Revisiting Spartacus-Templates

[onli]

While thinking about s9y 2.0, yellowled and some other people mentioned quite often "some templates in spartacus are awfully old, as old as default is". So i thought: let's have a look. I installed all spartacus-templates, checked "do they seem to work" and "how is the code". Regard this as part of the thinking about s9y 2.0, but i thought here is the better place to discuss this.

The list of my results, with all the templates with some factor saying "improve me":

• Green Mile: table(-based)
• xcur: not working at my installation, div soup
• Bazooka: table
• Blogger-Death v0.1: table
• blue: table
• blue-haze: table
• Blue Iceberg: table
• Brown Paper: preview missing, not downloadable (missing on the mirror?)
• carl_blue: table
• Serendipity 3.0: table, name misleading?
• carl_green: table
• codeschmiede: table
• coffee_bar: table
• Coffee Cup: table
• Blue Streak: table
• CoSTa's Nature Theme v0.5: table (heavy use)
• Serendipity v2.3: table (This is default!?)
• XBox 360: table
• Earphone: table
• Elitist Latte: table
• fifty50: table
• Gila Two: table
• HippoRatorium: table
• Idea: table
• informatif: table
• iphone.app: no preview
• Joshua: table
• Joshua-Nino 1.1: table
• Joshua Tree: table
• Kamouflage v1.1: table
• Mo3: Mentions to need some plugins and indeed dies without them. necessary to throw 500?
• MT - Clean: table
• MT - GeorgiaBlue: table
• MT - GettysBurg: table
• MT - Plainjane: table
• MT - Rusty: table
• MT - Trendy: table
• MT3 - Chalkboard: table
• MT3 - Gettysburg: table
• MT3 - Independence: table
• MT3 - Squash: table
• newfangled: table
• Newspaper Blog: table
• Party Brown: table
• Powered: table
• BlueGrey: table
• Orange: table
• s9y Thin: table
• Safety Yellow: table
• Slate Grey: table
• Styled Bar (colorsets): table
• Blogger-Tabacco: table
• Tool-Box: table
• Truth: table
• wp-clone: table
• XHTML MP 0.1: no preview
• xkur: table

I think it's important to mention that this shall not be interpreted as "they are bad, delete all of them!". In fact, i really like some of them, i like for example the simplicity of Idea and i love the death of Blogger-Death 0.1. Sure, on the other side, some of them not only have old code and are old but also look really old (e.g. most, if not all, of the mt-templates).

So, i'm not totally sure what to do with that list, and of course it's not my call to make.

But i want to suggest the following: let's improve the template-selection. I was quite impressed how many good-looking template there are in Spartacus alone, and we could improve the old ones as well to improve the selection as a whole further.
I also think that improvement of the structur of old templates, especially of default, would prevent new authors to go in the wrong direction. I for example built my own template on top of codeschmiede, which led to some issues with the tables.

If we agree we want to try that and that table-based templates should be ported to a modern div/css-stucture (or even html 5? Shall we also judge about the look? Who should do that - a new-to-build design-council?), then we could go the following route (which, as i admit, is work-intensive and needs a lot of diplomacy):
1. Contact the original authors and ask for improvements, explaing the situation with s9y 2.0 and the exact issue with the specific template.
2. See for all templates the original author is not interested in fixing wether there is someone else to improve it.
I for example have a modernised and div/css-based version of codeschmiede in use on my own blog.
2.5 Maybe even have a blogpost about that and inform about templates which miss a maintainer
3. After quite some time, before the release of s9y 2.0: Decide for every template still not improved wether to remove it from spartacus or not.

Only a suggestion. I fully accept if your opinion is "we don't want to control the template-selection that much". Otherwise, i surely would try to help with that project.

[yellowled]

While thinking about s9y 2.0, yellowled and some other people mentioned quite often "some templates in spartacus are awfully old, as old as default is".

Personally, I would vote to move at least the old mt* templates from the core to spartacus at some point in the future. If it was my choice, the core would ship with only 2 templates - default/fallback and standard (currently BP, probably something new in 2.0), but anyone's mileage may vary.

I think it's important to mention that this shall not be interpreted as "they are bad, delete all of them!"

There's nothing wrong with keeping them on spartacus.

If we agree we want to try that and that table-based templates should be ported to a modern div/css-stucture (or even html 5? Shall we also judge about the look? Who should do that - a new-to-build design-council?)

Yikes. Don't get me wrong, I'd appreciate anyone willing to do that, but personally, I wouldn't want to do it. I also think that the time this would take would be spent way better in doing new templates from scratch.

1. Contact the original authors and ask for improvements, explaing the situation with s9y 2.0 and the exact issue with the specific template.

I'm pretty sure a lot of template authors are no longer with the S9y project and could be hard to reach.

Another idea would be to split the template section on spartacus:

* current: more or less modern, usable templates; css/div-based
* outdated: table-based, but still there if you like it
* experimental: HTML5 etc.; use at your own risk

We could even have a fourth option "commercial" for non-free templates, but given the fact that Kinetic is still the only one ...

YL

[Timbalu]

Whow, Malte, good work!

What marked my interest was the factor saying "improve me" meaning commonly html table design.

Aren't there any other things to improve like having a config.inc.php, improving the config setting, missing CONSTANTS, added features to S9y in the meanwhile which can not be solved by defaults tpl files to the template, nowadays available hooks, etc etc etc (please add)...?

I think making a detailed list of these missing features under the hood and the (old) html code base to be a template as present as possible, would be very helpful for all designers, modernizers, or even users who want to decide which template is best for them.

@Yl
The idea splitting the template until then into current, outdated, experimental and "commercial" is a brilliant idea.

[yellowled]

Aren't there any other things to improve like having a config.inc.php, improving the config setting, missing CONSTANTS, added features to S9y in the meanwhile which can not be solved by defaults tpl files to the template, nowadays available hooks, etc etc etc (please add)...?

Please, by all means, anyone be very careful with the extended use of theme options. Missing options might just be what people using these outdated templates actually like about them! BP with all it's config options is not for everyone, some people like to keep it simple.

The idea splitting the template uintil then into current, outdated, experimental and "commercial" is a brilliant idea.

I still don't think it's worth a commercial section.

There is and has ever been one commercial template -- Kinetic. It would look ridiculous. Which of course could happen to any section -- I'm not sure there would be too many experimental templates as of now, but that's not too bad. But just imagine we end up with more outdated than current template ... most new users already complain about the lack of templates for S9y ...

YL

[Timbalu]

...extended use of theme options

I did not plead having extended ones, but having some might be very helpful at all.

... just imagine we end up with more outdated than current template ... most new users already complain about the lack of templates for S9y ...

Very true - so back to Malte first post!

[yellowled]

... just imagine we end up with more outdated than current template ... most new users already complain about the lack of templates for S9y ...

Very true - so back to Malte first post!

Which is why I think it is a better idea to spend time in developing new templates (haven't done much in that area myself for a while) than revamping old ones or restructuring spartacus.

YL

[onli]

I think making a detailed list of these missing features under the hood and the (old) html code base to be a template as present as possible, would be very helpful for all designers, modernizers, or even users who want to decide which template is best for them.

This exactly is part of my conflict. I feel like it would be wrong to make access to spartacus more restricted, and adding must-have-features to templates leads to that. But I'd like to have a better selection of templates, a better impression.

But if a template is really missing something, is in another point outdated, that could surely be part of the list.

Which is why I think it is a better idea to spend time in developing new templates (haven't done much in that area myself for a while) than revamping old ones or restructuring spartacus.

It'd be great to have both. New templates and a better quality of the existing ones. Since i am not a designer, I tried to approach it from the technical side, trying to be able to say "if you install a template from spartacus, you can be sure it is not using techniques from 2001".

But i am not convinced it would be worth it if you use your time for this project instead of developing new ones.

[Don Chambers]

I do not think contacting old template designers is going to work for reasons already mentioned by YL... many are long gone from the project. Changing a template merely because it is table based is not a good enough reason IMHO. Tables continue to serve a purpose, and I for one, would not want to waste what little time I have trying to convert the concept and intent of a different designer merely to port the original code from table to div/css.

Having said that, I do agree that we have some pretty stale templates in the core. Problem is, we have very little to replace them with if removed. A few years ago, my honest intent was to start cranking out templates for s9y. I locked onto one strong idea (Kinetic), and never really got any further. It is exactly what a handful of people want, but its not for the masses.

Do we really need to pull a lot of templates out of the core? Isn't this why we have a "lite" version?

[yellowled]

Do we really need to pull a lot of templates out of the core? Isn't this why we have a "lite" version?

Kind of. The lite version also doesn't have the files requires for a shared installation. My concern about outdated templates in the core is that they make us "look old". Having templates with table-based layouts shipped with the core gives new users the impression that S9y is not exactly a modern blog system.

Also, most of the mt* template only vary in colors and fonts -- actually, this would be a nice project for someone willing to put a little effort in here: Put all the mt* templates into one template with a theme option to select the individual templates as colorsets. Same with the mt3* templates.

YL

[Timbalu]

If it was my choice, the core would ship with only 2 templates - default/fallback and standard (currently BP, probably something new in 2.0), but anyone's mileage may vary.

I think this is the way we should go. Reset the overhead and have the surface fresh & modern, the rest is in spartacus. This should be even possible with BP in 1.6 now. There are some more like the non Smarty ones etc.
Maybe we could add a main third one, being some kind of proofed and simple html5 skeleton to start designing an individual template without starting from scratch.

[yellowled]

Maybe we could add a main third one, being some kind of proofed and simple html5 skeleton to start designing an individual template without starting from scratch.

Personally, I don't think a barebone template should be shipped with the core since it's not usually meant for the John Doe user.

YL