Slashdot Mirror
E2 and LJ, Comparing Content Management Systems
Anonymous Noder/LJ'er writes "Linux.com is running a story written by Slashdot's Krow, one of the authors of Slash comparing the LiveJournal site engine to the Everything2 engine. He went over the installs of the two engines and talks a bit about customizing both. I really like both sites so it is interesting to see someone talk about what makes them tick."
You couldn't plan that better!
-- @rjamestaylor on Ello
(p.s. do I get mod'd up for posting the first real comment?)
"I say we take off, nuke the site from orbit. It's the only way to be sure."
Comments about administration policy asside, I think that comparing the two isn't strictly fair. At least, not on absolute terms.
As a developer on both codebases, teh differences as I see them are basic: Ecore is about grouping and linking as sets, while lj is more about mass indexing of list-type data.
One of ecores main weekness is scalability, or lack thereof -- this is not a slam on teh code, but just an introspection on design. Because lj os more about this loose-linked list paradigm, it can easily scale and cluster on mutliple machine while ecore, with it's extreme data interlinking, is heaving right now with redesigns to allow that.
Of course, Ecore (or at least E2) has a much better XML interface, which is probably it's second strongest point. It's first strongest and most important is the concpet that everything is a node.
Hilary Rosen's speech was about her love of money and her desire to roll around naked in a pile of money.
Here is the original WikiWikiWeb: http://c2.com/cgi/wiki?WelcomeVisitors
Here is a Wiki you can easily install on your own machine: http://minnow.cc.gatech.edu/swiki/15
Here is a free Wiki farm that let's you start your own on a shared server: http://www.seedwiki.com
I have been looking for a web-based system to manage personal data. I don't mean calendar, contact list, etc, but more general things.
Basically I need a replacement for a filing cabinent that will let me upload documents (and maybe grab copies of web pages), store them in some sort of semantic web/catagory hierarchy, annotate them, replace them with updated versions etc. Ideally, the system would work with a wide variety of data types: pictures, pdfs, text files, html/xml, word documents etc.
Is there a system avaialable that does this sort of thing?
It would have nice if they included installs of Postnuke and/or Phpnuke. I've installed slashcode, phpnuke, and postnuke with phpnuke probably taking the cake for the easiest to install.
I wrote my own journaling program (blogger system). It's based on html forms and MySql, and is written entirely in perl. I never really intended for it to be much more than an exercise to learn a little MySql. It's basically done with one major exception, security. I need a decent way to handle remote passwords. Currently, I use a combination of AuthUserFile/deny/allow in .htaccess to limit who can make changes. I need to implement a better system, but can't decide the best way to go about doing this.
-- Will program for bandwidth
Any comparison of software like this should really talk about performance and not just installation and administration ease. Any idea if these systems can be sped up, and what else is out there?
The major downside to it (which seem to be common to most things in this area) is a lack of documentation.
Having looked at several CMS for a website I am going to relaunch, I needed an approach that was most general and would allow me to choose how I could store the content and separate the design. OK, the difference to the projects mentioned here is that I don't need a large system to manage things like user comments or other methods of dynamically adding pages. Currently I think I will go with AxKit, which is not really a CMS, but basically an on-the-fly caching XSLT processor. For me this provides the most flexible solution. I have designed an XML format to store my content files, and can then use either an XSL stylesheet to produce HTML or WML or whatever needed, or write an XPathScript style sheet allowing me to process the XML data while additionally using perl code to add dynamic features. The nice thing is that with AxKit you can use HTTP GET parameters to allow different style sheets (plain, xhtml, print, ...), pick a style sheet depending on browser type (lynx, netscape4, mozilla...), etc. And for a website offering mostly static content that needs to be organized in a proper way, I think separating content from layout using XML seems like a good idea. What I like to is that it's easily possible to include multiple language versions of your page in your XML data files and transform to HTML based on, say, a ?lang attribute. Plus, you could even store the XML content tree in CVS...
For websites that are just trying to be in control of their mostly static content, AxKit surely helps (provided you have access to the server box as you need to install the apache module...). Storing pure content as XML and then providing different stylesheets for layout seems a proper way to go for me.
Of course, this is not to say I don't like the LJ or E2 engines, butjust depending on what you need for your website, XML might be helpful, and AxKit might be the way to implement it.
www.typo3.com
Some really cool features: (Stolen directly from typo3.com) I've started to use it for a couple sites in the last six months, and its really made web development fun.
-Pete
Okay, you can sit around with your TI-86 all night long and talk about XML parsing times, but who cares? The only real test of a site is how much it provides content, and by content, I mean a peek into the life of CowboyNeal. Using this test, let's compare:
CowboyNeal on e2
Versus
CowboyNeal's livejournal account.
Aside from technical details, which one of these gives us more insight into the delicate poetic soul that is Mr. Pater?
Hopefully I didn't put any [] around my words.
If you are into rolling your own, then take a look at the Look at mod_auth_mysql Apache module. It's basically .htaccess file kind of access control except the user info is in a MySQL DB. So you can do updates/inerts/whatever on the database via your perl and get close to what you need as far as access control without having to write files in the docroot.
You might not be able to make it fine-grained enough, but if you have a thing where each user (for instance) gets their own directory or something then it might work pretty well for you.
And if you are not into rolling your own anymore, check out Moveable Type.
-B
Ash and Hickory, straight-grained and true, make excellent bludgeons, dandy for the cudgeling of vegetarians.
MS Sharepoint Portal Server is the a next round in MS's binding of the corporate office bureaucracy to them. It's basically a Web CMS and DMS that fully integrates with the rest of MS Office.
It's a pretty damn poor Document Manager, and a really abysmal Content Manager in most respects (except for - again a killer feature - WYSIWIG page editing inclusive of component embedding), but the MS Office integration is the key. And of course, no-one can integrate with MS Office better than MS.
If there was a decent "Open Office Portal Server" then things would be just dandy - but, as it is, Sharepoint will act to lock people into another round of MS-dependency. Sharepoint Portal Server has been used by people talking to me as an argument to stick with MS Office even with the existence of open office and star office.
Choice of masters is not freedom.
Also, on a somewhat serious note, e2 isn't about the technical interface...not that it wasn't nice for Mr. Nate et al to give it to us. However, e2 really succeeds as much as it does for the human bonds it encourages. Or in other words, what other web community can gather 100 people from across the continent to have a scavenger hunt?
Hopefully I didn't put any [] around my words.
Sharing INFORMATION is something different than sharing worddocuments. However a lot of people tend to think that 'information' should be stored in worddocuments, which then have to be harvested into databases and / or distributed by a CMS on top of that.
Why on earth would anyone start to hammering in texts into a worddocument when there will be no paper-version of the document, ever? There is no need for a wordprocessor to share information, on the contrary.
Never underestimate the relief of true separation of Religion and State.
Seeing it in action there, and having used it there for almost a year, it has convinced me personally, that E2 is the way to go for this type of content management system, but YMMV.
One of the assumption that's being made both in the story and in many of the comments is that the CMS runs on a dedicated machine over which you have total control. This assumption covers database formats and filesystem layouts that are unfriendly to multiple installations, plus custom Apache configs (even an Apache dependency is noxious IMO) to "just install mod_obscure_widget" pseudo-advice.
The problem is that the assumption is just not true for many people, who run their sites on servers owned, configured and administered by a hosting provider. Any CMS that can be installed in such an environment automatically becomes about five times more useful than one that requires total control of a dedicated machine. That's the review/comparison that would really be useful to most of us.
Slashdot - News for Herds. Stuff that Splatters.
One of the reasons I think most CMS systems seem to be slow (on the client side at least) is because they insist on using tables for everything. You end up with tables inside tables inside tables inside...
Most browsers take inordinate amounts of time to render pages with this kind of structure. Much cleaner to use and , especially for tables consisting of one cell which many of these systems generate.
comparing the LiveJournal site engine to the Everything2 engine.
This actually is one of my pet peeves. CMS is a larger term than just "web site management system". CMS need not have anything to do with web site management; it may act as backend system that may include authoring part, workflow management.
This is probably also why it seems all commercial CMSs have big problems (I work on a project that uses one and we have plenty of issues. That is, if they are "just" web site management systems, geared towards web design, they should be marketed as such. The company I work for bought a reasonably pricey CMS with some expectations, and then developers find out it's only glorified web management system. The irony is that not only are we rewriting much of existing functionality, we are not even using most of 'advanced' functionality that is mostly related to actual web publishing (in our case CMS is not the front-end system).
"Full" CMSes should probably concentrate on having complete robust platform for developing actual applications, which can then drive web sites or other publishing (often publishing to actual front-end systems, not being one). It would be good to have reasonable interfaces to actual publishing front-ends (web servers etc)... but it shouldn't be too tightly coupled.
I like paying taxes. With them I buy civilization -- Oliver Wendell Holmes
Anyway, for e2 and livejournal, that wasn't the nature of the slowness that I notice.
Is there some reason e2 doesn't use InnoDB?
I use phpWiki and you can put a link LikeThis or a link [Like this]. You could even put a link [with arbitraty text but poinint to a keyword|AnotherLink].
...
It really is easy
unfinished: (adj.)
I agree. If you think a CMS is the same as a blog, then consider the following questions:
1. Can the software support different media types and still keep all the material searchable?
2. Can the software support future media types, i.e. media types which are currently unknown? By support here, I mean store, retrieve, present, and make searchable.
3. Can the software support versioning of media items?
4. Can the software support multiple levels of security on media items?
blah blah blah...
A blog is just for text, and maybe some graphics. Try running a real company on just those.
Please mod this post only if you think others should/n't read this. I have enough ego^H^H^Hkarma. Thanks!
Hey, have fun! A really useful CMS is hard to build. I hope you've nailed down your requirements really well with your future users because that sucker is gonna snowball otherwise!
.NET, if you have the option. Yeah, it's a learning curve, but since you're on the Microsoft wagon anyway, you'll be better off in the long run if you bite the bullet now.
I would do myself a favor at this point though and go to
Best of luck!
Please mod this post only if you think others should/n't read this. I have enough ego^H^H^Hkarma. Thanks!