Home arrow Compare CMS - 3b
Compare CMS - Part 3b- Business CMS Reviews



Compare CMS - Part 3b - Business CMS Reviews

More Business CMS Compared

Here are more business and capable CMS reviews, including a training management application.


WebGUI
Moodle
RadiantCMS
More about RoR

Need to ask a more detailed question? Try the new CMS Forum.

[please be sure to read the disclaimer at the end]


WebGUI

Making a CMS choice based on a feature matrix is not necessarily the best way. WebGUI - which I would say as "web-jee-you-eye" but they want you to call "web-gooey" - has the most features of any CMS, out of the box. In any feature comparison, it knocks everything else stone dead, because it appears to outclass every other CMS by miles. On a feature matrix, as virtually every box has a tick for 'yes', when even Joomla (the most capable CMS existing in terms of jobs it will do) appears to lag well behind it, it seems to be the Best CMS In the World. Due to its apparently amazing capability, WebGUI generates a huge amount of interest. In practice you will probably be disappointed, because the hosting requirements are tricky to comply with unless you are an enterprise, with a dedicated server and physical access to it; and the core features / plugins may not do what you need.

Note also that PlainBlack, the WebGUI authors, run one of the CMS feature matrix comparison sites, so a Webgui CMS comparison looks pretty good there.

On the other hand, you could view this as a hosted solution - and on that score it works well. By this is meant that the developers also host your CMS if required - and if your usage patterns and budget line up with what's on offer, then it's hard to fault this arrangement.

As the more usual format of user-hosted CMS, though, there are some disadvantages. Firstly, you have to have physical access to the server, so even a dedicated server will not be sufficient. You must compile the CMS on Linux, meaning that there is no one-shot installer, various operations have to be carried out. Few hosts would do this (they wouldn't know how in any case).

The CMS is open-source, which is always a major plus point. There are many advantages to this, and no disadvantages. However, the extensions are often commercial, and so we could regard theis CMS as a semi-commercial solution. This is especially true when you consider the hosting requirements.

You might initially see the clear feature superiority as an advantage, but there are too many negatives. In the end, you'll find there's nothing wrong with a bare CMS framework to hang plugins on, since it means you can swap them out. Here, there aren't really enough alternatives to allow that. Many of the extensions are too limited - the shopping cart for example [though we looked this in 2006 and haven't checked back since]. If you need better functionality, you pay the chief dev to work it up for you. Fine - but this gets one or two guys' input, a full-OSS solution gets the best around the world. The developer will not host instances with extensions that were not originated by him; but this CMS will probably work best as a hosted solution, with the developers. It is therefore probably not the best choice if you plan on implementing it yourself. T
his is the one of the few webapps I know that's coded in Perl.

A Windows installer for LAN trials, or for in-office team use, is also offered. Don't even consider trying that, though. Installing it on a local dev LAN on a Windows box will be the most painful experience you have ever had (unless they have recently fixed a large number of errors). It doesn't work, and won't work without a lot of core hacking - if at all. Why on earth they supply a Windows version I just don't know. I gave up after I'd hacked 5 or 6 files with completely wrong filepaths and still had no working result. You have to be a Perl coder to get it working on Windows. You'd think that at least one person would be asked to test it.

Somebody let us know if they
fixed the 'Windows Installer' yet. Compile it on the penguin or don't bother. Also, it scrambles your PHP filepaths as well - worse than eZpublish in fact - so just don't put it on a Windows box. I didn't put it on my Linux box either after I saw what it did to the Windows one. That had to be reformatted in the end, to fix it - luckily, only a half-hour job with Acronis TI. All hail Acronis!

This CMS will suit those who want a hosted solution that precisely fits their requirements, as it comes, out of the box, and most likely without any additional mods. It is an enterprise level CMS because it has the core functions needed to qualify: ACL, versioning and so on.

It has excellent URLs and lots of features; but to be more popular, it would need to be easily installed on a LAMP server - not scramble script filepaths - be more flexible. Currently, it ideally needs a dedicated server, or at least one where all other CMS are WebGUI. 

It's interesting to compare this app with eZpublish, as they are both semi-commercial CMS: part-OSS, part commercial, with the core free but the extensions normally commercial; and both have some scripting / filepath issues for a server. Somehow, eZ got it right and WebGUI hasn't quite made it yet. Perl must be a very capable code for webapp dev, and compares well with PHP, because WebGUI has many more core features than any other popular CMS. Other pros and cons are for devs to argue over.

This is a mature project now and looks better than ever for enterprise use, on a dedibox, where experienced Perl support is available and can be afforded. Other users are less well catered for.


[please be sure to read the disclaimer at the end]
 

Moodle

Strictly speaking, not a CMS but an LMS - a learning management system. It is the best-known CM application for Internet-based learning (or LAN use of course), and is aimed at managing courses and training online.

There are some similarities in the Moodle organisation to eZpublish; and possibly Mambo, as this is also an Australian-based operation. Like eZpublish, the core application is free / open-source, but extensions, integrations and support are mainly commercial. There are widespread support options and an in-depth history here, as this is a mature project.

The two main sites are moodle.org and moodle.com, but don't go to moodle.com - it is basically a dead-end (there are either a raft of usability issues here, or alternatively the information you require is deliberately obfuscated for some reason. For example, there is no link anywhere on the site to the main website at moodle.org.) First experience of the project at moodle.com might well lead to you giving up; but moodle.org is the place to go. Moodle.com is the software authors' site, and clearly they support and protect their commercial partners as far as possible. As mentioned, this has similarities to the eZpublish model.

Moodle is a PHP app that runs on Linux or Windows, using a MySQL database. The LAMP installer is 12MB, the plain Windows box installer is 60MB and is integrated with XAMPP. From this you can probably deduce that it will not run on a bare IIS server, only one extended with a full PHP - MySQL installation. There is a MacOS installer at 81MB, integrated with MAMP. All sizes quoted are for the zipped installer version. There is a large range of plugins. It should be noted that XAMPP installations are not advised for production, by the Apache developers from whom it comes - a full IIS installation is preferred (i.e. Microsoft Windows Server 2003 with IIS). For LAN use, though, this is ideal.

Because of the semi-commercial nature of this project, information is a little harder to find than usual. The forum is a good source of information but for registered members only.

Moodle is a sophisticated application for online teaching and course management, with a good number of plugins and a smooth installation sequence. For DIY use it would be inexpensive, though all advanced operation will be of the commercial variety except for users who are themselves expert implementers and developers.



RadiantCMS

An interesting new contender for the 'New and Better Codebase' trophy, Radiant is a Ruby on Rails CMS. The code is Ruby but the pages come out in very, very nice HTML. Radiant has excellent standards-compliant pagecode, based on divs as it should be. The code validates 100% (when set up right, which few current sites are, though that is normal - user templates cause the problems), and the whole thing looks good.

This is a very new project though, and probably not ready for universal Internet-facing duties yet. As an example, they don't have a web installer yet, your host has to install it. Hopefully they won't be staying like that as it would mean there will never be any real community, people with physical or even shell access to their servers are a minuscule fraction of the user base, you absolutely have to have a slick HTTP install routine or the project will never go anywhere. How many sites run WebGUI? That is a mature app with a big rep, but few are using it due to the install issues.

It's true that enlightened webhosts (like Nativespace) already have an auto Ruby webapps installer built in - but how many hosts are enlightened? And in any case, the simple Ruby installer that hosts employ is unlikely to be able to handle a CMS as there are a bunch of configs to do, to get it talking to the database for instance.

Alternatively, this CMS could be used as a collaborative workspace tool - that is to say, a team cooperation worktool for internal office use, or on an intranet. In this case, installing it for LAN use, there would be access to the server so that a compiled-installation routine would not be a major issue. [qv CMS terminology for an explanation of this]

There is a way to go before this is a user application (i.e. for normal CMS implementers and sysadmins / webmasters). It's purely for developers at the moment - fine if you're a Ruby dev, and in fact this looks about the best application of it we've seen. It will suit if the owner has a dedicated server, or for LAN team use. Radiant will do well to compete with Plone, however, which is a one-click install on a Windows box, and a first-class office team CMS with ACL and everything else, out of the box. It looks as though Radiant will compete by being easier to manage, and to write extensions for. Plone doesn't score well on either of those points.

Documentation is, ahem, not a strong point. They need to get two things sorted out quickly if this project is going to go anywhere:

1. A slick web-based installer.
2. Some user-centric documentation. This means help for ordinary mortals, not geeks.

As an example of what needs doing in the docs department, take templates: at the moment there are some vague ramblings that pan out roughly to "You can have any template you want." This, actually, is of little help; we'd like to know precisely how to build one / get one / install one / modify one. And so on.

Some clever people working on this one, but they need guidance from someone in the real world. First-class page code, which is a pointer to the quality of the app; and if you're a Ruby dev you might be able to get somewhere with it.


Compare Joomla v Radiant
A pointless comparison because they are for different jobs; and Radiant [at 2008-01] is where Joomla / Mambo was about 2 years ago. Also, at present a team tool for office use, not for Internet duties except as a developer-run specialist application. You have to have root server access, which means the 95%-plus of webmasters who don't are excluded.

It has tremendous promise, though, as the devs obviously have the foresight to see exactly what's needed - there are not many other CMS around with perfect standards-compliant HTML pagecode based on divs - CSS, as every CMS everywhere should have.

Right now, there's still a lot to be done - you'd need to get a developer in to do just about everything, and that's not the definition of a usable CMS. It needs a web installer, a basic templating routine, plugins, and documentation. At that stage, it would probably fly: webapp devs using RoR would be able to create plugins on a wide scale.

Like most CMS of course it does not have extensive functionality, out of the box. This is not the point, as all CM systems are extended via plugins. There are many available now, and no doubt in the future there will be a large number, as the Rails development framework seems to have so much promise.

Radiant wishlist

I can't tell you how much I would love to have a web installer for this, so that it didn't have to be compiled on a local machine. Santa Claus?

More about Ruby on Rails

Because I don't want to be seen as Ruby-bashing, due to the somewhat negative review of RadiantCMS, here is some more material on Ruby. It is unfair to criticise because the code only dates from 2004; so how can there be any kind of complex community applications as yet. Not to mention that the only good app is a mature app.

Radiant, though, deserves a lot of support because of the superb page code. This may seem like a simple matter, but if you thought this you'd be wrong. My daily work involves repairing bad development by people who can't write simple valid page code; who never heard of web standards; who think that it is still OK to use cells and tables as the main page architecture when they have been obsolete since 2002; and who never even heard about validating their code. To find nice clean page code like this from a CMS is a breath of fresh air.

Rails enthusiasts tend to be very keen on their chosen codebase and its strong community. Fair enough, if something's good, promote it. We would like to see a good, working, community-usable result, though. This is unfair to a codebase / framework that only came into being in mid-2004, of course, as the shortest gestation period I ever heard of for a working CMS with most of the bugs out and enough support was three years.

It's said that RoR is the best framework for web applications that exists; as powerful as Python but better for webapps. It has been suggested that small to medium-size applications are its home turf, rather than large implementations; and this wouldn't compare favourably, in that case, with the Python-based Plone / Zope CMS class, which handle millions of pages and mega loads. Perhaps this only reflects current usage, though. Of course the main business for CMS in any case is for sites of between around 50 and 1,000 pages, there are far fewer big sites. It's supposed to be the #1 dev framework for database-reliant webapps, and have the fastest dev time of any comparable code and framework.

More links for portfolio sites and related sites are included here than we normally show. There are good reasons for this:
1. I have had to be a little negative about Radiant since it is not a community-use tool as yet. I'd like to make up for that somehow.
2. As a tired, bitter, cynical old SEO consultant there are few opportunities to be enthusiastic about anything much; but it would be a pleasure to work on one of these Radiant-based sites. That is, if half my work hadn't already been done...

Example sites:
www.redken.com
www.indiaandmore.com
www.hawkinscosmeticdentistry.com
www.uneins-magazin.de
www.ridgelinemediallc.com
[an interesting exercise in how to make Radiant look like WordPress?]
 
Some Radiant-related sites:
http://www.artofmission.com/articles/2006/06/06/radiant-cms/
http://wiseheartdesign.com/
http://lists.radiantcms.org/mailman/listinfo/radiant
http://notepad.onghu.com/2007/5/26/hello-world-using-radiant-cms
http://notepad.onghu.com/radiant
http://lists.radiantcms.org/pipermail/radiant-docs/2007-December/000034.html

This looks useful: an instant RoR / Windows LANserver (like XAMPP, uses Apache, said to be a one-click install for a Windows box):
http://instantrails.rubyforge.org/

Here are some RoR links:
http://www.onlamp.com/pub/a/onlamp/2005/10/13/what_is_rails.html
http://www.rubyonrails.com/



PLEASE NOTE: these critiques represent an entirely personal opinion. They are personal reviews. There are some negative views expressed here that are one person's opinion and may be entirely wrong. There are positive opinions here that may be equally wrong. There are obviously many people who are entirely satisfied with webapps that have been criticised in some way, and you should ask some of them before taking this material at face value. There may also be those who are unsatisfied with CMS apps, or aspects of them, that have been praised.

YOU MUST TRY THEM AND MAKE UP YOUR OWN MIND.
 
Web Business Managers