Glowing reviews?
August 14th, 2007 by Ted KulpNeedless to say, I’m a bit frustrated. Take a look at this review…
Very clean and simple CMS. Editing stylesheets and templates is a bit awkward, but after some time creating your own stuff, one can get used to it.
The quality of the user-submitted modules is abysmal. Many of them are fundamentally flawed and their PHP code is often plainly wrong.
If you avoid 3rd party modules, CMS does the job very well.
This comes from our page on opensourcecms.com, which is pretty much the largest pusher of traffic to our site from the outside world. A lot of our new users find our name on the list and check us out. And this is pretty much the first thing they see now.
The developer’s forge is a great idea, but it almost seems like it’s hindering as much as it’s helping. It’s not the first time I’ve heard this complaint, so we as a group need to try and figure out what we can do about it. Whether it requires a more strenuous testing/acceptance procedure (which we don’t really have the manpower to do), or if we just only approve projects that we now will be done right… well, we just don’t know.
Any suggestions? This needs to be corrected or it will become a downfall of this project. And I refuse to let that happen.
August 14th, 2007 at 7:11 am
Outside of some of the most actively developed modules (Calendar, Album, etc) I’ve found the quality of modules to be variable; also, too many are launched with early versions and then never developed again.
I try to test new modules as much as I can, but I can’t use them on any of my production sites until I’ve had the opportunity to try them in a dev environment, which isn’t as often as I’d like. I’d like to get more involved in developing modules, but my PHP skills aren’t up to scratch.
One thing that would be helpful is if developers ‘ate their own dog food’; for example, this site itself is still hosted on version 0.13, when it should certainly be on the latest version possible if mistakes are going to be spotted.
Perhaps a site could be set up which uses the latest version of CMSMS and to which all modules have to be submitted by the developers and approved by a panel of volunteers before being released? I’d be happy to take part in this.
In the end though, I think we really need more developers to get involved with the project, and more users to test thoroughly before using in a live environment.
August 14th, 2007 at 8:17 am
My views;
-first things first; We need a team which tests all currently downloadable modules. All the modules which dont work with version 1.1 needs to be deleted, or in de different part “broken modules” but i prefer simply deleting. Off course one week before deleting a non-compatible module the main developer needs to be send an email if he`s willing to rewrite the module which will make in compatible.
-This will decrease the number of 3rd party modules, and with less modules, it will be easier for a (new) team of say like 5 people or so, to look over the code of all modules, and keep track of new modules.
There is no bulletproof system but i think this way quality will increase.
I`m going on vacation tommorow for 2 weeks, but I`m happy to help when my vacation is finished in 2 weeks, your always welcome to send an email if there`s anything I can help with.
August 14th, 2007 at 8:46 am
Thanks for the comments so far!
I’m glad to see that others are as interested in cleaning this up as I am. We’re going to have a conversation over the next few days (or possibly at the dev get-together at the latest) about this topic and lay down exactly what needs to happen.
Because we have a forge rewrite in the works, it’s the perfect time to lay down business rules and other forms of moderation that will allow us to increase the quality of the modules that are hosted on the forge.
August 14th, 2007 at 10:25 am
Why not create two classes of modules: “official” and “user published”?
The “official” modules will have to be tested by a dev-team and meet certain standards (as to documentation etc).
The “user published” modules, on the other hand, will be more on-your-own-risk, but still make it easy to find and share codes for users. When a “user published” module has proven stable etc, it can be moved up to “official”.
August 14th, 2007 at 11:19 am
@Mats Ingelstrom
I like that idea, maybe there can be a message like “reviewed and approved by modules team” in the forge when you are viewing a certain module.
August 14th, 2007 at 12:32 pm
How about releasing a document stating the coding conventions? Once you have a guideline, then you can have something to review submissions with. It would also help promote and produce uniformity of the modules, and perhaps make the reviewing easier.
Something like this, but perhaps a bit less stringent http://www.dagbladet.no/development/phpcodingstandard/
August 14th, 2007 at 12:40 pm
I think a good way to manage quality in a open source system like this is to implement a Defect and change management system as this will allow you to review code before it is released ot the public and this also includes custom user modules. This would require so time but will boost quality.
Another option is to public a basic how to guide for people creating custom code, so they know what standard you require.
August 14th, 2007 at 1:14 pm
I agree with Signex. There needs to be “review and approved” stamp on all quality modules. Mozilla does something similar with “Recommended Modules.”
August 14th, 2007 at 4:01 pm
I love the idea of having “user published” & “official” tags on the modules in the forge! Great brainstorming there!
I know that a lot of GREAT developers can’t focus a lot of time on their CMSMS modules because of time/money.
And truthfully when you compare the amount of CMSMS community involvement with something like Drupal, for instance, it’s kind of difficult to make the A+ grade in the opensouce community. personally, i think CMSMS is the best thing out there, and i (like all of you) would like to see bigger & better things coming out of the forge.
I had two ideas, and i’m not sure if it’s totally kosher to sell modules for GPL software… what if CMSMS had a “development booster” website where developers could charge for modules, here were my two ideas, either…
1.) sell each module as a single license until the account totaled up to the amount of hours they spent in development then releasing it openly as GPL
or (a more likely one)
2.) give an estimated price based on the number of hours a developer(s) would think the module would take to get it into beta. then accept donations until that total has been raised - once it’s been raised the person or team can begin development knowing the funds are there in an account to compensate them for their time. once beta is released and tested by moderators/community those funds are released to the developer and a new price is set for a stable version and the process is repeated.
the final module is then under a GPL and was funded by the people who wanted it the most (namely, businesses who have clients and need very stable modules - this gives them the opportunity to give funds to encourage projects that peak their interest instead of just waiting around or hiring a freelancer outside of the community).
–
these ideas may sound stupid, or anti-opensource. that’s not my intent, i’m just trying to think of ways to encourage development & donations so CMSMS modules are the best they can be and keep them GPL in the long run.
any thoughts?
August 14th, 2007 at 10:17 pm
I’m willing to help with testing from an end user’s perspective. However, I wonder if most of the problems folks have are from old modules still hanging around.
Maybe all modules listed in the Trove should have an expiration date so as they get ’stale’ there can be a statement that simply says, “Last update to this module was 120 days ago…” locating in the summary page, for example.
August 14th, 2007 at 11:07 pm
As I see it, the Developer’s Forge, and Usable Modules are two different things. A Developer’s Forge is a place to track developing projects, and to encourage collaboration. Fully developed and stable modules need to be listed elsewhere - and they should either have to meet minimum requirements to be listed OR, they should have the option of user submitted reviews.
Frankly, I find looking for a new feature to be fairly frustrating, because I have to sort through a bunch of incomplete or abandoned projects just to filter out the few gems in the bunch. The good ones rarely have helpful descriptions - many don’t say what they actually DO, or what the features are!
Abandoned ones don’t belong anywhere. Incomplete ones belong in the Forge. Stable and useful ones belong in an area where useful information can be given about them.
August 14th, 2007 at 11:41 pm
We have also experienced a wide range of quality among the modules. Often it results in us developing our own stuff, since it tends to be faster that correcting buggs.
Mats Ingelströms idea seems to be a good one - release modules in two different classes.
August 15th, 2007 at 2:10 am
I totally agree with the last 4 comments. It seems like the best and easiest/less time consuming method. People could recommend modules from the user published section and/or people could report an approved module as faulty.
If somebody could volunteer as a module moderator that would be a great solution. I would like to do it myself but I am not a programmer and feel it wouldn’t be right.
Another idea could be to use a fundable.org fund-raiser to pay for modules to be created.
August 15th, 2007 at 3:45 am
I like Mats Ingelströms idea too. I think the two sets need to be clearly separated, and not just marked with a “reviewed and approved” message when viewing the module. The default should be just to search and browse the “approved” modules, since most users want something which they should be able to get running without too much hassle. If they can’t find what they are looking for there then they might want to have a look at the user published modules.
I think the problem really is that the developers’ forge is trying to do two jobs - sharing code between people who want to play with PHP and extend CMS, and publishing useful modules to people who just want to use CMS to run a website.
August 15th, 2007 at 3:57 am
The idea of two classes sound great.
But as Ted said there’s not enough man power available for a “reviewed and approved by modules team”. As you perhaps know we have a qualitiy assurance team too, yes! But most of members (westis, patricia) have a lot other things to do (can’t see them in forum).
Another problem is the variety of existing installs (different versions of CMSms, PHP, databases, server). In my daily support there are users they comes from CMSms 0.10!!! Since this time there are a lot of developing steps. An (approved) module/tag that runs great with 0.10, but wount maybe work with current version.
Think we need help from all developers. Not sure if it’s a possible way but maybe EVERY module/tag should get status “unapproved” when a new CMSms version is released. If a project is active developed dev can/should check his modules with current version and set it to status “approved”. For unapproved extensions every user will know using is maybe risky … it’s only an idea.
@Mats Ingelstrom
You are welcome to be the founder of module/tag proving team :).
August 15th, 2007 at 4:10 am
We had a dev meeting yesterday and this was one of the main topics of discussion.
We think there are several things that could help this situation…
1. User rating and comments for modules. These would be designated on the project’s front page and should also show through to the module manager.
2. Automatically designate projects that are older than 6 months (no releases or svn activity) as being inactive. These projects can then be easily gone through by either the dev team or another group we designate to be removed if they’re too far gone.
3. As you’ve guys have discussed — some kind of review board. Modules that have been reviewed by a member of this group could have some kind of official seal or something put on it. “This module worked satifcatory (not necessarily bug-free or feature complete, mind you… just working good enough) with version 1.x”. Again, this would show through to the module manager in some way.
The nice thing is that since we’re also in the process of rewriting the forge, we can do whatever we want with workflow. One of the items that we’re going to handle at the Copenhagen meeting is the walkthrough of a project from beginning to end to decommissioning. Once that workflow is set, then we get someone to help us finish the new forge.
Thanks again for all the feedback! As usual, it always makes me smile when I see that people care about this project as much as I do.
August 15th, 2007 at 5:42 am
I find a lot of thoughts good. One of them came from Damien. He speaks of a document containing a standard for programming. This would certainly help QA-ing a module during approval time. However what I find frustrating at this moment in time (I’m building a module) is that I miss a step-by-step description on how to build something. The administration side of the screens is not a problem. The client side however is a continues strugle. How come? I can’t find examples that I can copy (the idea) from. As a result I’m now in the stage of preparing something that I understand and am able to get up and running. However I understand the risk that I introduce: yet another way of getting the same thing done. From a QA point of view I understand this is bad.
So my ideas are:
1 - to prepare one or more step-by-step wiki page(s) on development issues.
2 - to have a list of issues and where to find a good example how to solve it (solution works in one defined way).
Although I’m quite new at CMSMS I volunteer to join a development team.
August 15th, 2007 at 6:27 am
Looks like a great plan. I notice some modules continue to work with every CMSMS version. Bookmarks still works perfectly with version 1.1 and hasn’t been updated since August 1st 2006. Because of this I am glad to see the idea of a review board that will be able to verify that some of the older modules will work with the latest CMSMS version even though the module may be dated.
August 15th, 2007 at 6:53 am
Wouldn’t making a nice review and posting it sit on the top of the list so people see that first?
August 15th, 2007 at 12:27 pm
I’ve been working for some time with CMSMS. I looked at different CMS before and CMSMS was just the thing, I was looking for. It is simple and meets my requirements. I looked at different modules and sometimes I became a little bit frustrated when some module wouldn’t work with the CMSMS version I was using. Upgrading ist always a bit of a chalange, will a module still work or not? I think everything has been told before and an increase in quality assurance and release management that shows which module works with which version of CMSMS would be a good thing.
CMSMS is a great tool and you guys do a very good job!
August 15th, 2007 at 5:48 pm
What about summing up, which version of a module works with which version of CMSMS?
module_x 0.10 works with CMSMS 0.13
module_x 0.11 works with CMSMS 0.13 / 1.0 / 1.1
…
So everybody can easily see what has been tested.
Of course this does not give information how deeply it has been tested.
August 15th, 2007 at 10:53 pm
I think that developers has not much time to test their modules with all version. Other user ors developers can report when they uses that modules. I think it better
August 15th, 2007 at 11:43 pm
About the coding standards document,
I believe we already have something, but its scattered in the forum (iirc calguy’s post) and wiki.
Also the skeleton module has lots of information (and is pretty much the style we like to see).. the problem is that the information isn’t in found easily and isn’t as clear as it should be..
just my 2c
August 16th, 2007 at 4:52 am
I understand Ted’s frustration but this is just *one* review which is a bit negative. You won’t get 100% positive reviews, every once in a while someone will say “this cms sucks”, for whatever reason. It is plain bad luck that right now that comment is showing at the top … if everyone who hasn’t already written a comment would write one it would be more in context.
Nevertheless I agree that the modules need to be approved.
If there will be some kind of “user approval” I’d be happy to help.
Keep up the good work!
Alex
August 16th, 2007 at 7:09 am
I am a new CMSMS user - and this CMS is the best. Period. I migrated from WordPress and looked hard at Joomla and a bunch of others prior to making the jump. I actually found CMSMS on the opensourecms.com website. CMSMS is simple, flexible and standards compliant - not something that can be said for many of the other systems out there. I have converted 3 of my 4 sites to CMSMS and the fourth site will be converted over at the next site update.
With that introduction out of the way… I agree with earlier comments that a designation of “official” or “user submitted” would probably go a long way in lessening confusion. The other thing that has been mentioned that would be of immense help is weeding out old modules that were started and never completed.
I realize everything involves work but it seems like these two ideas might be the most bang for the effort required. Sadly, I am not in a position to contribute my help right now. I am in the middle of a major move… Hopefully within the next few months I can get involved and be of some use though.
I have been building websites since 1998 and I just love CMSMS. Keep up the good work!
August 16th, 2007 at 11:50 am
I’m in a similar position as Rob who just published other than I’ve set up quite a few custom sites for clients using CMSMS. I’ve also tried and tested everything from DNN to WordPress and obviously I’m a huge fan of CMSMS so much so that our entire web development approach, sans ecommerce, has CMSMS for the foundation.
However, I totally agree with the criticism posted at opensourcecms.com.
And I agree with those who suggested:
There needs to be “review and approved” stamp on all quality modules. Mozilla does something similar with “Recommended Modules.”
While I’m adept at designing around, customizing, and optimmizing CMSMS (call me a power user I guess) I don’t have the ability to be a heavy duty developer in the PHP world.
However, with the above conditions being taken into consider, I would like to volunteer to be utilized in the process…maybe as an official beta tester or something.
If anyone has any insight on those of us who are not developers to participate please let me know.
August 16th, 2007 at 1:42 pm
I am a believer in CMSMS. It has great bones. The idea is awesome and much of the current implementation is good, but I do agree that the modules are it’s downfall. Yes, there are the bright points like the user tags and Album, but it seems that the quality and content of many of the modules are lacking. I think that CMSMS is solid, but I would suggest that some time and effort be focussed on developing some more excellent core modules. It seems the efforts are great but spread over a vast selection of modules. Why gather everyones’ focus to develop the poo out of some really excellent modules to set the bar for other programmers? Quality attracts talent. Develop top notch modules for some very common needs, for example, I have yet to find a super amazing mp3/media player plugin for CMSMS. I was looking for one for the JW Media Player and there were links for Ruby, Joomla, WordPress, but none for CMSMS. I believe that by focusing on making some really slick welll programmed modules will only draw attention to the fact of what a great CMS CMSMS is.
August 16th, 2007 at 2:07 pm
I downloaded and installed CMSMS a week ago for a friend. I had used an earlier version several years ago and was excited to try the new one. It installed great, and then I went looking for some modules. I installed half a dozen or so, but one I installed completely broke CMSMS. (And I can’t remember the name of it right now, don’t you love this helpful post!) I was able to hack around and get it uninstalled/removed by installing “older” versions of it and then removing them one at a time. Needless to say, I am steering clear of all but the most basic modules.
I agree with all the above, some kind of tiered system or “official” modules and the rest are use at your own risk/take it up with the module dev.
I use IPCop (firewall/router) in some places, and there they have a team that just does the basics (IPCop), and then others who work with addons (content filtering/spam protection, etc) etc. Addons are “use at your own risk” but they do increase the functionality for those willing to work more/risk more.
Todd
August 16th, 2007 at 3:47 pm
Wow, this is a fantastic thread! This is exactly the kind of community involvement that a project like this needs. I have noted those who have expressed interest in volunteering with testing, documenting etc., as I got that responsibility at the recent development team meeting.
After the Copenhagen meeting I am sure the entire organization around CMSMS will get a real boost and to know that there are many willing to help in different ways, with their variety of skills, is great. This thread is part of a current general overview of the organization, to better use the skills of those in the Development Team and all those willing to volunteer with their skills and their time.
Thanks for all the great ideas!
August 16th, 2007 at 9:58 pm
- My 2 cents worth.
still being fairly new to cms, unfortunately I must agree with the comments of some users about the mods.
I have had a play with quite a few of them, and most need ‘tweaking’ before they work as they should.
I really like the idea of maybe having a table that displays which mods are compatible with which versions of cms, as well as a voting + comments system where users can leave feedback for other users & developers.
What about having a commercial mods area?
I use a number of other open source programs one of which is izabi (a social networking platform), there they have a model where some of the mods are free, and some of them are ‘commercial’ mods, it seems to me that developers would be a lot more inclined to develop mods that ‘just work’ if they saw a monetary return for their efforts.
After all that is all anyone wants when installing a mod, especially with such a great backbone application like cms, which ‘just works’.
The commercial mods don’t have to be enourmously expensive for it to make it worthwhile for developers, as if they create useful mods’ that work’, they will obviously sell quite a few. That way everyone wins, developers get paid for their time, users get mods that work.
It would also be a great way for the core developers of cms to get some much needed funds, being the core developers, who knows the program better?
I for one would love to see (and be willing to pay for) “official” commercial mods made, that work as beautifully as the core cms program does.
End of mini rant.
August 17th, 2007 at 5:29 am
I posted the initial comment on http://www.cmsmadesimple.org/ and I sincerely apologize for the frustration that this comment may have caused for the developers of CMS made simple. I spent some long nights fixing broken modules and I wanted to share that feeling of being alone in the dark with a lot of the 3rd party additions to this otherwise very useful and clean CMS that I continue to use inspite of my ongoing troubles with some modules. The question of quality assurance of 3rd party modules, add-ons, templates et al. is something that every large PHP project faces. There is no perfect product. Of course, my own RoR stuff I consider to a certain degree ‘perfect’ but my colleague would probable wait for me with a pickaxe behind the door if he would have to fix or extend things in my projects.
Now, I would tone down my criticism of CMS made simple because the quality of some of the modules is not the fault of the core project team. On another note, I really appreciate that you take criticism seriously and that you immediately started a discussion about improving things. Also, I will participate more actively in the forum to discuss certain problems to help to improve the quality of some modules.
August 17th, 2007 at 5:43 am
IMO it would be a start to really make use of the dev.cmsmadesimple.org website. ATM virtually all discussion about modules (and problems with them) is taking place at the general forums. That’s not where it belongs though.
Result: people who are just visiting dev.cmsmadesimple.org get the feeling as if the module they’re after has been abandoned since no discussion is taking place there. So module creaters should really give support where it’s due and not in the general forums.
August 17th, 2007 at 6:34 am
My idea is as well to have some (approved) modules being recommended by the DEV-team. This would lead to viewer similar modules that all do more-or-less the same, but by focussing/concentrating on only the best/recommended I think that they will progress more than if all effort is spread among all…
That was also the idea behind my post of August-7th,
http://forum.cmsmadesimple.org/index.php/topic,14093.0.html
Ronny
August 17th, 2007 at 6:38 am
Some additional text to the previous post. New users will then find it easier to choose for several modules, as they don’t have to do all the search for modules by themselves.
Once again,
Ronny
August 17th, 2007 at 8:26 am
I also think it would be good having some “approved by devteam” modules.
Another thing I thought about is about which versions of CMSMS a particular module are compatible with. Just as Jonas also mentioned in a post above. See Firefox Addons, for example this one https://addons.mozilla.org/en-US/firefox/addon/3615
They have something like “Works with: CMSMS 1.0-1.1″ for each and every module. This makes it very simple to see what versions of CMSMS the module have been tested against. Maybe the module manager should reflect this issue somehow aswell in that case.
August 17th, 2007 at 8:35 am
Reading what has been posted, there were two thoughts that came to mind: second that the idea of a two levels seems to be a good one. I also take the point about lack of capacity. Nevertheless, I think that even if 4 or 5 modules were cleared it would help users, albeit flagging that it will always be an on-going process.
The first thought, however, was the negative review. As faqlork writes, there is no such thing as a set of perfect reviews. I was a little alarmed that there has only been a small discussion on how to deal with the bad press. As a very green newbie, it is not only a matter of loyalty to CMSMS but also about ensuring that a good product remains publically attractive to the broader community. It would be a pity if people were turned off from using, enjoying and being supported in their work simply from a weak review.
I would simply like to suggest that anyone who feels sufficiently motivated by the comment or have been supported by CMSMS should post fair comments on the original website opensourcecms.com. For myself, I am amazed not only at how far it is possible to play around with CMSMS but also two further things. 1. without any initial idea of CSS, PHP MySQL etc it is incredible how far you can go and pick things up. 2. the online forum (especially ‘carpenter’ Mark and Ronny K) is speedy, dedicated, friendly and diplomatic enough not to point out that they have already told a hundred other people the same answer. It is a lovely and fun place to start my website life. It would be a pity if others were deterred from this.
August 17th, 2007 at 11:24 am
Hey,
yes, as many people have said before, the idea of reorganising the modules really makes a lot of sense.
What’s about setting up a different part of the website (plugins.cmsmadesimple.org ?) up as some kind of plugin-portal? The newly to found Plugin-Dev-Team (as someone suggested before) could use it as a platform to suggest good plugins, maintain a database and provide help for plugins (to an extend). This could be the primary focus-point for end-users, whereas developers could use the Forge to maintain their projects and “submit” their plugins for review / approval to be included in the offical DB and gain the “official approved”-status?
Some said before that the forge is trying to do two things at once - and I have to agree. A single point of reference for users would make plugins a lot more attractive for the end-users and developers.
I also liked the idea of making paid-plugins available. There are some functions that would be nice to have in CMSMS, but they would surely take a lot of time and effort to implement - why not paying someone to do it and selling the plugin on the same platform as the free plugins? Sure, this might not be the true-spirit of community-based development, but I am sure that there are companies / people out there that have written plugins for their specific purposes, but are reluctant to give them up for free. A centralised platform for this would really be a big advantage and encourage users to participate as well.
As far as specific programming-guidelines and examples go: I am not really into plugin-development, but, would it be an option implement some features as plugins, rather than hard-coded into the core? I don’t know what you guys are planning, but maybe this would make some good examples for further plugin-development and could also provide a start-point for fellow plugin-developers?
(I hope my ideas made some sense)
August 17th, 2007 at 2:02 pm
I believe that this has been the 400 lb. gorilla in the room that no one wanted to talk about. This is a very healthy exchange of thoughts and ideas. I believe that we cannot let this momentum die down. You can tell by the post that there are plenty of people who care about the future of CMSMS. We should be excited, because many entities travel blindly down the road with hardly an inkling of what their users/clients really want or need. This is not the case now. There is a clear direction. Ok, so we all agree that changes have to be made. What are the steps that we are going to take to travel in this direction now? The discussion has been great. I would like hear some sort of a tentative plan of action from the CMSMS gods. We know where we are. We know where we need to be. Now how are we going to get there? I am enthusiastic to see what comes out of this point of crisis. Is this the breaking point for CMSMS or is this the start of a new season of growth and creativity for the project? I hope for the latter.
August 17th, 2007 at 5:49 pm
@Josh: Agreed about the 400 lb gorilla. This will be one of the largest topics we discuss at the CPH meeting. My job before that time will be to get the forge rewrite as close as I can to being complete, sans the workflow and other features we’ve discussed above. Then we’ll really nail down what needs to be different from the current system (and, honestly, we all know already… just need to talk about implementation) and get the new forge out there.
It’s going to take a few months easily before it’s all resolved. But it needs to be, or could play a very large part in the downfall of this project.
@all: Thanks again for all the great comments. This has really been a refreshing change of pace for me, as I don’t frequent the forums as much as I should.
We’re going to get this problem resolved. We have to…
August 18th, 2007 at 6:53 pm
Dear Ted,
I have always been and continue to be a HUGE fan of cmsmadesimple.
I have checked out virtually every PHP MySQL CMS out there and
cmsmadesimple has the best admin (backend) interface I’ve found
and while it may lack the scope of Joomla or Drupal it just plain rocks.
After many years of studying the Internet, Open Source, CMS and the like
I’ve come to one conclusion. the number one weakness of most
Open Source projects it the lack of good Documentation and Training.
If the people who write the code could just understand how important this is.
No matter how good the tool you create if no one can figure out how to use it
it becomes virtually useless. The folks that write the modules for cmsmadesimple
must understand that clear easy to read instructions are critcal to the successful
use of their module. Forums are great, help is fine but good Documentation
is absolutlely key to the use and adoption of any software.
Thanks again to everyone for providing us with cmsmadesimple and it’s add ons
now please document clearly how to use them.
August 19th, 2007 at 1:08 am
I will just echo what everyone else has said about CMSMS and its modules.
I think that this system has loads of potential, and for simple page-based sites this is a fantastic system. But for anything beyond basic content management, I find myself tearing my hair out with add-ons that don’t work.
The only thing that stops me from committing fully to this project and using it for all my projects is a) the lack of e-commerce modules and b) never being sure if modules actually work, andhaving to get a developer to fix things if it breaks.
I absolutely adore the way CMSMS handles templating, and there is so much potential! I wish there were more I could do (I am not a strict developer, more a designer) to help this project.
August 19th, 2007 at 3:44 am
Hi all,
i’ve read near all the posts here. The “official” module flag is a good thing to find working modules. I don’t think that a lot and a lot of modules is needed to have a good cms, i prefer have 2 or 3 third party mods they work correctly and are usefull. CmsMS is “young” instead of Joomla or others cms, but it’s a very good tool for the moment, so with a little time modules will be better.
2 global suggestions or ideas (perhaps not the good place…) :
> in a way of showing that CmsMS is a quality tool, why don’t we use a latest version with a clean design? Interested designers will send an “up to date” logotype and global design propostition (screenshots), and community will vote for the best. As designer, i know how the first impression can be important for future users, and would be happy to contribute to made cmsMs more “sexy”.
> about the devs: i think the permanent (or official) devs must be more visible on this site. Why? There are 2 categories of cms users : pros and particulars. Pros need to have contact with a strong team in a way to build custom mods for example for their cutsomers. I think that a dedicated page for the ‘official’ devs would be useful, and imagine that it can bring more money and more quickly that waiting for donations…
August 19th, 2007 at 10:25 am
I think this review is finally a good thing, even if the terms were a bit hard.
It triggered a serious discussion about Quality Assurance, and that is good.
For me:
If a programmer has no time to document a module, the module should not be an official module. If it is within the forge, it should be clearly marked as “untested, use at your own risk”.
Each new release of a module should reset its state to “untested” until QA procedures are completed.
I also think more emphasis should be put into documentation: user documentation, tests documentation…
This is not a critique for my part, just suggestions because I like this product and I would like it to grow stronger.
My advice is: freeze development and emphasize on documentation and reflexion about QA.
Patrick
August 19th, 2007 at 12:22 pm
This text was translated by Babel Fish.
The modules are reliably a large problem with CMSMS and it are for a long time well-known.
In the forums those write user already for a long time over problems - at most over module problems.
These contributions cost more time as if a core Developer of modules from the Top Ten gradually on the current conditions bring.
You should not believe, that programmers wait to write modules for CMSMS.
There nobody waits for it.
Co-ordinates, which modules at most are needed and take these into the core.
The Top Ten cannot be so wrong, which one could not update her fast.
95% of all modules of projects can be deleted immediately.
August 19th, 2007 at 1:15 pm
Just another idea to improve the quality of the product:
Why not create an “exchange”: user could request help against payment. Some expert developers might answer these requests…and keep the money, BUT they MUST:
- fill the support database with a description of problem/solution,
- mention if problem/solution might be of help for someone else
- mark errors/missing info in the documentation.
Patrick
August 20th, 2007 at 1:14 am
I think it is a good thing to have a discussion like this, also in mind that 2.0 is coming and that it’s good to have a solid structure from that moment on.
I think that the signals that are given, also the one Ted mentioned to start this thread, are a good trigger to rethink the structure. In organisations that deal with growth, this is a normal process. You can’t keep growing without (at least thinking) about restructuring. I think that the signals make it clear that CMSMS has grown, and that some action is required to make the next jump in growth/quality possible.
I think that the release of 2.0 will be a good starting point to work with the changed module-system. That’s a good moment as well to leave some unattended modules behind, and only go with the ones working (correctly) on 2.0.
Ronny
August 20th, 2007 at 4:10 am
Half-working (and half-broken) modules are a real problem - as well as forum answers like “Oh yes, there was a problem in previous patch - it’s fixed now” (but it still isn’t), or “It works here - you must have a problem with your IE (or Apache, or whatever)”.
Anyway, I’ve not been here for a while, and it’s nice to see all these new CMSMS enthusiastic lovers.
My 2c suggestion : let’s clean the forums, archiving all closed, redundant or useless issues. They’re frightening for new comers, and probably counter-productive - they turned to be such a mess that it’s probably easier to ask and ask for the same question forever rather than searching… and finding the right answer. Such an archiving will be a long and hard time, but probably the worth.
BTW, tagging questions “SOLVED” (ie, user education…) would be fine, too.
And to end up on an optimistic note… I don’t feel the problems with non-core Modules would put CMSMS in real danger, it’s really too nice. But having problems with 2, 3, 5 modules for sure leads to be very cautious with new ones…
Fred
August 20th, 2007 at 12:02 pm
I’m new to CMSMS and have successfully installed it. I’m very comfortable working with XHTML, PHP, CSS, and JavaScript and am working through setting up the templates and stylesheets. Based on the comment about the modules in the OpenSource CMS comment, I hoped to find information for recommended, and stable, modules on the site for CMSMS newbies. I would rather avoid the issues dwave wrote about, installing modules, and then spending days debugging them.
So far, I haven’t found a stable/recommended modules list despite searching the forums over the weekend. Does it exist?
Ronnie mentioned a forum thread (above in one of his comments) about module recommendations. However, the thread doesn’t include any suggestions.
August 21st, 2007 at 12:41 am
I think there is a great inconsistency to the way various modules/plugins work. Both in the PHP code and in the templates and CSS they use. Greater standardisation/simplification would bring much greater ease of use. I think a much simpler interface is called for.
For example when a user installs CMSMS for the first time, ask them wether they want a calendar, blog, news, photo gallery etc. Tell them what each does and that they can add them later if necessary. Let’s have all have a common interface and style, including their management.
Perhaps we have lost site of the goal of ‘CMS Made Simple’? Having spent a day to get the calendar module to do what I want it todo, (nothing special - just proper lists, lose the DIV’s and replace with classes (how about some common classes!), proper url’s and changing page titles/headers for accessibility - why not straight out of the box?,) I think there is some way to go.
For example, my young son produced a stunning web site with photo galleries, movies, RSS feeds, uploads, comments and a blog in under 1 hr with iWeb!!! OK the code is messy and about as standard compliant and accessible as a chocolate teapot - but it looks very good, and was very quick and easy. Did I mention that this was his first site in iWeb. Did I mention that he could do it. There is no way he would have got near a site that good in six months+ with CMSMS.
Now don’t think that because of my comments that I dislike CMSMS, rather the exact opposite, sometimes you have to be cruel to be kind. I’ve persevered with CMSMS for many years and contributed bits and bobs when I have time. It’s improved greatly in that time and I think it has great potential. When you get to know how it works, you can do some very good stuff. Trouble is that getting to know this takes some time… we need to flatten and greatly shorten that learning curve to properly justify the ‘CMS Made SImple’ tag.
August 21st, 2007 at 12:10 pm
There is alot of talk in the comments here and I just on’t have time to read it all right now.
But I wanted to just say I think CMSMS is the best CMS I’ve come across, biggest reason being for ease to design for. I love that it’s completely CSS based. It is absolutely a dream to build a site on it, and have it look however you want (Anyone ever noticed how you can peg a Joomla site from a mile away?).
Alot of talk is no doubt going on above about how to make the module environment better. I don’t need to chime in on that.
I want to stress that CMSMS is an outstanding CMS that I love dearly and am so very happy I found.
August 21st, 2007 at 2:10 pm
I totally agree with the statement that Russ made: “Greater standardisation/simplification”. One of the burdens that the code of the various modules doesn’t look alike is due to the fact that it is hard to find a standardisation (could someone of the coreteam name one module that can be seen as the standard?) . I stepped in on version 1.0.8 and tried to install ModuleMaker. Useless, since it only works up until 1.0.5 (I only found out after downloading, unpacking XML and trying to install. This last support version should be mentioned in the download header information). So how are versions after 1.0.8 supported with regards to a more automated (and hence more standard form) generated php and template?
Since I use bits and pieces off various modules, I’m preparing for some new, my own, standard. Not a good thing.
My idea, try to get the ModuleMaker working for each version (at least 1.1). This will help new developers very much.
August 21st, 2007 at 2:46 pm
“The quality of the user-submitted modules is abysmal. ”
It’s true. In itself CMSMS is very good. But just go to module manager after you’ve installed it: not all the modules available on the site can be installed because not all of them show up in the module manager. You have to go to
http://cmsmadesimple.org/downloads
to see all the modules and what do you see once you get there? I’ll bet more than half of the listings “haven’t released any files” or are at some alpha stage. And even the ones that are past the beta stage are still outdated or don’t work right.
I’m using the Google Sitemap module. I don’t know if any other cms does that, but I’m afraid this module will fade away for the next version. These are the problems you have to tackle:
remove old/outdated/inactive modules
keep the essential modules or integrate some of them in the cms installation
August 21st, 2007 at 2:48 pm
And also take some of the modules which are very useful, which their respective developers have given up on and that no other CMS has (formbuilder!) has, fix them and make them a part of the basic installer. The formbuilder developer seems to have given up on it and it’s a shame.
August 21st, 2007 at 3:26 pm
Just installed a new system, and this time frustration set in … :-((((
IF there will be “approved” modules, I suggest (beg! demand! whatever!) that only modules with a useful documentation get “approved” status. Sometimes I feel downright silly, when the “help screen” does not even mention the setting of the module’s admin options and - after unsuccessfully searching the forge (and clicking on the “module homepage” link just to find out that the homepage does not even exist) - I have to ask the forum or plunge into the source code to find out what the heck this or that option means …
Just like many others I have tried and installed gazillions (ok, make that about 30 ..) CMS systems until I found CMSMS, which was love on first sight. My time is scarce, but I will offer any help I can give … testing modules … translate module docs into german …
Cheers,
Alex
August 21st, 2007 at 7:42 pm
some more loose change:
Just an idea, It would be really useful if it was a requirement that the developer(s) had a working example/live demo version available for every module, or at the very least screen shots .
So you wouldn’t have to install the mod just to see it in action.
August 22nd, 2007 at 2:44 am
CMSMS = totally fantastic and awesome, so please dev team, don’t get down-heartened by a few negative comments. You’ve produced a fantastic system that I use and recommend wherever I can.
However, the modules system is certainly the weak point of CMSMS and would benefit from improvement. So a few suggestions, in addition to all the others already made:
In the to-be-revised module listings every module should have listed with it:
* The module name
* This module is compatible with CMSMS x.xx, x.xx, x.xx
* This module is version x.xx
* The last release of this module was on yyyy/mm/dd
Maybe this information should be included in the module code as standard, so that a module could check with the central modules page and alert a CMSMS administrator that “Hey! There’s a new version of me available that works with this version of CMSMS! Do you want to upgrade?”
Every module should have full documentation in a standardised format. Too many slip by with only a version number and one line of text.
As others have said, an official “seal of approval” for modules that certifies they are maintained and compatible with the latest release. The onus should be on the module maintainer to submit this evidence?
Every module should have a forum associated with it to address support requests. No more getting module help requests mixed up with CMSMS requests.
Perhaps some modules should be included in the CMSMS core? Calendar and Album are used by nearly everyone.
A more obvious and easy to use bug tracking system. It’s quite hidden at the moment. This new system could have categories for the various 3rd party modules?
Some modules could do with learning techniques from each other. e.g. Every module should learn from the FormBuilder module - it has the wonderful feature of printing the browser alias string that needs to be pasted into a page, so it can be easily copy’n'pasted. (BTW, this is a major sticking point for our users)
August 22nd, 2007 at 11:25 am
OK some modules do seem a little iffy, but given how good the core package is I can’t imagine this really damaging the project, although I have to admit that sometimes it is a little frustrating. I would gladly join a team of testers and if needed would be willing to write documentation — that way we don’t discourage developers who may be great at writing code, but not have the time to write documentation.
August 22nd, 2007 at 6:48 pm
I’ve been a big advocate for CMSMS and have donated into the development kitty with every client site I’ve done.
There’s a lot of discussion about modules, and a lot of outstanding input. I want to call attention to something related that I’ve noticed and is a common theme in many of these comments: the difficulty in finding (and sharing) good information.
There are two main information systems for CMSMS: the wiki and the forums.
Right now, the wiki is harder to use for seekers and contributors than it should be. As a seeker, tips and tags are rather scattered around, and most are organized by the solution mechanism (user-defined tag, etc.) rather than the problem to be solved. And that leads people to turn to the forums for everything. I’ve seen far too many repeat-topic forum posts since I started working with CMSMS.
In the forum, reusable tips — those that can benefit lots of users, not just fix an individual’s issues — scroll away too quickly. Sometimes the best tips are in threads you’d never think to read or search in. Later, if you remember seeing a tip, you’ll have to hunt around to find it again. If you do a forum search, there’s a lot of hay to paw through before you find your needle.
As a past active forum and wiki contributor, I’ve sometimes seen or come up with good tips, but it’s never been clear exactly where I should post them in the wiki, where they would be easier for others to find and use. Many times, rather than take on the additional task of figuring that out too, I didn’t bother. Wikis, in shunning hierarchies, usually need extra editorial guidance to set up intuitive “spaces” for users to add information.
The wiki hasn’t kept pace with some of the potentially great core features like the Events Manager trigger capability. The one example on the wiki page “sends an email when page content is changed,” but doesn’t show anything about how data on the changed object (page title, user who made the change, etc.) would be accessed and manipulated. I’d love to develop a tag to check page content to help remind my less-technical clients not to image leech from other websites, for example, but I don’t have the time to hunt for documentation on the feature.
Lastly, frequently-desired, highly-customizable add-ons like Tags, Modules, and Themes need better visibility and access. Competing CMS sites put these things right in their main navigation. Pull them up where people can see and use them. Then, as others have said and Ted has supported, give easy-to-use feedback mechanisms for them. If a module developer didn’t document something crucial, a user can post a note on that module to help or warn the next person. If the module is fabulous, or isn’t compatible, or people don’t like it, give them a mechanism to rate and rank.
Having worked with a couple of other content systems, I strongly favor and support CMSMS, and would like to see it get more visibility, use and enhancements. Don’t mistake the critique for negativity — these issues are very fixable. The message developers should be getting is “help us help you.” Good information leverages a great system: newbies can get results and experience faster, and more advanced users can understand more of the nuances of the system to build on it and on each other’s ideas. I think Josh’s quote earlier is perfect: “Quality attracts talent.”
August 23rd, 2007 at 9:56 am
I recently added a feature that I wanted to the Calander module. I have not seen any bugs witht he change I made reported and I hope that the changes I made were for the better. However I was surprised that I was granted access to SVN, in response to asking for it with no assessment of who I was and what changes I intended to make. I email everyone on the team of that module about what process / procedures / standards etc where in place and the response was basically “Whatever, if you could make it better that would be great!”
I did not want to waste anyone’s time, but I was surprised at the lack of “project management”. Perhaps I intoduced a bunch of bugs and features that only really make sense for the sites I plan to use the module on, and generally reduced the quality of the module and there was nothing to prevent those poor changes finding their way onto production websites using CMSMS.
I dunno what the solution is, but that is one of the problems.
August 23rd, 2007 at 8:28 pm
My suggestion get someone to write an o’reilly style cookbook !
I am very new to CMSMS (2weeks) coming on to an existing project which was running very late so my views may be just a little coloured - we are modifying some modules & writing a brand new one I have been struck by the differences in the way modules are coded and this has a huge impact on maintainability some are well abstracted & some are just plain shite - some use function.xyz.php where others use action.xyz.php to do the same thing - soem do everything in one block of php code (validate & display form - other break this into 2 ) - some some store templates in the datatbase some dont - some style everything with divs - others dont , and all seem to reflect a different viewpoint on how CMS operates under the hood
When we are trying to do the new module we have picked one of the better ones to use as guide as to how to do things even though it is in a different problem domain.
So I guess want I want to say is after v 2 is released STOP development for a few weeks except for vital bug fixes and DOCUMENT how CMSMS works and how bits interact with core - some simple state / block/ activity diagram (ie swim lanes) diagrams (as appropriate) are all that are required so I can see ok I need this kind of set of functions here, those sorts over there and they are connected by… etc etc - this would have saved us an enormous amount of time (note I could put nearly this same post on any number of open source projects forums - etc)
Having added my 2c - let me say I am converting all my sites CMSMS because it IS so extendable
August 25th, 2007 at 9:58 am
I have been working with various open source projects for the last four or five years. The problems encounter here are the same as those found elsewhere. Projects such as this one deserve to be supported and it appears that there are many willing hands to provide that support. The only issue is how to make best use of that support.
I looked at other open source projects to see how they dealt with the issue. The one that leaps out is OSCommerce with some 4,000 user modifications. The ‘mods’ process there is quite different from the one here (and elsewhere). In OSCommerce the mod is posted and then eveyone who has a bug cure or ‘upgrades’ can post the modified ‘mod’ on the same page as the original mod posting. (see OSC Contributions). The net effect is a give and take between the original author and the community. Moreover, if an originating author drops out, it is not uncommon for development of the mod to continue through the efforts of other community contributors.
My suggestion is to take a look at the OSCommerce model, which has been remarkably successful, and pattern the cmsmadesimple solution to the ‘mods’ issue on the OSC ‘contributions’ model. It works; its effective; and its simple
August 25th, 2007 at 3:38 pm
Open source is a wonderful concept. But the reality is that it is always a calculated risk to use an open source CMS for any client’s project. The real question, is why are web site designers and developers looking at open source solutions? Mostly to save money; to be competitive; to offer great solutions to businesses and non-profits with little budget.
To save money doesn’t mean that a solution as to be free – it just needs to be the right amount. Like a few others have said, other CMSs offer modules for a minimal fee. Most of the time, these are polished, well documented, almost bug-free, and with good support.
We all have to make money to live! I am completely supporting Franklin’s proposition (response #9) to create some funding pool per modules. I do not believe that it is against an “open source” concept – those who can afford to financially support a module development will understand that it benefits the majority of those who can’t. I will be one of the first to do so. (To tell you the truth, this money will ultimately be charged to my clients over one or more projects.)
To this funding approach I will just add one other suggestion: what about a thank-you in the form of a banner advertising — or/and service directory listing — for my/your services?
Keep the good work! I have discovered CMSms a few weeks ago and like many I fell in love with its simplicity and huge potential.
Michel
August 26th, 2007 at 9:42 pm
Increase Motivation == Increased Devotion to the cause.
Maybe the simple fix is to give developers more reason/motivation to make great modules.
Ratings and or COMPETITIONS with ‘awards’ for the module and/or developers might be just the thing to get them motivated.
e.g. Something like the Karma system in the forums.
People like recognition.
I look forward to “My module survived the CMSMS Quality Audit” t-shirts.
As has been mentioned:
A LACK OF an eCommerce module probably keeps many serious developers away… think about it:
If the developer can -easily- use CMSMS to make money, therefore offset their costs for developing on CMSMS, they might be more inclined to contribute more work to the system.
Oh and it’d be cool if there was an ‘upgrade’ button in the admin panel for CMSMS.
oh, and the CMSMS website and admin panel is starting to look a little dated.
Reflections people, Reflections!
August 26th, 2007 at 9:45 pm
On that last note, how about a competition:
Redesign the CSS for the website/admin panel.
August 27th, 2007 at 1:06 am
Well this message is a good reminder that I need to get of my lazy @#$ and get back to developing the module that I announced I would. I think it would certainly help to add some kind of inactivity timer to module development. Some developers (like me) just need a little encouraging reminder that they need to finish what they started.
Jeroen
August 28th, 2007 at 6:13 am
What an interesting thread this is!
I came to CMSMS at version 1.1. Over the weekend, I upgraded to 1.1.1. I’m currently working on two local test sites and one remote test site. The upgrade seems to be working fine.
The assumption I make is that 1.1.1 is the most up to date and stable release, so that’s what I should use. But when I read earlier on in this topic that some support requests relate to 0.13, I wonder if I’m right!
Surely what all of us who use, or plan to implement client sites with CMSMS need, is a stable core version and that’s surely 1.1.1? Is there any real reason for having previous versions available for download? Does this not detract from progessing the CMSMS project?
What we’d also like is Modules that have the approved stamp, because they’ve been tested and they work. This would be a great confidence booster in introducing new features and functionality, especially to clients. Modules without the approved stamp would remain “use at your own risk,” but again surely the goal here is for eventual approval? If not, there seems little point in having them!
Documentation is vital, but it’s a massive area. It seems to me there are three requirements. Installation, locally and remotely which is covered to some extent. Implementation - there’s got to be a best practices guide out there somewhere! User Guide, well this would be veritable treasure trove. I’d be delighted to collaborate and get involved on the latter two areas.
Thanks all. In my view CMSMS is just great and we can all help to make it even better.
August 31st, 2007 at 3:15 pm
Something that might help differentiate official from unsupported plugins and increase quality at the same time is is to make every plugin provide environment (e.g. dependencies and versions) and unit tests using simpletest or phpunit. If you look at most linux packages, CPAN perl modules etc you’ll find that it’s standard practice - if it doesn’t pass tests, it doesn’t get installed. If a plugin wants to make it onto the official list, it should have to provide and pass tests.
An ideal way to get this working is to provide a skeleton plugin that comes supplied with basic unit tests (e.g. to test functions required to actually operate as a plugin) and appropriate stubs for plugin-specific extensions. I know that strictness here will reduce the number of official plugins, but you can be pretty sure of their quality. Developers that can’t be bothered will remain on the unsupported list.
September 2nd, 2007 at 10:43 am
This is coming from a guy who develops about 50 sites per year… some are small Static HTML (Dreamweaver/templated), but I get quite a few mid-size sites and large/busy sites as well. I do e-commerce, real estate, manufacturing, insurance, banking sites… customers that need robust and reliable solutions. For paid CMS sites, I use Joomla, for nonprofit sites I use Joomla or Wordpress, and for personal sites I choose Wordpress. I also work on some legacy sites that are mainly static HTML with some custom PHP for page editing, forms processing, shopping carts, etc.
I’ve begun a few sites with CMSMS, but have been very frustrated by the lack of functionality of the addons. Because you can access the modules from the Admin interface, it appears to a newbie that they might be somehow approved/tested/up-to-date/functional, etc. Not the case, as we all quickly learn. It looked like a nice feature initially, but I suggest you move it out of the Admin and use something more along the lines of Joomla or Wordpress… where you search out addons, read user comments and ratings, then go to a dev repository or homepage site to read more & download, then go to the core and install/publish/integrate.
In the best of all possible worlds, we’d all have offline duplicate dev sites… but I will tell you that is not how I do it because I work in a NOC, so I am a hosting admin working on a local net, and can easily and quickly restore a site if it gets hosed. I always test addons on a playground site prior to live implementation on a client site, but I don’t generally do a parallel dev site except with phpBB, due to the need to change core files to add features.
Joomla has a huge installed base and communities that produce some top-notch components… I’m thinking VirtueMart, JCE editor, Hot Properties… I can use these for high-level sites and not disappoint clients with buggy interface, spotty browser compatibility, or major security issues. These are pay-to-play components, but they are worth it. Likewise the pay-to-play templates. Support is quick and readily available for the pay-to-play addons.
Wordpress is so simple to teach and use, perfect for content-heavy sites with a few plugins such as Event Calendar, SEO Pack, runPHP, FlashEmbed, RS Sitemap, PodPress, etc. If I have questions or issues, I can search out an answer very quickly or get some help on the forums.
I have not had the same success with CMSMS. I would consider using it and hacking in my own plugins, but then I might get bound to a certain version and have to skip upgrades. You guys are dealing with major security issues still, while WP and Joomla (and even phpBB) seem to have moved beyond major security holes. I’m not saying they are exploit free… but their upgrade notifications have been quick, and their upgrade paths have been glitch-free for me (for the most part). I’ll generally wait a few days after an update to see what problems folks who use my favorite addons are experiencing… just to be on the safe side and not break a client site. If it’s a major security hole, I will patch ASAP and be prepared with an immediate revert in case of problems.
I’m comfortable with Wordpress and Joomla, and they have served me well in producing client sites. I had great hopes for CMSMS, but I’ve always run into problems early during a site build, and have had to jump ship and move back to WP or Joomla because I know I can find gazillions of good templates, plugins and modules… and I know how to backtrack when something goes wrong.
I’m also quite familiar with phpBB, so I know it can take a ton of coding & server optimization to get a site properly configured and functional… so I really don’t have an issue with that level of involvement with coding/upgrades/backups. However, I choose the ease of upgrading Joomla or Wordpress, and the speed at which my favorite components/plugins/modules get improved/normalized for new versions of the core.
Bottom line, I will continue to use Joomla and Wordpress for CMS sites because the support for modules/plugins is great and because the user communities are plentiful and helpful. I’d truly like to use CMSMS due to the integration of CSS and the quickness of CMSMS sites compared to Joomla… but I need something that is a bit more stable and well-supported.
I suppose if I became more active in the CMSMS community my problems would evaporate. I will commit to re-doing my personal site with CMSMS and if it works for me there, I will begin using for client sites.