Which CMS for the website?

[RYX]

Since it has been discussed for a while and everyone has agreed to make a poll, here it is. The available ones are only a quick selection of the most popular CMSes I know - if you want your personal CMS of choice to be added here please give a note

[mikedee]

Shouldn't you include wiki in that? ;)

[RYX]

I thought the general opinion was to to keep the wiki and the forum and integrate both into/with the CMS-based website (maybe I misunderstood that?). I never questioned the use of a wiki for the documentation, but I don't think a wiki is a CMS. Anyway, if you want a wiki as choice, I'll add it ...
:D

[mikedee]

I thought most people thought wiki was a good choice, but some people had major problems with wiki.

As I said before I cannot make much of a choice because I do not know the pros and cons of each. Personally I would rather stay with wiki and work out the problems that people have with it.

If wikicode is too horrible then people can write it in the forum and I can move it over. The wiki cheatsheet is very useful plus you can still break into html if you need to. The wiki was designed so that even people who do not know html can add basic formatting. It looks wierd to us but to some people it might make sense in an ascii art kind of way.

The translation problems seem easy to overcome, and since we are talking about content, most of it will need to be rewritten for each language. We just need to agree to keep the english page as the main one and then each language can be updated to match. Wiki allows easy viewing of modifications so it gets my vote there (dont know about any of the other choices).

[nzjrs]

I would rather stick with a wiki

Positives

Already have a nice theme
Matches other FOSS/Novell project wikis (F-spot, Beagle, Compiz, Hula, etc)
Of all the Wikis, mediawiki has the most use and familiar syntax
Good versioning and ability to rollback to previous edits


I just dont see that changing to a CMS would buy us anything. For every person that is unfamiliar with a wiki there is one that is unfamiliar with a CMS.

Remember that the cost of transitioning to a CMS must be considered in the evaluation. IF a CMS is chosen it must provide sufficient benefit to justify the change.

I would argue that
time-to-learn-mediawiki-syntax< time-to-change-to-a-cms
and that a CMS does not provide sufficient benefits to overcome this

[RYX]

From a web-developer's point of view, a wiki is no real website - it is a documentation-generator and nothing more. Things like a theme-database or plugin-download-section are either impossible or ten times more work when using a wiki. We shouldn't talk about replacing the wiki, but replacing the unprofessional and scattered look&feel of the site. Also the forum should be integrated into the site, not simply get linked to ... which is also hard when using a wiki for the entire page.

Benefits of a CMS:
- dynamic content management
- professional and consistent look&feel
- great extensibility
- far more possibilites compared to a wiki
- ability to easily create a theme/plugin download-section
- wiki can still be integrated as module
- forum can be wrapped into the main site giving one look for all
- NOT doing it the usual (and IMO unprofessional) Novell-way
- NOT having such a boring and childish look as most other FOSS-projects
- NOT scattering information and ressources around the net, but keeping everything in one place (=user-friendly)

Of course this is only my view of that, but I would offer to realize it - so it would be my work. Whoever wants to focus on wiki and documentation could continue doing that. Whoever wants to help me in creating the kick-ass website which a kick-ass app like compiz deserves, let's go ...

:)

[RYX]

I would argue that
time-to-learn-mediawiki-syntax< time-to-change-to-a-cms
and that a CMS does not provide sufficient benefits to overcome this
The problem is that I, personally, don't want to learn that (stupid) wiki-syntax. Maybe setting up the CMS will take some hours, creating the design will take some more hours - but the experience you get is always worth it.

[Ballsacian1]

The problem you get when using a wiki as posted above, is the ability to extend it. Exponent is a very powerful CMS that is very easy to extend and comes with a good amount of modules. Theming Exponent is the easiest part as once you have an xhtml compliant design it takes literally minutes to transfer it into Exponent.

Exponent has advanced permissions including individual, group and workflow policies. Exponent also can keep an infinite amount of revisions so rolling back changes is effortless. Add that to its built in forum module and galleries and you have a top notch system.

Although Exponent isn't as developed as some CMS's mostly Joomla, it is a far superior design as it completely separates structure from styling. I've been spending the last 6 months learning about this system and working very closely with its development team and feel that it would be the system to choose.

FYI: I am Imnotpc's son and would be willing to contribute my time in helping make Exponent a success on this website.

As the months go by this system is just going to get better and better as me and the rest of the development team are pushing out features as fast as we can think of them. For me Exponent was a no-brainer.

I've seen what Compiz can do and believe that it deserves just as stunning a website to show of its power to the world. The beautiful thing about Exponent is that you can literally make the design of the site whatever you want. So no two Exponent sites look the same.... unlike Joomla and many other CMSs

[nzjrs]

I just added some help docs (http://www.go-compiz.org/index.php?title=Help:Editing) to the wiki to make it easier for new editors.

I also created this forum post (http://forum.go-compiz.org/viewtopic.php?t=148&start=0&postdays=0&postorder=asc&highlight=) asking for screenshots. This will be a good experiment, to see if a CMS is needed or if 95% of the use-cases of a CMS can be implemented with 7% of the effort using mediawiki in 4.6% of the time[1]

[1]numbers made up

[RYX]

Hi Ballsacian1! I totally agree with what you say about CMS compared to a wiki. Exponent has a good looking site and seems quite useful (that's what I can say from the first look). And maybe you're right that the current Joomla is crap (that's why I added 1.5 to the choices - it will be much better). We should let the poll decide about that.

Nzjrs, thanks for updating the wiki. You're right - it looks better now and it can be themed and beautified. One big problem still remains: How do we visually integrate forum, wiki and upcoming parts like a theme-database or a plugin-section. The current state of the site(s) is horribly scattered and inconsistent - on the long run, a CMS is the only useful solution for that. So why should we duplicate effort and not switch to a CMS right now instead of in approx. 6 months? I don't want to question your knowledge, but I have been creating websites for the last 8 years now and I think I know quite a bit about that stuff.

Don't take the ugly and foolish sites as an example, look at blender.org or joomla.org - there you can see the dimensions of a project grown on a good website.

:)

[amgeex]

I agree with mikedee and RYX here. My vote, however, goes to drupal. I've found that this is one of the most flexible cms systems out there, and I think is far more beautiful than joomla. It has forum module, and a module that can import all the data from a phpbb forum to the drupal forum module. About the wiki integration, I'm not sure if there is a module, but I will answer that in my next post. 8)

[imnotpc]

Looking at the poll results so far the sentiment is running 3 to 1 in favor of CMS over wiki and I don't see any indication that's going to change. However the votes between the CMSs are close enough that the wiki people could alter the CMS choice. Do any of you that voted for none/wiki wish to express a CMS preference before voting closes?

[mikedee]

Can I maybe make the suggestion that for the next 6 months we use wiki on a trial basis and concentrate on actually adding content, howtos etc.

If the wiki problems are too great then we can look at joomla 1.5 which should be out of beta. I think that the wiki has all we need though, the cms systems seem like overkill for what will hopefully be mostly static content.

The wiki is already looking much much better and we can hopefully improve it with time, what I think causes the most problems for potential users is the lack of documentation on how to install and run compiz. If we can create all of that in an easy to read and use format then we can look at translating it and bringing the sites together design wise.

[nzjrs]

Can I maybe make the suggestion that for the next 6 months we use wiki on a trial basis and concentrate on actually adding content, howtos etc.

If the wiki problems are too great then we can look at joomla 1.5 which should be out of beta. I think that the wiki has all we need though, the cms systems seem like overkill for what will hopefully be mostly static content.

The wiki is already looking much much better and we can hopefully improve it with time, what I think causes the most problems for potential users is the lack of documentation on how to install and run compiz. If we can create all of that in an easy to read and use format then we can look at translating it and bringing the sites together design wise.

+1 Agree completely

I also followed the discussion of which CMS to use for the new GNOME website.

After the initial vote, each CMS was evaluated and a demo site was setup so that people could test each CMS.

So do those in favour of the top 2/3 CMS want to set up demo sites so that their capabilities may be compared with that of the current wiki?

[RYX]

Maybe we can build two teams - one with focus on design and one with focus on documentation. The design-team could work out a CMS-template and -layout while the doc-team extends the look of the documentation in the wiki. Once the cms-design is ready we can seamlessly switch to the new system with integrated wiki and forum, while keeping all existing wiki-data (except the start-page).

I have to work with joomla, anyway. It is no problem for me to invest some time and work into learning and trying - practical experience is always good. I would even like to code new modules or components (or modify/hack existing ones), if needed.

Edit: Nzjrs, maybe I repeat myself but using a CMS would not mean replacing the wiki. It should be seen as an addition to the current state, which extends our future possibilities ...

:)

[amgeex]

Well, after reading mikedee's last post I volunteer to go into the documentation team, although I can help on the cms/website team too. I think this is a very appropriate move, separating the development of the site into two teams, so an even amount of documentation and work can go into the wiki while the cms/website is being worked on. And after the cms is ready, the wiki can just be merged into it and all our work will still be there. 8)

Sounds good to me! :D

[imnotpc]

nzjrs, ask and you shall receive. Here are default installations of each of the top 3 CMSs in the poll:

http://www.ubaight.com/exponent/
Username: compizadmin
Password: password

http://www.ubaight.com/joomla/
Username: admin
Password: password

http://www.ubaight.com/drupal/
Username: compizadmin
Password: password

Note that I haven't done any configuration on these whatsoever. These accounts should give you full admin access. If you need access beyond this contact me directly. I'd recommend the admins who are familiar with each one set them up to show the features people are interested in.

A couple interesting, but not particularly relevant, things I learned while installing these:

Each of them needed you to create a database and give them a username and password.

Exponent installed cleanly in about 8 minutes.

Joomla is a big package. With a 2.8MB tarball it's twice the size of Exponent. Setup took 20 minutes, plus another 20 minutes to discover the beta package tries to install a non-functional package and a new feature needs to be turned off in order to use the administrative interface. According to the site this has been fixed in more recent versions.

Drupal is a very small package at about 500KB. The installation is mostly manual and there isn't much to the default setup. Took about 22 minutes to install but now that I've done it once it could go much faster.

Happy hacking!
Jeff

[nzjrs]

nzjrs, ask and you shall receive. Here are default installations of each of the top 3 CMSs in the poll:

Wow thats fast! You should lend a hand to the upstream new GNOME website effort (if you have time). See http://live.gnome.org/GnomeWeb/WgoGetInvolved

[mikedee]

Here are my initial 5 min overview

http://www.ubaight.com/exponent/

broken?

http://www.ubaight.com/joomla/

Seems very complicated, not very intuitive (i spend 30 secs looking for the edit button). From what i hear, it is not easy to customize.

http://www.ubaight.com/drupal/

Very simple, has forum and blog but i prefer dedicated software myself rather than one does it all. This doesnt seem to have many benefits over wiki (looked to have disadvantages). The menu option under input format was VERY worrying.

PHP code

* You may post PHP code. You should include <?php ?> tags.

Hmmmm :shock:


What exactly were the major problems with wiki again. After a very brief look at these I would recommend wiki even more really. Especially until we really know what the limitations are.

[amgeex]

Well, as main supporter of Drupal around here, I've tweaked it a bit. That's why it is sporting the minimal look , I changed the default theme. I also modified the menus and setup some demo forums and posts. Also, the main page is a blog, this so the news can be on the front page. All of this was done with minimal effort, and it is very easy to move around the site once you've become familiarized with it.

About the php in the input formats, it has been removed, as we don't want script kiddies hacking away our site. And we need to install a wysiwyg editor asap. Here's a screenshot of TinyMCE:

http://img295.imageshack.us/img295/9742/wysiwygab4.png (http://imageshack.us)

[imnotpc]

OK, I had to tweak my php.ini file to get them all to play nice. They should all be working now. Let me know if I broke something else.

[imnotpc]

And we need to install a wysiwyg editor asap. Here's a screenshot of TinyMCE:


OK, It looks like it's installed now.

[imnotpc]

Wow thats fast! You should lend a hand to the upstream new GNOME website effort (if you have time).
Yeah, that's the rub. <grin> I've got my own decent sized company to run, another open source project I do some documentation for (XMail), an in-house computer aided dispatching application I wrote that I'm trying to deploy, a robotic vehicle patent application I need to complete , and a family (wife, kids 20, 19, 12 , 8 some of which like to see me once in awhile). I really shouldn't even be here, but this stuff is just so cool I can't help myself!

[amgeex]

Hey, thanks man! I really appreciate this!

[RYX]

Thanks for the nice examples. Trying to be objective, I have to agree that exponent seems to be the simplest of all those. Drupal is waay too chaotic and unstructured for my taste (and looks rather ugly IMO). Maybe it is fair to also post the Administrator-link for joomla - because joomla has a frontend-admin and a more feature-filled admin-backend.

The "real" Joomla-admin area can be found here: http://www.ubaight.com/joomla/administrator/

Now please compare again ;)

Edit: Maybe Joomla has a strong learning-curve, but do you prefer Microsoft's Paint over Adobe's Photoshop, just because Paint is much easier to understand? (Yes, stupid question - of course you use gimp)

:)

[imnotpc]

In the spirit of fairness I tried to install mediawiki so the wiki enthusiasts would have a demo site with the same access and freedom as the CMS sites. Unfortunately the last couple versions of mediawiki require php5. My old SuSE 9.1 server doesn't have it and I'm not comfortable making a major upgrade like that on a running server. Sorry... I tried.

Jeff

[mikedee]

I have given them all a quick test drive and tried to copy the compiz homepage to each of the sites. Here are my results.

Exponent - This had my vote up until this point, I couldnt insert a table so the home page is just one long page. Even if you can some how replicate this, it didnt seem straightforward.

Joomla - After getting to the correct admin page it seemed a bit overcomplicated and not so user friendly. For example, I saved my home page and then it dumped me back on the edit pages and I couldnt see the page I just saved.

Drupal - Seemed to have the same html entry field except in an even smaller teeny tiny box to edit with.

Mediawiki - This seemed to accomplish the task with the least fuss and time. I didnt do it but I know it would have been fairly simple to do.

All of the wysiwyg editors had problems with adding and removing whitespace changing the formatting. Drupal didnt seem to have a h1 button to add it back in.

[Ballsacian1]

Actually Exponent should be very easy to change the default look.. The issue is you need to have access to the folder to upload a new theme index file. It accepts standard html/xhtml compliant code. After created the template you want you just need to add a few lines of code dictating where container modules go.. these modules are what hold everything else in the page. By sunday night the www.ubaight.com/exponent/ page should have a general layout done... I work a lot on the weekends so I don't have much time right now to do it.

Also my dad explained to me what your problem was with Exponent. I didn't realize they changed the default toolbar for html area but i have changed it back using the toolbar config page to include all choices in a default install. The newer version that I am going to install on this site on sunday include Xinha FCKEditor and a few others to choose from as htmlarea is kinda outdated.

I've also set it up so anyone that goes to the "wiki" page and registers a name can now add edit and delete content on it. Exponent also has unlimited revisions so it can still serve this purpose. You can also combine a wiki style with a work flow policy so that everyone can edit but after editing a person with special rights has to approve the edit before it shows.

[amgeex]

So yeah, Drupal is kind of ugly... Joomla has a nice administration panel, but not much more. Exponent on the other hand seems to be the only cms that achieved a nice resemblance to the actual homepage.

I guess that after this round of comparisons, my vote goes to MediaWiki, definitely. You can do more with less, and mikedee and I have posted syntax guides quite some time ago. If you had read them you'd know how to work the wiki now, lol.

The wiki is just more flexible, in my opinion. You can do tables, pictures, links, lists, bullets, whatever you want, and you don't have to be worrying about moving blocks, setting up permissions, activating or installing modules, etc. My point is: the wiki is fine, it works, and its simple. 8)

[RYX]

Well, if simplicity is more important to everyone than extensibility and professional features - we should use a wiki. With a wiki we are limited in many ways, that's why I intentionally didn't add the wiki as an option to the poll. I still say that a wiki is no cms - instead a good cms enables the creator to add a wiki to his website (to offer documentation).

I don't want to enforce using a cms (maybe you got the impression), I only would like to have a professional base for future extensibility of the compiz-website (what a wiki definetly not is). Your beloved wiki would stay intact, even if we would include it as a module to the cms - the only change would be that the wiki will be used _only_ for documentation, not for the overall site management.

(And Joomla has far more than only a nice admin-panel, but it takes some days to learn and understand)

:)

[mikedee]

Which features are we missing exactly?, a wiki is good for long term documentation etc but it is not good for news/blog features. Really we do not have massive needs, but it would help if you could point out how your life would be harder we can find a solution.

Maybe we could look around for some nice blog software. If it is easy to customize (a major plus of wiki) then we could have all 3 sites running alongside each other looking like 1 site, plus we will have the best software for each individual task that are all easy to maintain independently.

[amgeex]

Serendipity is a great blogging package. It's quite easy to set up, and has tons of features and it uses the Smarty template engine, so making themes isn't too complicated.

Check it out: http://www.s9y.org

[RYX]

I didn't say I need a blog, but Joomla has an embedded blogging-feature for that purpose (and ten other ways to publish news). The things I would like to have are:

- logically structred website
- well designed website with consistent look&feel
- one backend for administration for all parts of the website
- user-friendly download-section with various plugins, not packaged (also possibility to add 3rd-party plugins)
- theme-section once another decorator is there (there is more than metacity)
- screenshot-/video-section to demonstrate the power of compiz
- all above points available without the need to write a single line of html/wiki/*-code for end-user, editor and admin once the site/design is ready

I really doubt the use of having another site to manage by separating all those parts (wiki/forum/blog/...). Webdesign is not Unix-development. There is no use in splitting things into small parts, it only causes trouble - CMSes were invented to solve exactly that issue of "historical" web-development. What's bad in having unlimited extensibility under one central management with one consistent look?

:)

[amgeex]

Alright, but then we should decide on what CMS to use. I don't like Joomla that much, but I've got experience using it, so I don't mind it. Exponent, I really don't know much about it, but I'm willing to learn if there are docs around. I still think that a cms is overkill for our needs. :shock:

[RYX]

I am currently trying to "pimp" the Joomla-example. I am not really aware of the new template-syntax (xhtml instead of html/php), so I don't do much on the template.

[RYX]

After playing around with Exponent, I realized that Exponent is the better CMS and very similar to my own, unfinished CMS I wrote some months ago (my CMS is strictly OOP, follows Eclipse's "everything-is-a-plugin"-principle, grew to 1mb of code-size and 100+ classes and I freezed it because I got too confused).But I still think Joomla has better support and is developed and reviewed by more people and fits best for our purpose. Though Joomla 1.5 has not many available modules yet and we would have to wait a while, anyway.

@amgeex: Where do you see the overkill in using a cms? We do not necessarily have to put in all features we can find (like other cms-based sites tend to do).

:)

[RYX]

I spent some hours on learning the new features and changes in Joomla 1.5 and modified the Joomla-demo a little bit. It is a first approach for what I would call a good website (the design is ugly right now). Check it here ... (http://www.ubaight.com/joomla/)

:)

[imnotpc]

RYX, can you setup the Joomla forum module? I've never used it and I'd like to check it out.

[mikedee]

Here is my quick attempt at getting the forum and wiki to look the same, aside from the menus and slight main page formatting differences you would not know they are on totally different sites.

http://www.anykeysoftware.co.uk/wiki/index.php/Main_Page

I didnt realise that we were wanting to replace the forum at the same time.. Is this what you are planning?

[nzjrs]

nice.

For the record I actually like the wiki color scheme so would not mind if the forum was skinned to look more like the wiki.

but unification is good either way

John

[RYX]

RYX, can you setup the Joomla forum module? I've never used it and I'd like to check it out.
Yes. If I can find a forum module which is compatible with 1.5 version ... I will look into it this evening (so in about 6-8 hours).

I didnt realise that we were wanting to replace the forum at the same time.. Is this what you are planning?
Actually I didn't plan that - instead I'd like to keep the forum with all entries. Since phpBB is one of the most popular forum-softwares out there, I am pretty sure there is some bridge-module for allowing easy transitioning to either integrate with the site or move to the SMF (which is afaik the best forum-module for joomla) ... we should wait with any decision on that until the cms-question is answered. Using an integrated forum would result in having one user-account for all parts of the site ... which is good for not so advanced users.

We still have a draw ... did anyone check out the Joomla-demo now? The menu is crap and the design looks like there is no design (Its highly stripped down) ... but the stucture is better (imo). And no need to code html/wiki-syntax, everything can be easily added and is right where it belongs (i.e. where the designer wants to have it).

:)

[imnotpc]

I didnt realise that we were wanting to replace the forum at the same time.. Is this what you are planning?

Ideally it would be nice to have a single CMS which has all the modules we need. In practice, independent programs like the forum often are superior to the comparable CMS modules. That's why I really encourage the proponents of each system to "show their stuff" so the community can decide which combination best suites us. The top 3 CMSs each have a forum module and I'd like to see how they compare to what we have.

From what I've seen so far the drupal forum is nice but not as full featured as current one (phpbb?). I haven't installed the exponent one yet but judging from the forum on the exponent site I expect it to be about the same as drupal.

If you have a chance to go to the exponent demo site (before Ballsacian1 replaces it) you can click on forum and get an idea how integrating the current forum into exponent would work. Joomla has a similar wrapper that you can use for inserting html pages directly into the template. I assume that you could wrap the entire forum as well but I haven't tried it. I don't know if drupal has this feature or not.

[amgeex]

Drupal has a module that imports all the data from the phpbb forum to its own forum module. If it could be installed it'd be great, so we can try it out, see if its simple to use, etc. :wink:

Here it is: http://drupal.org/project/phpbb2drupal

[imnotpc]

OK, when I get home tonight I'll see about installing it.

[mikedee]

- logically structred website

This is everyones goal and it is possible by using different software for forum/blog/documentation

- well designed website with consistent look&feel

My wiki is the only one so far that has a consistant look and feel.

- one backend for administration for all parts of the website

I do not think that even Joomla offers this, the forum modules all looked like hacks which just registered a joomla account whenevera forum account was registered.

- user-friendly download-section with various plugins, not packaged (also possibility to add 3rd-party plugins)

I think that allowing people to post random code in an official repository is asking for trouble. We need to check the code here before putting it anywhere official. Wiki allows file uploads, we could easilly write a custom form which posts to the existing wiki upload page.

- theme-section once another decorator is there (there is more than metacity)

Same as above, we will only get 1000 vista clones otherwise ;) I thought that some sort of web service that linked to the theme configuration tool would be better for something like this, we would still have to choose a sensible range of themes to distribute. This is all acedemic at the moment since we are quite a few months away from this.

- screenshot-/video-section to demonstrate the power of compiz

There is already one on the wiki. The videos are taken from youtube so that we do not have bandwidth issues (we are not funded so we cannot afford to be streaming 50Mb video files all day)

- all above points available without the need to write a single line of html/wiki/*-code for end-user, editor and admin once the site/design is ready

The problem with editors that do not require a line of html is that they are very inflexible. Look at your home page, you should be able to easilly copy the existing wiki page, but that seems to be taking quite a few hours.


I have to say that using one piece of software to do everything we will ever want is a very very bad idea for these reasons

- The options shown so far lack flexibility and cannot replicate the design of the existing forum easilly at all. This means if we change the design there will be a lot of work needed.
- All our eggs in one basket. There is a reason why that is a common saying. We do not want everything to go down if there is a joomla problem. If RYX is not around when we have this problem we are in a lot of trouble.
- Doesnt scale well. Hopefully in the future we will be more popular. If all our software is tied to the same database and php code then we will not be able to split each service onto its own server, we will have to scale the database which is always much harder than scaling independant pieces.


I strongly feel that the pros of switching to joomla (all admin in one place, everyone who registers for forum gets a wiki account). Are far far outweighed by the disadvantages (all mentioned above + the task of migrating everything, training everyone to use and customize joomla).

I think we need to concentrate on working out a site structure and design, then actually adding the content. If adding to a wiki is too hard then add it to this section of the forum and I will personally move it to the wiki.

We are forgetting there is 1 extra vote which is not counted, Davids. He originally voted for wiki so if we add his then wiki wins the poll. Even if we did not count his vote then there is not enough opinion to change, we would stay with the status quo.

[amgeex]

As I said, I still think a CMS is overkill for what we need. I propose that the frontpage of the site should be a blog, so posting news and updates would be easy as pie. It should be easy to customize and to maintain. On the sidebar there would be links to both the current forums and to the wiki. Also, some blog software allow the creation of static pages, which could prove useful to showcase the youtube videos, etc. I fully back up mikedee on this issue, as I've said throughout this thread. I too will help to move things to the wiki if you don't want to, or can't, do it.

Still, I could help out with Joomla, if that's the way you guys choose to go, but I do prefer separate pieces of software doing their own thing. You cannot compare the flexibility and power that wordpress or serendipity give you to the blog that Joomla offers, or the forum modules of that CMS to phpbb, which is far more powerful.

We just need to decide on this and then move on to customize the different parts of the site. Again, I see that the forum already sports a custom look, so that's one less thing we have to work on. And mikedee already started work on the wiki, its just a matter of porting it to the official wiki. Now, if you think about it, all that's missing is the proper homepage, the blog. That should be very easy to customize and setup. I myself volunteer to move stuff to the wiki, help out on the forum, and maintain the blog.

[imnotpc]

My wiki is the only one so far that has a consistant look and feel.


What do you see inconsistent on any of the CMSs?

RYX wrote:
- all above points available without the need to write a single line of html/wiki/*-code for end-user, editor and admin once the site/design is ready


The problem with editors that do not require a line of html is that they are very inflexible. Look at your home page, you should be able to easilly copy the existing wiki page, but that seems to be taking quite a few hours.


Once all the editor features were enabled this was not difficult with exponent and took me a couple minutes. Requiring people to learn wiki markup now isn't a big deal since most everyone here is a developer or experienced tech person. But as compiz becomes mainstream, the majority of people using (and hopefully contributing to) our site is going to shift to less knowledgeable users. Having a user friendly plain text editor will become important.

The options shown so far lack flexibility and cannot replicate the design of the existing forum easilly at all. This means if we change the design there will be a lot of work needed.

I've not tried to duplicate the current patchwork of different themes on the exponent site. I think at this stage it's important to focus on functionality and usability. I could drop the current forum header onto the top of the exponent site if you'd like to see how it looks. I had assumed that we were going to spend some time improving the site no matter which system we use.

- All our eggs in one basket. There is a reason why that is a common saying. We do not want everything to go down if there is a joomla problem. If RYX is not around when we have this problem we are in a lot of trouble.

There are several of us that can manage both Joomla and exponent. I'm not sure about drupal. A CMS is essentially static html and php code (some use javascript) which either works or doesn't. Site problems are usually related to the web server or the backend database and those need to be addressed by the site admin.

- Doesnt scale well. Hopefully in the future we will be more popular. If all our software is tied to the same database and php code then we will not be able to split each service onto its own server, we will have to scale the database which is always much harder than scaling independant pieces.

I'd like the site to be popular enough to have this problem! I think we have plenty of headroom with any of our options and we couldn't afford the bandwidth to overwhelm the CMS and certainly not a mysql database.

I strongly feel that the pros of switching to joomla (all admin in one place, everyone who registers for forum gets a wiki account). Are far far outweighed by the disadvantages (all mentioned above + the task of migrating everything, training everyone to use and customize joomla).

We have so little content that migrating is a non-issue at this point. Training everyone to customize Joomla would be a big issue but I don't know why we need to do that when you can edit right on the page to add content. Same with exponent. With granular editing rights and plain text editor there's less of a learning curve than with a wiki.

Jeff

[imnotpc]

I think we're all starting to repeat ourselves here. Let's have a new poll with only the choice of CMS or wiki. If CMS wins we can have another poll to select which CMS we want. David clearly stated he wanted the community to decide, so let's make the decision and move on.

Jeff

[amgeex]

Just check this site out. It's more of less an example of what I had in mind. Its a wordpress blog on the frontpage with links to the forums (foros) and to the wiki. It's in spanish, but just for you to get the idea.

http://www.archlinux.com.ar/

[mikedee]

Right...

So the solution to losing the 'which cms should we use' poll is to have another poll :D

Am I losing it totally?

[mikedee]

Sorry, I am losing the will to live here so I will only address a few of your points

Once all the editor features were enabled this was not difficult with exponent and took me a couple minutes. Requiring people to learn wiki markup now isn't a big deal since most everyone here is a developer or experienced tech person. But as compiz becomes mainstream, the majority of people using (and hopefully contributing to) our site is going to shift to less knowledgeable users. Having a user friendly plain text editor will become important.

How can less knowledgeable users contribute to documentation etc? They can contribute to the forum easilly enough. Also Wikipedia has millions of non technical users who contribute daily.

I'd like the site to be popular enough to have this problem! I think we have plenty of headroom with any of our options and we couldn't afford the bandwidth to overwhelm the CMS and certainly not a mysql database.

Ill answer with a formula (including a smiley!)

compiz v1.0 release + slashdot + digg = toasted server + :oops:

[imnotpc]

There are twice as many votes for using a CMS as there were for using a wiki. I don't see that as losing. Besides, the original poll was to select a CMS not to decide if we should have one. The wiki option was added at your request. However in light of the discussion on the forum it's clear that we don't have a consensus and we need to take a step back and see if we are going to use a CMS at all.

compiz v1.0 release + slashdot + digg = toasted server + Embarassed
That's a formula for disaster no matter what software we run! :lol:

[RYX]

The problem with editors that do not require a line of html is that they are very inflexible. Look at your home page, you should be able to easilly copy the existing wiki page, but that seems to be taking quite a few hours.
There is always an option to code in HTML if you want, but a well designed website and template don't require the publisher/admin to write a single line of code. I never wanted to repeat your wiki-design. I wanted to create the base for a "real" webdesign. If you want to use a cms you have to do things the "cms-way", not the "wiki-way". So it is quite stupid to try repeating the wiki-look in a cms... there are better ways.

But I think I am fighting windmills here ...

Edit: And I agree that we should have a poll "CMS or Wiki?" because (repeating my self again) a wiki is not comparable with a real cms ... we are running in circles atm.

:?

[mikedee]

compiz v1.0 release + slashdot + digg = toasted server + Embarassed
That's a formula for disaster no matter what software we run! :lol:

This is not true, wiki was written from the ground up to support very high loads. If we are expecting high traffic we can move the wiki (ie. www) to another server so that it takes the full force of it, the forum will be totally unaffected. Mediawiki supports many levels of caching and has been designed to use those caching methods to their full potential. It can also be easilly copied onto another box and added to a cluster if traffic gets really high.

I would recommend lighttpd + php-fast-cgi for performance, they out performed apache + mod_php by a long way when I last tested them. With memcached used at the application level and mmcache at the php level you can support massive loads with no hits on the database whatsoever and very little cpu time used. There is also no disk thrashing which is a major bottleneck.

If we have lots of people visit and 20% decide to post on the forum, using an all-in-one solution we would lose the database and therefore everything else.

[RYX]

PhpBB is the same performance-monster as any cms ... do you want to switch to another forum-software because of that? Fast and reliable servers get cheaper and cheaper so the better performance of a wiki is a very weak argument ...

[mikedee]

Edit: And I agree that we should have a poll "CMS or Wiki?" because (repeating my self again) a wiki is not comparable with a real cms ... we are running in circles atm.

Why dont you have a poll which would be

'IF we were to replace the wiki/forum with a cms, what would it be?'

Then we need to see that the cms is able to be branded exactly as we want (your joomla example looks like all other joomla sites at the moment). My example was proof that we can still have a consistant site without using one piece of software for everything, it took less than 1 hour including installing and customizing (including 20mins wasted on the old template).

Once we can see the limitations (or lack of) then we are able to make a decision that it is worth all the work moving the wiki and forum over.

Then the poll can be

'Should we change the forum/wiki to x or stay as we are with a seperate forum/wiki/blog?'

With a link to each finished site branded with the new design (which doesnt exist yet).

Either way, we cannot rush to do anything for at least 3 months, Joomla is still not ready, we do not have a final design and there is no content. I think we really need to concentrate on how the site will be layed out. Adding consistaancy will be very easy.

[mikedee]

PhpBB is the same performance-monster as any cms ... do you want to switch to another forum-software because of that? Fast and reliable servers get cheaper and cheaper so the better performance of a wiki is a very weak argument ...

Individual forum posts do not normally have very very high traffic, if a post is very important or useful then it makes sense to move it to the wiki. Most external news sites will link to either the home page or a release announcement, or both. If the forum goes down it is not a major problem, people will still be able to download things and get the basic install instructions, if the whole site drops suddenly then it will not be nice.

[RYX]

Of course the design of the Joomla-site looks like all other Joomla-sites, because I only modified some parts of the template and stylesheet to be a simple base... (and I already mentioned that).

Mikedee, I don't understand why you are heavily trying to let the wiki "win", while (as already noted by imnotpc) it is obvious that the great majority wants a cms and no wiki ... Sorry, I have nearly 8 years of experience in PHP-coding and web-development so maybe I have a too professional attitude on this but I am used to take the best solution, not the easiest (and I actually thought that type of attitude caused you to move from beryl to compiz so I cannot really understand your point of view in this case).

[imnotpc]

I also don't understand why there is so much resistance here. We have 3-4 people with CMS experience willing to work on a site that had been virtually ignored for a month. And no matter how many choices we give you it's never good enough. Rather than accept our help you and nzjrs spent a substantial amount of effort modifying the wiki to show you don't need us. Why? If you don't want my help just come out and say it and save us both a bunch of effort. <sigh>

[mikedee]

I think the point is that I didnt spend a substantial amount of time modifying the wiki. RYX and you are all spending a lot of time modifying your cms's to be what we want, but none look anything like that yet.

RYX - I am really sorry to point this out, but I do out-rank you in web design experience. I really really do not want to go down the 'I am a professional' road if possible, thanks ;)

Having said that - I have totally lost all interest, I thought you would all want to spend more time on actually creating contect rather than talking about the container for it. I have tried my best and I really dont care either way now, spend from now until after Christmas working on a replacement.

[mikedee]

Why? If you don't want my help just come out and say it and save us both a bunch of effort. <sigh>

Sorry but I have experience in building very large and maintainable websites, I really do not need help. I cannot help but to let you in on some of my experience, you choose not to accept it - thats fine.

You do not need my help to rip everything apart and start again, so why bother that I do not agree, just get the server password and install your favourite cms/wiki/blog/news feed/template upload/screenshot/video upload site. Then just blame the 'server admins' if things get a bit slow and unscalable.

[imnotpc]

imnotpc wrote:
Why? If you don't want my help just come out and say it and save us both a bunch of effort. <sigh>


Sorry but I have experience in building very large and maintainable websites, I really do not need help.
Ok, I'll work on something else.

[amgeex]

Either way, we cannot rush to do anything for at least 3 months, Joomla is still not ready, we do not have a final design and there is no content. I think we really need to concentrate on how the site will be layed out. Adding consistaancy will be very easy.
mikedee is right on this one guys, which brings me to the next point. As I said before, how about if we make two teams, one that focuses on the new design and that would work on the cms of our choice, and a team that works on documentation and content. That content could be added or 'stored', if you will, on the wiki. When the new website/cms is ready, we could move the content there, or keep it in the wiki, or both. How about that? I really think that this shouldn't be such a big problem, I mean, we're about to lose mikedee here, and he's a great contributor. I think that his work is more important than a cms over wiki disussion.

[RYX]

Yes, amgeex - you're right. I already agreed to make teams and already proposed to make a 3-6 months plan so no need to hurry. I also think that the current wiki is ok for now, all I wanted to do is getting all things together.

Mikedee, from a certain point, professionality becomes a matter of taste and style. So I didn't want to point out my "ultimate" experience, but wanted to note that I maybe am a little overmotivated concerning the overall style of the website. Since you are so much more epxerienced than I am, we will both agree that a wiki is still no cms ... so basically the whole comparison wiki vs. cms is pointless.
;)

And FYI: I installed the Joomlaboard-component into the Joomla-demo. It is only meant to test the component, no need to panic.
:)

[amgeex]

I would like to see mikedee's response. And who exactly are members of what team? We need to define who goes into the cms/website team and who goes into the documentation/wiki team. 8)

[RYX]

Maybe we should just leave things as they are. If zootreeves installs the Attachment-mod for the forum, we could have a screenshot/plugin/theme/whatever download-section as separate forums (as previously on old beryl/compiz forum) and the wiki can become (or stay) the official page for compiz and source of documentation.

My approach on unifying the various parts wasn't intended to cause any trouble (but I somehow knew it would) and, honestly, I have better things to do than creating a professional website for a Novell-project without getting anything for it ... I would rather do it for Linux and all newcomers to it (and people that hate the scattered and childish presentation of the huge majority of open-source software).

Edit: And I think I don't have to mention which team I would join ... :D

:)

[amgeex]

Edit: And I think I don't have to mention which team I would join ...

No, you don't, but the rest of the people here do. :roll: And well, it did cause trouble, but its pointless pursue the argument now. We should start focusing on adding content to the wiki. The first candidiate: How to run Compiz with XGL (anyone got info on that?)

[imnotpc]

I think this discussion has drained the life force out of all of us. So how about a new proposal:

1. Since we can't even agree on how to interpret the results of the poll, let's just call the results inconclusive.

2. Let's have a logo design "contest" and select a new logo before we start getting nastigrams from Cingular.

3. After the logo is selected, anyone who wishes to will have a month to develop a prototype site incorporating the new logo and using whatever tools they wish.

4. After a brief period to evaluate and discuss each prototype we'll have another poll to determine the future site design.

5. All polls will require a majority to win. If there is no majority, the lowest vote getter will be dropped and another poll taken until one design gets a majority. Each poll will last 3 days.

6. During the process we will continue to work on documentation and content for the current site which can also be used in the prototype sites.

7. We'll all sit around a campfire and sing Kumbaya until we get all this aggression out of our systems.

8. Forget item 7. We'll just sit around a campfire and toss back brewskis.

9. OK, we can skip the brewskis... but I feel better just thinking about it.
Jeff

[amgeex]

I agree with imnotpc. Let the wiki fill with docs!!! :P

[RYX]

Thanks imnotpc, I fully agree with your "guidelines". (And sorry if I sounded agressive, that wasn't my intention ... I will try to focus my energy on creating a logo)

:)

[imnotpc]

I wasn't referring to anyone in particular. I think the discussion took an unproductive turn and we all got a bit testy. Let's focus on a logo design and site content. Once we have a logo we can take a look at this again. Unless mikedee wants to add something I declare this thread officially dead, RIP.