying new ideas and not stray too long from the "normal" forum(although thats what I most often LIKE to do lmao) . That was over 2-3 years ago. The team perfected the HTML/CSS for Curve, so now its easy to make those variations - but its colour-book style themeing IMHO.

[Angelina Belle]

SOME items generally contain the "expected" stuff, and so USUALLY could be made CSS-movable, because they would behave well, if they were set up to do that.
Like the button menu, the user information, etc, the top banner.

[Antechinus]

Except, you know, with a nautical theme/framing images, a racing trophy and coffee-stained bylaws and a club snapshot...

Producing a nice theme simply not very easy. That's one reason "nobody" does it.

Sounds like fun. Give me a yell if you want to try it again and get stuck with anything.

[Angelina Belle]

The first difficult step is obtaining good graphics. If I can manage that, I might just take you up on your kind offer.

Thanks!

[Antechinus]

Found this the other day. Very handy source of various images, although I haven't looked for nautical stuff yet. http://www.photos8.com/

[Angelina Belle]

(ot) but thanks. I'd need good images of club-specific items, which I don't have yet. Only lousy ones.

On the other hand, here is an example of a themer chomping at the bit.
Or else -- maybe, with a "one size fits all" theme, it would only be necessary to drag in new images and adjust the CSS a little bit.