neverendingo wrote:

ovidiub13 wrote:Why not use Django CMS? it's python and it's extremely flexible.

because it's python? Seriously though, if you want to build a team you should make sure that
- you find enough members to be able to code with that platform and
- the sysadmins are also able to handle the system while running

That sounds like django (handling while running).

From my experience, programming in python, and respectively django is by far easier and more efficient and of course less error prone than any software built in PHP. But that's just me. I'll go with what ever the majority chooses.

david_edmundson wrote:Count me in.

I think we have two possible approaches here

Top Down:
- we make a high level overview of how the entire site should be, with the main heirachy, gloal theme, then go for a big change

Bottom Up:
- we pick the most important pages; and fix the actual content before we worry about styling

IMHO https://www.kde.org/workspaces/plasmadesktop/ is one of the most pressing candidates.

I'm thinking top-down; KDE has a lot of content rot and duplicated content... Outside of snippets I heavily question how much of the site is worth 'saving'. Also, if we go with individual pages we won't be aiming for 'big picture' changes which will give meaningful long-term results.

I think what we need to do is focus on putting systems in place over pages, and be willing to work with a clean slate. If we go top-down we *will* be dropping significant amounts of content until it's re-built, but I don't really consider that a bad thing... We still have KDE3 content on the site. It would probably be best to re-add whatever is still good than try to crop whatever is bad.

ovidiub13 wrote:

neverendingo wrote:

ovidiub13 wrote:Why not use Django CMS? it's python and it's extremely flexible.

because it's python? Seriously though, if you want to build a team you should make sure that
- you find enough members to be able to code with that platform and
- the sysadmins are also able to handle the system while running

That sounds like django (handling while running).

From my experience, programming in python, and respectively django is by far easier and more efficient and of course less error prone than any software built in PHP. But that's just me. I'll go with what ever the majority chooses.

My personal leaning is towards something PHP driven, mostly because KDE.org is already heavily PHP-based and some parts (like the forums) aren't going anywhere, so it's probably better we stick to one language. Python and Django might be fantastic, but all things considered we also need to consider the fact that a python core would fragment our languages, when we really need something which can help glue it all together.

FWIW, dot.kde.org is Drupal with some extensions.

It seems to work ok-ish.

@neverendingo:

Do you remember the conclusion last time a new CMS was discussed? As I mentioned on the mailing list, I seem to recall that it ultimately boiled down to Drupal vs Wordpress, and Wordpress clearly won in several aspects (I think better performance was one of the advantages).

Of course, if it's been decided that Drupal is the way to go I won't object, I just want to make sure we make a bad decision that could've been avoided by looking at previous discussions.

@Hans:

Yes, that was the outcome of our former discussions. Performance and (like mentioned above) the people involved more used to wordpress codebase.

But no need to rely on that in this case, as the current people involved - or those interested in being involved - could have a completely different experience.

Hans wrote:@neverendingo:

Do you remember the conclusion last time a new CMS was discussed? As I mentioned on the mailing list, I seem to recall that it ultimately boiled down to Drupal vs Wordpress, and Wordpress clearly won in several aspects (I think better performance was one of the advantages).

Do you have a link to a mail with the comparison? I sifted through the July archives, but could not find anything that deep.
I do think there needs to be a spec of what needs to be built and what features are needed. Then we could compare the amount of work required for each proposed CMS.
Ken had a start for a checklist in his mail on July 14.

My own humble proposal is ProcessWire. It is, of course, very marginal when looking at the userbase, but the advantage is that it is a very thin layer on top of PHP. So devs don't have to deal with wordpressisms or drupalisms - only phpisms

Looking at Ken's checklist, PW ticks each requirement. The only "unknown" for me is how it would integrate with phpBB in practice. Yet, "the PW API and data can be used from other PHP scripts, including command line PHP scripts" and PW devs have obviously been doing it for years (bootstrapping is the keyword when searching the forums for experiences).

If it boils down to WP vs. Drupal, I would go with Drupal as the project is at least attempting to escape the swamp of legacy code and conventions! WP has kind of painted itself in the corner by staying 5.2-compatible so long.. and of course its sheer popularity makes breaking changes hard

There probably is no comparison sheet available, it was mostly a real life discussion during our Webworld sprint some years back. If at all, there were some local documents. But again, new faces, new options

I strongly believe that we should not dismiss django (and django cms) just because it's not php.

It may prove itself better than the others.

Let's first figure out our requirements, decide what we want, and then look for a platform.

Do we have sketches of our current proposals?
Let's make some and compare them.

neverendingo wrote:@Hans:

Yes, that was the outcome of our former discussions. Performance and (like mentioned above) the people involved more used to wordpress codebase.

But no need to rely on that in this case, as the current people involved - or those interested in being involved - could have a completely different experience.

Thanks for confirming!

beluga wrote:Do you have a link to a mail with the comparison? I sifted through the July archives, but could not find anything that deep.

I seem to remember that a former sysadmin posted benchmarks on his blog, but he's retired now, his blog is down, and I couldn't find anything archived.

Note that this was a while ago, probably around 2011 when KDE WebWorld was held.

Now that CMS discussion has its own thread I would like to ask some clarification on a few points:

1. Do we have a vision given the fact that Lydia on her last post (http://blog.lydiapintscher.de) said that it is one of the things we need to focus on
2. Do we have a set of goals for the website. Are they different from those on the current website?
3. Do we have anything near a marketing plan, audience, tone of voice, things we definately need to have or things to avoid?
4. There are many ideas in what to build and what the needs are. How would we prioritise what is most important and what needs to be build?

I understand that this is an important initiative but it feels to me that it needs a bit more clarity from the board. It is after all the kde website and would need to address several needs like
- information exchange
- community support
- volunteer recruitment and user on boarding
- work showcase

What is important and what is not needs to be clearer. Not that we can't build a website without these just not as good a website as we need it to be to serve as the de facto outreach tool of the KDE community.

OK so there was some consensus that we can do content and top-down stuff in parallel.

So. Content:

Priority is this page https://www.kde.org/workspaces/plasmadesktop/

it's the highest viewed page on kde.org (besides the front page) and it shows the wrong frickin' version of Plasma.

I need:
- text
- A list of features of Plasma to show off
- good screenshots of all the good bits
- text
- making it pretty

I made an etherpad for text: https://notes.kde.org/p/workspace_website

if people like writing, please help.
screenshots to a place on share.kde.org please.

Lets roll.

david_edmundson wrote:OK so there was some consensus that we can do content and top-down stuff in parallel.

So. Content:

Priority is this page https://www.kde.org/workspaces/plasmadesktop/

it's the highest viewed page on kde.org (besides the front page) and it shows the wrong frickin' version of Plasma.

I need:
- text
- A list of features of Plasma to show off
- good screenshots of all the good bits
- text
- making it pretty

I made an etherpad for text: https://notes.kde.org/p/workspace_website

if people like writing, please help.
screenshots to a place on share.kde.org please.

Lets roll.

Alright, I think I'll move this into a new thread for content planning.

Hello everybody,

Is this project still active?

I'm a bit late but here are a few ideas I had for the website:

-merge KDE application web pages into the main webpage (already mentioned)
-use Breeze icons everywhere (application page for example)
-events like the publication of a book (!!!) or the great timeline made for KDE's 20 years need to be linked to the main page (https://20years.kde.org/book/) to be more accessible and stay there! Right know they feel "hidden" and aren't widely promoted !
-

Kver wrote:I'm thinking top-down; KDE has a lot of content rot and duplicated content... Outside of snippets I heavily question how much of the site is worth 'saving'. Also, if we go with individual pages we won't be aiming for 'big picture' changes which will give meaningful long-term results.

I think what we need to do is focus on putting systems in place over pages, and be willing to work with a clean slate. If we go top-down we *will* be dropping significant amounts of content until it's re-built, but I don't really consider that a bad thing... We still have KDE3 content on the site. It would probably be best to re-add whatever is still good than try to crop whatever is bad.

I agree with this point of view. A lot of content doesn't feels up to date anymore anyway. A fresh start could be a good choice.
-I like the feature presentation of pages like http://www.flymeos.com/functions.html#_file or http://www.apple.com/macos/sierra/ but this is a very personal preference
-like the way KDE Neon display a nice logo, we could use one of those logos for the new page (for frameworks, Plasma, etc): viewtopic.php?f=285&t=133836

For the presentation of the Plasma Workspace itself we need to focus on what makes Plasma special/good: activities, highly customizable, KDE Connectetc. Projects like Plasma Mobile could also be linked on the main page.

Kver wrote:

rcharbonneau wrote:Where on IRC can an interested party go to discuss this? I've had a pile of ideas go through my head in the last few months regarding this topic and I suppose it's fortunate that a choice has been made to make a move.

I'd highly encourage you to simply post them a thread here! It will be easier to reference ideas and conversations easily as time goes on if we keep brainstorming and notes in the forum.

If WordPress is still being considered, I'd love to help. I think WP would be a good choice running in multisite mode. The present subdomain setup would still work and each subsite can have it's own theme(although naturally, similar). Each subsite can have it's own custom post types to make it easy for devs to update the site while keeping a standardized layout. As far as single sign-on, there's a couple of phpbb/wp bridge plugins but I'm not well versed in that area and I don't know what other systems you'd want to be included. Each subdomain does not have to be running wordpress, so the existing buds.kde.org wouldn't be affected however, I don't know how single sign-on would be implemented or if that would be desired on that system.

If we're not using wordpress, I'd still like to help. Content writing, css/html, general organizational chores etc.