Slashdot Mirror
Content Management Nightmares
bildstorm writes "I've recently been looking into content management systems for my company and have found that there are all kinds of systems out there. I've found that most Americans consider web content management to be the catch all for content management (like Interwoven). In Europe, I've noticed that what's referred to as digital asset management is what is usually meant by content management (like Artesia).
Has anyone used any of these systems well? For more than just web content? Has anyone tried any open source systems and used them well? I know there is a conference in Zurich next month for open source content management, but I don't know much about the products."
Favorite things about Zope:
this is getting old and so are you
blog
This is becoming a larger and larger issue among companies. I know of one company that is spending close to $3 Million US to get their content organized. In addition, I think there is also a movement to integrate Content Management Systems with Learning Management Systems. I guess this is a good idea.
Kris
...to set up and configure, especially (gack) on Windows, but once it is running it is an incredibly powerful tool. Currently running on Solaris, has Linux and HP-UX variant as well. Eminent customization and power comes with a price though. A very high one at that. Have looked at a promising app by eGrail that seemed to be competitive and at a fraction of the cost, but wasn't quite ready for prime-time (this was a year or so ago, things might have changed since...)
Brain: Promise me something, Pinky. Never breed.
There are heaps of these bloated systems. For a while, Vignette's system (formerly called Story Server) was a leader. Many outfits build their own, for example, based on Oracle. A colleague has recently installing Microsoft Content Management Server for a large government client, and he has been remarkably impressed.
weeell.. the first thing you need to understand is that some of these content management systems are really toolkits, some are more out-of-the box experiences... its kinda a spectrum.
my opinion - beware the hell of out of box stuff, (like red dot), you wanna budget about 50/50 buy vs build (or, better still save half your budget and use an open source system)
the open source alternatives, arsdigita, midgard, Zope Content Framework, are really every bit as good as the mid range CMS systems, but if the bureacracy is gonna wanna spend 400,000 dollars on a CMS systems like Vignette (bleech!) then nobody's gonna stop them.
<not a troll, no really>everybody, of course, is keeping a damn close eye on Microsoft, and their systyem is really shaping up, i gotta say, (if you like that sort of thing </not a troll>
if you want more, good info, check out cmswatch.com and *the*, definitive cms-list
..is implemented with a magic program called, "bash" and few of its friends.
I can search by content (grep). I can search by date (find). I can filter to other types (sed). I can assign hiearchical meaning to records (mkdir).
I can even assign meaning by fields with the '=' operator, source the file with '.' and deference environment varibles!
Works great. I've implemented a change control system, a BBS (threads, fancy search engine, and all!), a user management system, a product management system and a bunch of other cool things with it.
Right now most of those systems support formatted plain text and html as output. But I could add an XYZ module with almost no effort.
So it's not secure. Nobody said anything about security.
Do daemons dream of electric sleep()?
..I'll admit, I've never used it, but just from reading about it, and knowing who the main architect is, it seems like a pretty decent product.
:CityDesk
Check it out here
I lost my concept of community when my community lost all concept of me.
from nxn claims to be a good system for keeping track of "assets", i.e. all those pictures etc.
content management is always a sticky subject. Not only do you need a system that manages content but need to oragize it such a way that makes it user friendly. Coming from a company that has their own content management system for clients, the organization of content is just as important.
is what we use here. And I'm actually the one in charge of it.
A few things to make note of:
1) it's a good product, AS LONG AS SOMEONE QUALIFIED INSTALLS IT. Our installation job was completely botched by the company that did it, and it ended up being practically unusable. We had to hire contractors to fix it. Whatever software you end up choosing, make sure someone certified by the company installs it. It's more expensive up front, but will save you endless hassles and cost much less in the long run. For god's sake whatever you do, don't assume it's just like installing any other software and any bonehead can do it. It's just too complex for that.
2) For whoever will be managing the software: either hire someone certified by the company, or send the person who'll be managing it on as many training courses provided by the company as possible. The more they know, the better. For interwoven, a knowledge of PERL, XML, DTDs, and some sysadmin type capabilities are a must. Familiarity with JAVA is a definate asset.
3) TeamSite is a great product for straight ahead, content management, but if you want any bulk functionality, you'll need to do extensive customization. It's meant for one-at-a-time changes. A good PERL programmer will save you a lot of headaches in this area.
4) $$$$$. Any good content management software is going to cost you through the nose in training, installation, and the software itself. Expect it, deal with it. Make sure the marketing pinheads know it.
5) Get the tech support, you'll need it.
6) TRAINING TRAINING AND MORE TRAINING. Make sure the editors take at least a basic training course in using the TS GUI, or your manager will spend 95% of his/her time fielding calls from frustrated content editors who don't understand what a DCR (Data Content Record) is, and don't know how to unlock a file.
7) Last, and most importantly, install it on solaris. Do not, under any circumstances, install it on WINNT. Gah.
There are a lot of good resources out there for TS. It's a popular product, and I'm on a few mailing lists that are quite helpful.
If you have any questions about TS, you can email me privately and I'll do my best to answer them.
Moral indignation is jealousy with a halo - H. G. Wells
we just use CVS to manage everything. It may not have the scope of some of the larger systems but we all know how to use it (and there are purty web frontends ;))
after spending close to $150,000 on an interwoven set up, we finally gave up the ghost and ripped it out last month.
it has got to be one of the worst, most tempermental services i've ever used.
the breaking point was after asking the IW consultant "how do we make and manage templates?" and having him reply:
"you can't. you have to call us."
we're now back (and quite happy) using out custom setup based on VSS, ASP, SQL Server and FTP via PERL Scripts.
(pretend there's something witty here)
I have been using Incyte for project management, seems to work pretty good.. and of course, it is open source.
anime+manga together at last.. in real time.
Building a content management system in Lotus Domino is not too complicated a task for the average notes developer... Depends on what you're looking for...Workflow type stuff - domino is perfect for that...example - my company's web site is hosted on domino, any changes someone makes to it must be approved my several people and then the site is updated...
For small companies/small needs (not a lot of servers/content) I use simple Lotus Notes applications. For larger needs, I step up to Domino.Doc.
Security, multi-platform support, workflow, automatic versioning, support for any number of content types, it's all in there.
I looked at some other web content management systems, but the prices for Interwoven and similar were 6 figures or more. On a smaller scale, I was pretty impressed with RedDot, but still reverted back to Lotus Notes for product maturity and cost considerations.
If someone says CA's CCC/Harvest in front of you, RUN, don't walk RUN!!!!!
Why dont you try the POSTNUKE system? http://www.postnuke.com Postnuke is GPL and very mature. It's PHP and SQL (lot of sql kinds)
Fast forward to 2002. Most of the document managment vendors from 1996 are gone. Now we have "content management", which seems fine as far as it goes but also seems (IMHO) to make the basic assumption that everything is, or will soon be, a web page. Management of plain old documents on plain old file servers (SANs now I guess) has been forgotten.
Hello! Not everything is content! Not everything will eventually become a web page!! Would the "content management" vendors please remember plain old business documents?
Thanks.
sPh
phpnuke. then theres oma - for media management ( realvideo, mp3, realaudio ). zope i heard is also nice, but i dont trust it under heavy load.
there are loads of php/mysql related content management systems out there just look at freshmeat.
one thing about PHPNuke - it's not really GPL, but you get the "source" with the download. one guy is the author of it and he's happy if you buy something for his amazon.com wishlist. You can customize it to pull of a site where your users can post articles, post comments as well as a site, where your staff posts articles, and users cant login / post comments. there are loads of "plugins" for it - galleries, and what else you can think of.
just remeber to get this guy some cash if you use it commercially.
Most people have a problem defining 'content', or even 'information'. When I see an organisation struggling with their enormous pile of word documents full of sometimes vital information, I know there is something basicly wrong: the start of the storage of a bit of information (or better: a bit of data which can be interpret as 'information') is wrong in a lot of organisations, therefor the usability of this information is limited at best: to use it in expert systems, in general documents for print, in websites and f.e. in general database applications, it has to be extracted from the worddocument by external tools, which is not that easy in most situations.
Some organisations try to use a 'content management system' (CMS) to transfer their pile of worddocs into data inside the CMS, which is then usable as 'content' for websites. However, this process is difficult and error-prone, and the end-result is not what most people want to have but another collided form of the data which was once stored into a huge pile of worddocs.
Here in Europe you have a lot of different CMS's. Some large ones try to grab a lot of external data and 'publish' that on websites, mostly by offering worddoc/office document importers, others are build around 'data' and stick viewers on pieces of data, which can then be used in websites or anywhere else. How I see it is that there should be a general base of data-elements which make up the core base of data-elements for an organisation, which is used in all kinds of systems that use that data, including viewer applications for websites. My CMS (CESys) does this, also others like the Open Source CMS MMBase follow this approach. I think that's the way to go: it forces organisations to think about HOW to store data and how which data is used, instead of keeping organisations at the level of "when you want to store information, open word and start typing". Because: webpublishing is just connecting a viewer and a piece of data to get viewed by that viewer and with the proper storage of data it's an easy job to do.
Never underestimate the relief of true separation of Religion and State.
I'm posting this anonymously because of the usual reasons related to my employment ...
...
Another company which delivers a content mangement solution which integrates nicely into a web publishing and management environment is Blue Martini Software. I've used their system on a few projects and it does a good job of managing content and pushing out new 'releases' when you have something new to publish. BMS currently drives some rather high-traffic websites, which if nothing else, proves their scalability.
Yes, BMS in commercial software, but quite good at what it does; it also includes a sophisticated API and a (supposedly, haven't worked with it) quite good Data Mining tool, all of which is nicely integrated
Content Management Systems are golf-course-ware. It gets sold to senior executives by smooth-talking sales executives who claim their products solve every conceivable business problem, is a doddle to install, standards compliant, holographic user interfaces, everything.
The reality is that this is an inherently complex field, which requires a huge amount of business-thinking before the technical solution even becomes relevant - how do you want to manage your content ? Do you have a requirement for workflow-style solutions ? Is revision control important ? Do you need collaborative features allowing several people to work on a document at the same time ? Do you have a knowledge management infrastructure so you can re-use an accepted taxonomy ? What are your security requirements ? Where does your content reside - is it largely "document" based, or is it mainly database-driven ? How technically sophisticated are your content generators ?
Only after you have worked out what you want to do with your precious content should you consider what the technology can do for you - I suggest using any one of the myriad requirements gathering techniques used in software engineering and specify your "ideal" content management system, then drawing up a list of candidate technologies.
If you start with "what can the technology do for me", you almost certainly will end up spending a lot of time and effort (and money !) and getting very little in return....
It's all very well in practice, but it will never work in theory.
A while back, our company was looking into getting some content management software as well. This stuff (Hummingbird) looked really cool, but our small (less than 50 people) company couldn't validate the tens of thousands of dollars we'd have to spend to get it, not to mention the several thousand per year fee to continue using it.
You really have to ask yourself what you want. These software packages are REALLY expensive--sometimes more expensive than the annual salary for a new-grad CompSci major. Do you need all the features of a full-blown content management system, or do need something that someone at your company could dedicate some time to and write?
by getting my work done on time (which means I should probably read slashdot less).
The MS CM version today is a souped up NCompass version, since MS bought NCompass and changed the name into MS Content Management Server. It's ok, but page-focussed (not good IMHO, since a page is a collection of data-elements that are VIEWED by the page but are not part of it), and quite expensive ($35,000.- per CPU)
Never underestimate the relief of true separation of Religion and State.
I like ezPublish.
sPh
There's a great new product on the market by a company called InStranet (www.instranet.com) that uses an interesting approach to organizing the content; instead of securing the objects in a hierarchical fashion using ACLs, etc., they employ a multi-dimensional framework to organize and secure the content. It uses the same technology as a lot of business intelligence applications on the market and is far more robust than Interwoven or any of the other behemoths on the market. Something to keep in mind is that a lot of products like Interwoven et al. utilize a proprietary repository for storing the information. So, you can only access your content through their application. InStranet uses Oracle or DB2 so you have greater flexibility in customizing your app. Much better choice if you don't want to get locked into one company's app.
All the packaged content management systems suck,t hough. After a certian point, it's easier just to pick some decent base libraries or app server platform and roll your own. It will do what you want, you can see your source code, and you might be able to make ungodly sums of money later reselling it (Vignette).
For newspaper / magazine / journal / news sites, consider PROPS. PHP/MySQL/GPL.
As we're all experts here, I should point out that content management seems to be just a new buzzword for boring old configuration management with bells and whistles on.
You might therefore want to consider a configuration management system (CMS). Some of the CMS vendors relaunched their tools as content management systems during the dotcom bubble. You might want to look at them. Continuus (now Telelogic) did this, for example. And, of course, you could take the cheap and Open road and use CVSNe mæg werig mod wyrde wiðstondan, ne se hreo hyge helpe gefremman.
Correct, Zope is more of an app server. CMF (Content Management Framework) is a plug-in for Zope. Considering it's all free for the taking, anyone interested in content management would be foolish to pass up the chance to evaluate the Zope + CMF option.
But has anyone here used it? What's bad about it? What's good? What else is better?
-B
Ash and Hickory, straight-grained and true, make excellent bludgeons, dandy for the cudgeling of vegetarians.
for archival/project/licensing solutions you can look at eMotion's MediaPartner:
http://www.emotion.com
go to www.gaussinterprise.com
content/document management
CityDesk by our beloved Joel Spolsky seems like it may be powerful enough and quite easy to use (if it lives up to Joel's standards as he claims it does). I haven't tried it myself but you might want to look into it as it seems *much* cheaper than other content management software.
I have seen with multiple content management applications is as follows:
... where it falls flat is when it comes to dynamic pages that rely on database queries and criteria for the content (.cfm, .php, .asp, etc.) At least on the "web side" of our content management process -- this is causing all sorts of issues when trying to add dynamic pages as objects into the various products. (We already have the source control issue handled with mks -- but since 90% of our pages are dynamic -- most of the benefits of traditional "content management" have yet to be realized in areas where the content gets created on the fly --- turns search engines to mush.)
Most content managers create an AutoIndex feature that works similar "in theory" of a web search engine such as google. When adding static html files as objects into the system --- this works great because it is able to index the content of the HTML pages with pretty good "searchability"
(+1 Funny) only if I laugh out loud.
Documentum has been doing content management before the web began. They provide Workflow/Lifecycle support, controlled document support(e-Signatures, SOPs, Change Request, Change notices), Web Publishing(take a document through a lifecycle until it reaches the published stage and then push it onto the web server), collaboration, digital asset management, portlets, and also syndication. It's pretty comprehensive if you can get everything working together correctly. A lot of the components are hobbled together, but at least it's a complete solution.
Web content management is one thing, but it doesn't go far enough for what I am working on. I have been looking into XML aware content management systems that chunk XML documents, and allow reuse of those chunks. For publishing to html, pdf, paper, and whatever might come out in the future. This is so documents created by multiple departments can be accessed by other departments, and reused by documentation groups for customer delivery. And, maintain consistency accross reused information.
Some features I am looking for:
1) XML based documentation.
2) Multiple authoring tool compatibility.
3) Standards compatibility
4) Check-in/Check-out with release versioning a la CVS.
One of the interesting points I have seen is that content management alone won't cut it. You have to have the leadership in the company to push through the cultural changes to get a real return on the investment. A big point of content management is to enable all groups in a company to share information and build on the knowledge of others. The term coined in the article I read is knowledge management. Contenet management is the tool to free up kowledge in the company from departmental web server and file servers and make it available to anyone who needs it.
Dastardly
Did I read that right? Will this product convert Acrobat files to Flash? Why would you do something like that?
Here is an overview of the various flavors of content management:
1. Content Management. A generic term for managing various types of content. It includes a system for managing digital content files (and perhaps offline content as well) along with metadata that describes the content. Usually workflow and security are included.
2. Document Management. Content management focused on text documents, office automation documents, and scanned images. These tend to be very workflow oriented.
3. Web Content Management. Obviously web focused content management that is oriented towards the web publishing process. Includes some workflow and usually publishing templates and perhaps a mechanism for actually publishing the content to the web server.
4. Digital Asset Management. Focused on being an archive or digital library that other systems such as web content management can draw on. The focus is on re-purposing of content. Often these system are rich media focused and include facilities for transcoding content from one format to another dynamically.
Hope this helps.
I used to work for Bulldog, "The Leader in Digital Asset Management Solutions". We all knew that CEOs and directors loved our product and the people who actually used it hated it. Kind of like Lotus Notes in the early days. It was definitely an matter of style and empty promises before substance. For a while, though, we thrived on clueless pointy-haired bosses--just for a while.
Have to say I mostly agree with those who saying 'define content'. What is it that you want to manage, and on what sort of scale? Most of the products above will cost you an utter fortune.
I've spent a while in this. In 1992 I was doing document image processing, in 1993 document processing with workflow, in 1995 working for a company called the Content Management Corporation (now bust, came this close to getting a deal to be distributed as an Oracle add-in when Oracle web cartridges were being pushed). On top of that, many of my friends worked in the same area. One in particular, who I won't name, has worked in technical pre-sales for a couple of the above products and so knows them well.
Basically, you must define content. You must define what you want to do with content - workflow, revision management, or just a glorified file system? And do take advice of posters above - the world is not composed soley of web pages.
Incidently, I've been given some mod points for this particular thread - a thread on a subject I've been dealing with for eight years. I'll do my best to give away a few intelligent points here and there...
Cheers,
Ian
How about japple?
Pretty neat, from what I have seen.
The best definition I can come up with is that a CMS is anything that offers, in some form, with a reasonable level of integration, several of: content (especially file and data record) control (revision control, access control, triggers, backup), content entry, searching, workflow, templating, deployment, delivery (including personalization), and commerce support. Each of these is a category (perhaps a buzzword) in itself, and you'll have to research what they are and how useful they are to you. While all the vendors will say their products do it all (or--the next version will do it all!), each is stronger in some of these areas, weaker in others. They also vary greatly in the amount of out-of-box functionality, versus how much you need to build, and they differ in ease of extension.
Frankly, it's really hard to make a good decision about these products without putting a lot of time into evaluating them against your needs. If you don't have a good idea of what you want from a system, you'll probably end up buying a lot that never gets used (happens all the time!), and missing out on a lot that could have been useful. So I'd work at defining your needs (talk to everyone who will use the system to see what they think a CMS does), then ask specific questions of the vendors, and try to demo the systems before making a decision.
Also, learn the lingo. You actually can get information out of the marketing material, once you learn the code.
Good luck.
The evaluation of an action as 'practical' . . . depends on what it is that one wishes to practice.
I worked for a web company, Syzygy, last summer and they had a really good content management system called Sycomax. It is not commercially available, only through Syzygy creating you a website, but it was a really good easy to use system.
formerly known as many different things....it's on sourceforge. it is an industrial strength CMS open source project that (in many different implementations) runs the content management for some large scale sites.
Postgres/Mason/mod_perl/lots of other stuff...check it out. the developers working on it know what they are doing. stable as all get-out and a pretty intuitive UI, in my opinion.
My company rushed into buying a content management system (StoryServer) before even knowing the meaning behind the term "content management." Even after three vey ugly years with StoryServer, no one can give a good definition of what "content management" means to our company. We had nothing in terms of content. We didn't have a content location or organization problem. What we had were business users sending I.S. countless word documents to convert into HTML for our web site. All that was desired was a system to allow the business customers to create their own documents and manage the placement of the documents on the webserver (site management). Well, after an amazing display of smoke and mirros by the Vignette salespeople at the time, our business people decided this was the product for them. Little did we know, Vignette had shown us a finished product that was created for sales presentations only. The propduct we purchased was nothing more than a glorified TCL editing environment with a web server front-end. The "content management" part of the system was a joke. We actually had to write code to "manage" the records in the content management system. What was it that this content management system was managing anyhow? We have struggled for years with production issues, slowness, etc. It's a joke. "Turn on caching," they said. Caching, in this system, is nothing more than creating HTML files in the web server's file system. OH MY GOD!!! I could have done that without this "content manager." My point is, know what you want and what problem you're solving before jumping into a product that will supposidly solve all your problems. You may find that your only problem is a management team hyped up by a lot of sales and marketing jargon. We could have had the exact same system we have today, written in Java with an oracle back-end database for "content storage" and have something a heck of a lot more supportable than the miles of cryptic TCL code we now have. Why TCL? Because this is what a StoryServer templates (i.e. JSP, ASP, etc.) are written with. Templates are a combo of TCL and HTML. It really sucks having paid a huge amount of cash for something that is built around a free language. Now we're trying real hard to get this product our of the mix so we can concentrate on just Java. It's a mountain of a task to undo this mistake, though.
Chuckwalla is another commercial
Media Managment System
great for printing outfits that use Macintoshes and eps's http://www.chuckwalla.com
I'd mod you up if I was a moderator. Good one. Guess that explains why I see so few references to Notes and Novell.
Yep..You read it correctly. It can be used to reuse Acrobat pages in a Flash file. It is actually an extensive system. The idea is to be able to find and reuse content.
Kris
I would like to state that I do not have any relation with Documentum.
According to an entire army of market researchers, among which Gartner, Seybold, etc, Documentum is the most important document management system on the market (Gartner: max. vision + max. ability to execute).
In my opinion the product is architecturally flawed and has serious shortcomings in terms of documentation (to say the least).
Architecturally flawed: the product still bears a legacy of client server, or better, outdated client application (i.e. Desktop client) and so-and-so server.
The so-called e-Content Server is in fact a topping of any major RDBMS (Oracle, SQL Server, DB2) which turns that RDBMS into an object-oriented DB and inserts and manages the usual document mgmt stuff into the DB: versioning, document life cycles, workflows, etc. Nothing extraordinary and one might ask why in heaven do they need to turn around the nicely structured relational database into a very intransparent OO DB.
The webifying of the entire thing, because that is what Documentum is after right now, is strategically an absolute mess. I very often had the impression that Documentum itself does not know which direction to choose: J2EE?
Stability: the product is very unstable, server-wise as well as web-client wise. The client-server Desktop client is better but outdated.
One could actually ask whether it is a product or a service you get in exchange of your million dollars or so. Sure you get some software but it requires a tremendous amount of "customisation". And, o yes, forget about getting a set of proven procedures etc. Documentum just shines in terms of absolute lack of documentation on best practice, methodology etc.
An open look under the hood: Documentum is to a large extent a puzzle of software from other vendors. As said: 3rd party RDBMS (ok, we can live with that, SAP also uses mostly 3rd party DBs), 3rd party search engine for the repository (!!!), 3rd party PDF rendition software (!!!), 3rd party (OPEN SOURCE!!!) XML engine (Xalan or Xerces, I do not remember)...
Can a company go for an open source content management system?
Sure it can, especially since large chunks of a commercial document management system are based on open source. But that is the wrong question.
Can the management of a company go for open source software?
Usually not. Because open source poses a risk. Not so many have done it before, there are less highly paid consultants around which you can blame the failure on and besides Gartner says
Just my few cents.
The nice thing about Windows is: it does not just crash; it displays a nice little dialog box and let's you press 'OK'
KnowledgeKinetics (www.knowledgekinetics.net) handles document (content) management as well as a whole slew of other things. Quite a nifty set of tools.
Though a Brit invented the World Wide Web in Switzerland, it seems that Yankees only really understand how to commercialize it!
1. ArsDigita Community System (http://www.arsdigita.com)
2. Open Architecture Community System (http://www.openacs.org)
Slashdot = Sarcasm
check out Japple.
Japple has some very interesting technologies, including a way to seperate your HTML from your logic by a very cool template system -- the most advanced I have ever seen.
If you are looking to manage Business Content, Stellent Content Server is probably your best bet. It has way to many features to list and is very customizable. It comes out of the box ready to run unlike many of its competitors. It also has a strong publishing add-on in case you are interested in publishing native content (Word, Excel, etc.) to be converted for a website. Check it out at Stellent.
Publish Button is an interesting CMS from Canada. In addition to content management, web application logic and page design can all be done from within its web interface. Too bad it's not open source. :)
Publish Button (www.publishbutton.com) is a CMS done in MySQL and PHP. In addition to be able to manage content through its web interface, web application logic and page design can also be done completely through its web interface. Pretty cool... too bad it's not open source.
I run a mailing list for people interested in content management. It was started by myself and my friend Cam at OSCon in 2000, and has grow from the stack of about 35 business cards to a mid-sized list of about 1000 regular subscribers and more on digest. It's populated from some smart, articulate people and there is plenty of traffic these days.
Recently we had some discussion about why or why not use open source content management systems. There are many issues beyond technology in the content management world--the list in general tries to address all aspects of content management, though those conversations are often held in the context of costs and performance.
The list is geared towards users of cms as well as the engineers/designers who admin and support the cms. Marketing to the list is forbidden. There's lots of interesting discussion in the archives. The cms-list is moving to a new home, cms-list.org, but for now, find it at cms.filsa.net.
Phil Suh - cms-list Listmom
This is not ment as a flaimbait. This is now we did it. YMMV.
First you need to decide what you want to do with your system, then how much you are willing to pay for deployment, and then how much you want to pay for maintanence and expansion.
We evaluated a number of CM systems back in 1999. arsDigita requires Oracle (expensive), OpenACS was not ready for real world deployment, CMS based on JBoss is relatively expensive due to cost of programmers. Zope uses its own programming language which translates in to problems with finding qualified programmers.
We had one more requirement. We wanted to have human-readable URLs
We ended up choosing Midgard. It's a GPL-ed Apache+PHP-based CMS running on top of MySQL. It has its share of quirks, but three years after deployment we do not regret the choice. Midgard is suitable for running a small to medium-size web farm. It's based on PHP, thus the programming has the same limits as any programming in PHP. Another shortcoming is exclusive use of MySQL.
ActiveWeb, a German company, has a great content management system that covers digital assets as well as content management. Its granular workflow and security features allow you to fully customize who sees what, and templates, preferences, and other features let you dynamically change how they see it. A great product for the money.
Got Rhinos?
The other critical thing to be aware of is that, compared to, say, your text editor, these CMS's are not mature products. They're bulky, slow, confusing, buggy, hard to install and administer. They're full of rushed, ill-conceived features. They usually have twisted histories that zigged and zagged based on major customer needs, stategy changes, marketing and technology fads, and bolt-on integration of acquired or licensed products--and it shows in a big way. (Welcome to the world of enterprise software!)
When they say "easy to install", they mean it takes one consultant one day--and that's probably for a minimal install that won't do anything useful! When they say "easy to use", they mean that after you have done the install and some initial integration work, and if you have a capable administrator keeping an eye on things, then the business user will be able to get his work done.
Be assured: There will be days when these systems drive you insane. On the bright side: After you've made them work, you can become a highly-paid CMS consultant!
The evaluation of an action as 'practical' . . . depends on what it is that one wishes to practice.
My company has been working on this problem for some time and so far the best (for us) we found is NetShell. the only reason we picked that one is because it gives you a linux style shell on any remote machine. It's based on a lot of the Cygwin stuff but it really lets you build your own content management package.
They do sell a GUI for their system but the price is just wrong.
"there is a marmot in the bucket ? I'll go fix that." (don't ask)
Delivery and Management model or Management Model. In the first instance the tool worries about managing your content, usually using tables it creates in a database you provide or XML files. It also has something like a scripting language or other tool for you design your pages with and then insert your manged content (or links to it). The second model effectivly reaches down into a repositry which the system may or may not have created and effects changes on the files that your delivery tool, like perhaps a servlet with an XML parser reads.
The arguments for an against can go on and on for hours, dependent on varying circumstances. If you have lots of legacy data the tools that can be adapted to use your existing repositories can be useful but then again you have to weigh up the pros and cons of implementation time. You might find it'll import your data a dream but it takes so long to implement that by the time you've done that you could have imported the knowledge into another product.
Main features are:
As someone said earlier, the big coporate players arn't cheap, Vignette will cost you half a million before you've started. Things may have changed since I used it, but yes, these things arn't cheap.
Last thing before I go and leave you good people alone. Opensource, There are some really cool tools out there, one of these days I'll release the one I'm writing for the J2EE platform. Checkout sourceforge. Try content management system or CMS.
That's all folks
Max
--
Insert Clever Signature Here.
Being a Notes developer, I know that notes sucks on quite a number of grounds: cost, memory hogging, really large install, etc, etc. However, it is probably the fastest platform to develop high powered business workflow apps.
As far as content management... Domino.doc is pretty great, there is also a 3rd party web managemnt tool called Aptrix that I worked on an implementation of. both products are pretty great, but only if your business is large enough to warrant running Notes/Domino server in the first place. I mean, Post Nuke does basically the same thing as both products and is free (as in beer, as in GNU).
Don't wait to be hunted to hide. - SB
PHP-Nuke is the best Content Management System I tried out. It has a very BIG community, a lot of addons, modules, blocks and themes to play with, the people is very friendly, administration system is very easy to understand and use, has intelligent modules and blocks system... You'll not be dissapointed at all.
A buddy of mine's been working on Omega CMS - a content management system generator: you enter in your primitives and it sets up the DB, creates a bunch of java accessors, and builds some standard JSP templates- you can customize them to your heart's content, as they are straight JSP at that point. It isn't GPL, but it's freeware if you use an open-source database like postgres or mysql.
What a strange bird is the pelican, his beak can hold more than his belly can.
Try it- you won't need more
http://ariadne.muze.nl/en/
Have you heard of Phoundry? Phoundry is a user-friendly, web-based, database-driven CMS. It's small, rather cheap, written in PHP4 and compatible with MySQL, PostgreSQL, Oracle and MS SQLServer (the beauty of PHP...). Phoundry only manages content in the backend-database and file-uploads, it does not generate a website. It's up to the website-owner to fetch the data from the database using the tools he's used to (either PHP, ASP, JSP, ColdFusion, whatever). Best of it all: anyone who ever touched a browser can immediately work with it. See http://www.phoundry.com for more info and an online demo.
Time flies like an arrow, fruit flies like a banana.
Content Management is like 'family values' everybody believes in it but it's really hard to define. Before closing the deal out on the links to buy Interwoven and its consultants, breakdown the functional components of what you need in a content management system, develop some use cases and see what you can piece together in-house and with open source. Content input and output (presentation) are customized (not out of box) and require consulting $$$.
.org can buy. Gotta have an editing system that constrains to a DTD. WYSIWYG editor for non technical types is cool.
Yeah, Interwoven is a good product if you get it installed right, get consultants who aren't cutting their teeth on their first install, and if you did your homework to see if it is a fit with your business. You do understand your business right? If you are doing templating or it's critical to your buy decision, get a proof of concept from pre-sales support. THis is hard because it means you need to understand your content. Storage, versioning, access, taxonomy are easy. What's hard is when you want fine-grained information, i.e. data with meaning at some sub-element level that drives searches or presentation on the smart retrieval and output format end of the system.
Gotta do XML. Gota do DTDs. Gotta do XSLT for complex DTDs. I like hippie-ware and am toying with jakarta slide and POI [jakarta.apache.org].
Check out xmetal at [www.softquad.com] or xmlspy [www.xmlspy.com] as not free but low cost templating systems even a
A trend in internet consulting, what's left of it, is the movement from building e-commerce sites to maintaining and refacing now "legacy" sites and pushing half-baked, tightly-coupled, all-or-nothing, guru required CM solutions that end up costing millions in consulting and leave everyone unhappy.
We use Interwoven, but frankly for $150k we could have written something more user-friendly in house. .'.
Or at least I have been for the past month+.
:).
I was tasked with evaluating and recommending a CMS on top of Weblogic Commerce Server/Personalization server/Campain server or whatever it's being called at the moment.
After going through all the presentations, whacking at installs and demos, pouring over frameworks and reading through source code I have finally come down to the following recommendations, opinions, and other such stuff.
Let's start with the definition. Is a CMS just supposed to store text, images and possibly other binary files, or does it store HTML and a framework as well? Every single one I have seen (ArsDigita, Zope/CMF Dogbowl, Fatwire, Stellant, Interwoven, etc...) have different ideas of what that means.
I believe that a CMS and a CMF should be separate yet work together nicely. This concept only shows nicely on the Zope project, and not at all in the others I mentioned. Write your own framework or use the CMF Dogbowl, it's all yours to choose. All the others I mentioned force you to use their framework if you want to use their CMS. A CMF is an architectural framework implemented in a language on top of a framework. A CMS is an application written in any language you choose for storing content.
Fatwire and Stellant are ok, but really bloated and untested. They do not perform well and are not even really out of beta yet. Interwoven does not perform much better and is priced somewhere past the moon. ArsDigita is ok from what I have seen, but nothing to write home about and lacks some of the functionality of Zope's CMF. That said I think it is a fine solution if you want to: go with it. Again you are forced to use their CMF if you use their CMS. Zope is my favorite because it's a CMS with other nifty tools like Python and DTML to boot. I can extend it and hack the source, both very nice features. They don't make me use their framework, but if I want to use it then I have a very nice one integrated and ready to bring online. The biggest benefits are discussed below, which was why I was so picky about our CMS.
I also see things that are a mutation of the concept of a CMF and a framework, like Portal Server. This horrific idea by BEA of how to mangle productivity and make the overworked lives of web developers much worse is only more problems on top existing ones without offering anything to ease the pain. BEA's marketing department is using mind-control devices, however, and used them on my bosses convincing them to force me to use the beast.
After learning (through great frustration) how to use Portal Server I have managed to implement a nice solution that minimizes the pain of administrating Portal Servers "portals". (I put that word in quotes because their "portals" are not Portals, but something else entirely which I have failed to properly quantify.)
My solution was to create a pipeline to Zope through a wrapper library and an HTTP connection, a tag library, and bang-whip-zing I have a working CMS and I can pretend to use the Portal Server "framework" (NOT), while really using Zope's stuff. It looks like this in JSP:
Now I pass of everything except actual java programming (like ERP access to corporate systems and in-house tools), to marketing to plug into Zope. I don't get called for "change this style sheet" questions anymore and yet I still have full control over everything.
IMO, if you are going to use a CMS and you don't want to make your life hell use Zope. Otherwise, my second choice is to go with ArsDigita. The rest are just too knew to the game and way too bloated and slow.
BTW: It took me only a couple of days to wrap zope in a library for use in JSPs and It can be done from any type of framework. Sure, it's odd to read content from an HTTP stream until you remember that when you channel bond your NICs, make your connections cached in a resource pool, and use Zope's caching the HTTP stream is faster than reading from disk
My $0.02 will always be worth more than your â0.02, so
You can't have built anything substantial, or anything of any quality, and after the fact say that Sharepoint is a better solution. If you have, you didn't know what you were doing, and you didn't learn from your mistakes.
Interestingly enough, I've build content management type things from scratch + a few libraries, and I've always found that what I did (on Windows or UNIX) generally worked better than packaged stuff.
Now I realize, you're probably a $25,000 a year hack at some big company and all you know how to do is the drag and drop stuff from VS InterDev 6.0, but for the rest of us with actual programming skills, and skills which extend past what can be learned from books with "Dummies" in the tiele, we prefer to actually program something.
BTW, nice way to throw around the XML/XSL stuff. Almost made you sound like you know what you're talking about.
...that doesn't also want to be the web server?
Our department site is piggybacked on a nice big Sun maintained by the university. It can handle a lot more (traffic|attacks|uptime|etc) than any little box we could set up. So I've been trying to find a CMS workflow that outputs to static pages and uploads them to the production box after the changes are accepted.
I checked out ArsDigita, SourceForge, and most recently HotScripts. They have a lot of stuff, much more than I have time to sort through. Love to hear specific suggestions.
There is a product out there called Empower and it is very easy to work with... I have worked with the ColdFusion flavor (very easy to use). There is also an asp version you can use... The users can edit there on content via a browser plug in. Admin controls the rules... Price is cheap compared to Interwoven...
I implemented on a large scale over 500 users,
... for just over a year. They offer University ST and Website ST for online content management at http://www.daxko.com
First: I work for one of the companies that has been mentioned.
/. or anywhere online.
Second: The CM market is very big, look at the public companies' revenue for proof.
Third: Slashdot is usually an open forum without too many covert agendas (other than Linux rules and Microsoft is the devil).
Fourth: Many Anonymous Cowards or first time posters on this thread are CM vendors that are slinging mud and speading FUD (Fear Uncertainty & Doubt) in their comments.
Fifth: As always, check your sources when reading info on
Zope, the killer Python app, when combined with a weblog app like Squishdot can create a solid fondation for a home grown CMS. Even more robust is the new CMF (Content Management Framework) for Zope 2.5, which I haven't used yet. There are additional 'bolt on' apps that will let you index PDF's, Word docs and Excel spreadsheets right into the sites main search catalog.
As an open source project and a framework it is very flexible and can be tailored into whatever you need.
Just as there are many needs, there are many CMSs. I work with many small businesses that want some browser based editing, version control and file management. They want something that is 90% out of the box. That want a simple web content management product.
h tm.
Aaron Renn really makes it clear when he splits it into fours groups (see Flavors of Content Management post).
One thing that shouldn't be confused are LMSs and CMSs. These are two wildly different beasts. In fact a smart way to architect would be to have an LMS front the CMS.
A great list of available CMSs is at http://www.hartman-communicatie.nl/Content/tools.
I prefer our own for small businesses, DojoCMS.
Disclaimer: I work for Gandalf Development, developers of Estrada, though not in the Estrada division
I don't have lots of (read: any) experience using Estrada, but they tell me it's good at, among other things, Section 508b compliance. See http://www.estrada-onstage.com.
My
A lawyer & digital forensics examiner. Also an expert on open source software (OSS).
Obviously, as PHP-Nuke's author, I can't be too much objetive on this matter.
;)
There are a lot of Content Mnagement Systems on the net to try out.
PHP-Nuke is just one of them with many nice features that maybe can solve your needs. But with so many options (free software options) you need to try then decide.
Download all of them, test each one and decide for the better option. You have many options like PHP-Nuke, Postnuke, Slashcode, PHPSlash, Zope, Thatware, PHPWeblog, Xoops, MyPHPnuke, etc... etc.. etc... I can only speak for PHP-Nuke but I prefer that you try and evaluate it yourself.
Hope you can find the best solution in the "free" world
Regards!
Recave
Since there are so many CMS references here, I thought a blatant plug couldn't hurt. And since we're currently in beta testing for our 3.0 release, maybe this will get a few more testers involved.
Check it out at http://www.simian.ca/.
Basic rundown:
putfwd.com - 1GB Free file storage with a twist
I am suprised that only one response on this thread even mentioned "performance".
In researching the available products, they really don't address that issue. In most cases, they don't even mention scaling or limits in their product info.
Pushing lots of content to lots of hosts is a real bear.
A lot of the commercial products are monolithic systems that try and lock you in. Most of it is pure junk.
At a previous employer, we (of course) had to build our own system. It moves around 4 TB (yep, TB) a day. It is indeed sick and disgusting that they move that much content around, but that is another story..
check out http://plone.org/ - Plone. It is a user friendly interface build on the Content Management Framework (CMF) for ZOPE.
.
If you go down the CMF road, Plone is probably where you would want to start. A new release was made today, http://plone.org/download
plone.org *is* running Plone/CMF/ZOPE.
[Disclaimer: I used to work for the company, which is why I'm curious what happened to them.]
At my current assignment, we are implementing a web site generation system, backed up by the SilverStream e-portal CMS.
The system will be used to create a web site, based on the information in the CMS database. The whole system is implemented using standard JSP components (including struts), replacing the e-portal GUI.
There are plans to do an Interwoven roll-out at this company (mainly for political reasons). But even then, our code will be used because Interwoven doesn't have the capacities for on-line content delivery (at least not in the way we need it). In this scenario, content would be transfered between Interwoven and Silverstream for the online delivery.
At my own company, they use Silverstream as well for a CMS, but there they use Silverstream's interface.
WWTTD?
you must assess your needs... others have pointed out that document management, asset management and web publishing have different requirements.
bottom line: most organizations will be forced to customize whatever solution they are sold.
where i work we publish 16000 pages per year, old skool print catalogs... of course, phb (creative/operations) went into power-play mode and bought a newspaper/magazine focused system hoping to kill off development.
the "professional publishing system" jackals have been around now for about two years--in november 2000 we got a nice chart with some goals, one being shutdown of in-house systems (in production since 1989, the age of sneaker net).
our custom solutions cover asset management, copy styling, copy writing (for page and web), product pricing, data replication... lots of time-saving automation and business logic--over one hundred thousand lines of 4GL (4th Dimension and AppleScript).
their "team" solution? deadlines come and go, the vendor lies, managers waffle and waver, it's a mess.
reportedly, we must use their team-shit as it is tied into our publishing contract, and someday, when it works, we will have the source, all in Cpp. of course, it's based on SQL Server, and you can use Quark if you want, but even though they promised it could handle character style sheets, they really want everyone to use Word for copy writing.
i warn everyone who asks me about the system not believe what anybody says...
yeah, i could whine for hours, it's unbelievable. i'm guessing that our mad dash toward a steaming mirage of staff cuts and cut-rate prepress is costing us millions. the stress is causing me severe psychosomatic symptoms, seems like my head and genitals are being blood-starved. i am tempted to bail-out, but that means abandoning oppressed users and letting these obscene fuckers win.
the moral: if you have a database and think you understand databases, DO NOT PURCHASE A SYSTEM CREATED FOR ORGANIZATIONS NEEDING A DATABASE.
beware silly fucks who want to sell you their "tools" and do not really have any interest in how your enterprise data is currently managed.
Stick with the PHPNuke forks, especially PostNuke, as the original is terrible - major lack of security, code uncleanliness (and hence a lack of extensibility), etc. Just ask Wayne Hunt (wayne at amiga dot org) his views on it - he had his site hacked over and over again while he used Nuke.
The PostNuke folks have been doing a great job of clearing up the code itself, a much better job than the phpWebSite guys at Appalacian State.
According to hotscripts.com, eZ publish is the most popular script. We use it in our buisness, and we are very happy about it. eZ publish is a very fast and scalable, and, in opposite to many other open source systems, very well documented.
:(, but very reasonable priced :) ) that makes you able to publish articles with a WYSIWYG interface. This makes publishing articles very fast and efficient.
One of the advantages by using an open source system provided by a commercial company is that you can buy support if something should go wrong.
They also have a desktop client available for Windows, Linux and Mac (although not open source
The biggest problem with eZ publish is that the system is very big. The code is very vell designed and documentet, but it takes a while to get to know the system.
Part of my job involves being webmaster for our company. When I got to that position we allready had a new and improved website. (I hadn't been involved with the creation of that site). It was constructed so that anyone in the company would be able to add information and news to the webpage. And that without knowing anything about HTML. Sounds nice, eh?
Well, the entire system was delivered by our advertising agency, and we didn't need to care about the system beneath all the beautiful eye-candy. That was a bad thing as it turned out. The site was built with a tool from a web agency which created ASP pages. It also involves special addition to Microsoft IIS (the only webserver capable of running the system). For us to play with we had a nice content management built into the site. We couldn't change absolutely anything (like moving a textbox 1 inch to the right) without calling the web agency and asking for their help. But, wow, we could add certain news items in mandatory textboxes. As it turned out, nobody in the company (who wasn't a HTML-guru) dared to change/add anything to the webpage, because the content management was extremely confusing and unorganized.
I, who know quite a bit about creating websites with HTML/PHP/Perl and whatever, was becoming more and more frustrated with the "nice" content management tool. It also required me to use Internet Explorer 5.5 (nothing else) and load a special entry to the Windows registry which lowered the securitysettings in my browser. When wanted to add a text to the site the CM system started a miniature Microsoft Frontpage. That means it was impossible to add any kind of HTML in the text, no links nothing. But, whohoo, I could press CTRL-I for italic text. Damnit, I run a Linux desktop at work and aren't too happy that I need to switch computer every now and then. The system was also extremely unreliable and would crasch once in a while. I have created a special WWW folder on our IMAP server where I store all mails to the web agency regarding the content management system and the webpage. It's over 100 mails since the beginning of the 2002.
The web agency is now bankrupt and we are working on a new webpage using Zope on a UNIX server. It's really a great and extremely flexible tool which enables you to quickly create advanced dynamic sites.
Ciryon
Their product is used by many organizations to manage content, including web content. Notable clients can be found at their web site, a good number of drug companies use their product to manage fda drug applications, and delta airlines uses it to manage their web site. Link here.
To let you guys get a better idea of what the nightmare we are dealing with is:
We have (currently) 8 companies being hosted with independant custom portals linked to our conent/functionality with about 500 more wanting on-board but being told to wait until we get done with our beta customers. Every company has their own marketing department and every one of those wants to have their fingers all over their portal.
So, if you have ever worked with a marketing department, especially one that doen't really know you are in fact human, you can appreciate the situation we are in. We want everything to be as brain-dead simple as possible with the site, and be free from doing anything other than writing really cool server-side tools.
My $0.02 will always be worth more than your â0.02, so
There is another application of content management frameworks. Specifically, "collaboration management frameworks," where the components are roles and workflows. Your description of "content management" touches on this, but I think collaboration management stands on its own.
The idea is that instead of publishing a document that describes a process that everyone must go find, read, understand, and follow, you publish the process instead. Then when you make changes to the process, instead of announcing the changes, and expecting those who miss the announcement to go look for changes in the published document (after finding it again) or discover process changes while talking to their buddies in the hallway, changes to the process are instantly visible to the affected parties. That is to say, you publish the change to the workflows between different roles, you don't publish a description of the change in workflows between roles.
There are a lot of ideas and concepts to explore in this area. Zope, as a Content management Framework, is a good framework for collaboration management frameworks.
Take a look at Zope. We have been using it now for 18 months and it is great. It is a very flexible system that allows you to build much more complicated systems on top.
It has a very unique feature called 'Acquisition' in which if the server doesn't find an object in its current path, it does a tree walk to find it. This makes it very easy to build sites with common elements near the root of your tree, but override stuff for specific branches if you need to.
It also has its own build in persistent object store, which is really nice for programming new content types (in Python) is you don't have to explicityly pack/unpack load/store data in an external RDBMS.
-Matt
We just implemented Stellent's Xpedio Content Server for a client. This is really a hybrid product that serves the web world as a content management system and also the backend world of workflow (a little tricky to customize the standard workflow to suit our needs) and document management. Also included is a link between the two meaning you can have MS word docs transformed into JSP (any format really, this was just what we did) and publish them out based on some publishing rules.
I would give it a 3 out of 5 for its CMS capabilities, but a 4/5 for its DMS capabilities. Pretty good out of the box product. Easy enough to customize--includes a JDK for writing cutom components and has an EJB for tapping into the database. Also as a plus it comes with Verity search engine bundled which does a nice/fast full text search.
This is on the lower end of price for medium sized solutions. It was a little clunky to get redundancy/clustering up and running, but in the end we made it happen.
I wouldn't use Citydesk for a major project requring an RDBMS, but for small websites it's the best. Most people using Frontpage or Fusion would be *much* happier with Citydesk.
Apparently, it's REAL easy to botch the installation on this thing, so DON'T INSTALL IT unless you've got a team that is KNOWN TO BE COMPETENT WITH IT! Get references. And if you don't have to use it, then don't! Too many horror stories have I heard from friends! Heed my words, or suffer the Dr.'s fate you will!
Useless opinions, worthless observations, and more!
I couldn't help it: http://www.theregister.co.uk/content/30/22650.html
I've tried dozens of content management systems, and Phillip Greenspun's baby still rules them all. Aside from reliability, functionality, and performance issues, the biggest problem with these systems is that they require developers and users to learn a whole new paradigm, language, and way of looking at the world- often completely new, and unrelated to anything else. Any reasonably competent programmer/webhead can pick up TCL and extend the ACS- without struggling with Zopisms, J2EE jargon, or Vignette-ese. And all users need to learn is how to cut and paste stuff into a browser form.
I installed phpWiki this morning, on one of our dev servers for internal documentation. It's not as powerful as many other systems, but it's incredibly simple.
Getting it to run with IIS isn't, however.
I work on the phpWebsite development team at Appalachian State University. One of the reasons phpwebsite was started was to develop a CMS for our use. We have chosen to tackle our problem of large-scale content by breaking content up into departments and then each department runs a copy of our core software. You then write modules that integrate the sites together and expand capability. Although we are still in beta this is where we want to be when we have the 1.0 release. This would be a structure of integrated sites. You may want to take a look. It?s GPL so you can see everything now if you want too.
I'm in the process of developing a suite of content management systems targeted specifically at Software Development Teams.
I'm using Mozilla as my platform. The QBAL applications will be accessible thru both a GUI
(Mozilla/XUL) and command line. My background has been in Unix development environments in Silicon Valley since 1984... From that I hope to bring forth experiences in the form of tools to improve that day-to-day work experience for development team members
Currently I've defined 27 QBAL applications as part of the suite. Clients will be able to select the QBAL applications that best fit their needs.
Regards,
Kramer
does everything you just described and a little bit more. Probably the only compelling reason to switch to Office XP if you are already on Office 2000. If you're on an earlier version it's time to switch anyway.
You could rescue an old Pentium clone from the dumpster and install Linux + Zope + Squishdot, all at a cost of $0. Not exactly a production environment, but more than enough for "proof of concept" and a smoke & mirrors demo.
I had a similar lack of funding problem. My organization has a capital spending authorization procedure that rivals that of the Pentagon. I installed Linux, Zope & Squishdot on a piece of Pentium-100 junkware, and customized some of the Squishdot screens so as to make it look like series of customized portals to deliver reports as PDF file attachments. The powers that be couldn't wait to spend $7K on a Dell Poweredge server. There is so much more I would like to say about Zope + Squishdot, but I have to avoid needlessly educating my competitors. I can't mention where I work, my employer's line of business, the clients we serve, or any of the specifics, but I can say with authority that Linux, Zope & Squishdot can be turned into a "wonder weapon" against larger competitors who have more money to spend.
I'm from Curtin University and after extensive evaluation of other content management solutions we ended up doing it ourselves.
The main reason ?
It reduces the money and time needed to customise the off the shelf product to the specific needs of the organisation.
We ended up using Apple's WebObjects which is actually a fantastic product. So now, we have a CMS tailored specifically for the university that meets only the universities needs.
Total cost ? 15000 US for hiring someone (im in Australia) + 100 US for WebObjects.
Considerably cheaper than something which may not necessarily do what we want.
HTH,
Naden.
Funtage Factor: Purple
We use our own Windows-based system "zeta producer" (german) for other things like e.g. creation of Windows Help files.
Beside City Desk, I don't know any other mid-/small-sized CMS package that is NOT implemented as a browser based application, but as a full features desktop Windows application.
Is someone aware of other systems?
-- Watch me working: www.magerquark.de
True fact, dat be.
based in Python, runs on Solaris and Windows.
We have it on a sparc server and it is slick as can be.
$5 / month hosted VPS on linux = awesome!
We bought a product from (company name withheld). There were some customizations that needed to be done for our company. It took well over 18 months for them to get the first portion running without too much of a problem. Then there were arguments over how certain pieces of the system were supposed to work. We got completely fed up and decided to write our own. About a year later we had a D.A.M. system we could call or own and works much better for us then the other system.
If you are considering going with a certain product, get a list of people already using the system and call them. Ask them how they use the system, how well it has performed for them, how much downtime have they experienced in the last year, and most importantly, how was the support when they needed it. If you can test drive the system, do it.
ZDNet ran a feature on Outsourced Content Management solutions a couple weeks ago. Talks about the Application Service Providers that are available for content management. Having spent the last 5 years implementing Vignette, Interwoven, and custom-built apps, it might be worth looking into...
Whatever you do, don't buy a proprietary CMS.
:-) )... :-). Just add a .org or .com to those I mentioned and you find your way through. (to lazy to write links just know :-) )
This is serious OSS turf and most of the professional CMS - Vendors I know and deal with in buisness have gone/are just about to go belly up. Note that those who do go belly up usually release their babe as OSS themselves in the end (ars digita - a Java thing I gather - for instance)
There are tons of OSS-CMS solutions out there that kick their proproietary counterparts up and down the street.
Depending on what project scale you have in mind I'd chose between 2 to 3 strategies:
#1: The small, minimum overhead PHP/MySQL (an SSI solution... THE SSI solution) way. Tons of ready to use OSS solutions out there, lot's of ISPs with PHP to go. I personally use phpnuke (kinda like 'slashcode reimplementation in php') as the cms for my webproject (www.modp4rlor.com). Note that PHP rulez the SSI market above ColdFusion, JSP, ASP and all the rest and is somewhat scalable if you use the proprietary stuff like that from zend later on.
#2: The big, fat , hairy project way with all the Java might the OSS community has to offer (apache, tomcat, jakarta, cocoon, turbine, jetstream,... you name it). All of these are leading edge, GPLd and kick serious ass in large scale projects. Allways keep in mind though: Beware of the Java overhead. A lesson hard learned in the recent years by lots of people who thought their 10-hits-a-day Site would look cool with servlets...;;-)
#3: My extra-special, quite very scalable, one size fits all, CMS, DocMS 'n multithreaded Appserver in a box: ZOPE!
Features: Exept for some performance critical stuff in C completely written and extendable in Python. A fully OO, bytecode interpreted PL, GPLd of course. Very nice. Easy to learn, next to no overhead for your small projects, powerfull enough for larger stuff like Document Management or Revisioning or Publishing Systems. I know IT-service companies that rely completely on Zope (internal and for customers) and pull some serious projects.
Has it's Web and FTP Server on board but can also be run as an addon to others (apache f.i.)
Comes with it's own small and unobstrusive SSI solution (DTML - Document Template ML) just suitable for everything that's to small for servlets.
Is fully OO and uses it's own Post Relational Database that actually stores EVERYTHING it uses.
Very easy to extend via plugins, so called "products" and it's easy to build these plugins too.
Oh, almost forgot: It uses a web interface. Only a web interface. And a good one on top. And it takes on klick to install on Windows and something like 4 and a half on Linux. Whatever you're up to, definitely check this one out!
Roxxen is quite cool aswell. AFAICT a sort of smaller Zope using Ruby (sheesh
Well I could go on but I think you see that there is no need to spend God knows what on Cold Fusion or Intershop
We suffer more in our imagination than in reality. - Seneca
That's pretty much the case for most general purpose CMS (rather than something site-type specific like Slashcode, PHPWebsite or whatever). Whether you're talking about Zope or Vignette, you're still essentially getting a high-level toolkit, with some nice APIs to handle stuff at the level of user authentication, workflow capability and so on.
You're still going to have to put something on top to run your site, whether it's a higher-level still toolkit like CMF (Content Management Framework, note) for Zope, or Multisite Content Manager (previously known as Enterprise Application Portal) for Vignette.
The only thing you can accurately describe as "Scotch" is a sticky tape made by 3M. And it's
Again and again you say: need training, need this, need certification, needs to be installed just right, needs to be done just this way or else .... etc. etc.
How many of those statements will it take to convince you that you're using a piece of (expensive) crap?!?
If it's so freakishly complex, why use it? There are plenty of other options, Zope, for instance, that don't require a degree in nuclear physics to operate.
Unless this thing makes you millions an hour and solves world hunger and achieves world peace, it doesn't seem to be worth the incredible hassle it appears to cause.
You might want to consider whether WebDAV is a key standard for your needs.
The WebDAV extensions to HTTP 1.1 provide a way for remote authors to manipulate files and collections, and an extensible set of file properties; and addresses issues such as locking.
A related standard provides for versioning: RFC 3253
As such, WebDAV and its associated standards effectively standardise what the various CMS and DMS vendors provide proprietary interfaces for.
The standard is now widely implemented - see www.webdav.org - both in operating systems, and particular clients and servers (although not yet in most of the aforementioned CMS/DMS). For an open source Java implementation of client and server, I use Apache Jakarta-Slide.
It is and will continue to be fascinating to watch how the incumbent CMS/DMS vendors react as their market gets commoditised.
As someone who's spent a lot of time working with business users on formal CMSs (and hand-rolled systems before that) over the last 3 years, it's not really that simple.
"should be" is all very well, but please do keep to the experience side of conjecture in future.
Try Bento, the best I have seen by a long way.
I've been working specifically with Content Management for the past five years. If you think that it is simply something that is just a buzzword or bullshit that is sold to corporate heads then you don't understand the real value it provides. Being an application developer myself, I think that the largest value of Content Management is appreciated by those who create and manage sites.
It provides abstraction of your content from your look-and-feel and can drastically reduce your development timelines and ease of maintenance. That may not matter if your site is only 5 html pages, but it does when you have thousands of them (and want to use the content on them in different ways).
The problem is that most of the exisiting products (in addition the huge price tag) have been oversold on what they do or even what they are. Remember that the Content Management market is still in it's infancy, so a lot of different compaines entered the space just providing a development platform or by rebranding document/digital asset management products. This left most of the work up to the developers or the profession services (good revenue for the companies selling the software) teams that have to implement them.
What I found led me to the conclusion we really needed some true turn-key solutions. After building about 5 custom Content Management Systems for various large corporations, along with my dev team we've spent the last two years designing and building a commercial product called Conclarity CMS (http://www.conclarity.com), which we are just about to release. It's built in Java using the J2EE framework, every content object is available as XML, and it uses XSL and XHTML for templating.
If anyone is interested in checking it out, I'd be more than happy to give away some free piolit licenses to slashdotters.
Landon Hall
President, CTO
Lucid DataStreams, Inc.
Most CMSs actually don't really bother about searching via their internal datastores, mostly because of the problems you've raised above.
What you tend to get are two (not mutually exclusive) approaches:
Sidenote - you can get around those query string URLs with most serverside scripting environments (as well as that PHP tutorial, it's worth bearing in mind that evolt's site is built in a similar way using Cold Fusion)
The only thing you can accurately describe as "Scotch" is a sticky tape made by 3M. And it's
How old is this submission? The conference 'next' month in Zurich was March 21-22 2002 (ie it finished a week ago). It sounds like something I would like to have attended. Thanks Slashdot for the usual timely news :-(
Phillip.
Property for sale in Nice, France
Bought by Broadvision in April 2000.
The only thing you can accurately describe as "Scotch" is a sticky tape made by 3M. And it's
Workflow (and workflow auditing) is also very useful to answer the question "why isn't the content getting there fast enough?", which is the opposite scenario to the one Dave painted. In the CMS or not, content publishing is still a business process...
The only thing you can accurately describe as "Scotch" is a sticky tape made by 3M. And it's
Zope seems like the real deal when it comes to Open Source companies. They release for free and then do commercial add-on and support. Cool stuff.
Zope has two sites: Zope.com and Zope.org. Send your developers to Zope.org and your boss to Zope.com.
.org is their community development site (which also runs on Zope). It is a very active very homey collaborative environment.
.com is the suit-friendly corporate face of Zope. This is where you send your boss to assure him that you can get commercial support / training. Zope certification will begin soon. This is one of the few certification programs that I'm ever likely to even consider.
All in all, Zope is a really solid piece of software. The new CMF does a great job of separating roles. Don't forget to visit the Demos (very informative).
Good luck with Zope. You will not ever regret using it.
-Peter
. Penguins Surely Ca
I get cold calls at least once a month from headhunters wanting me to do Livelink. I don't list a phone number, email address, or physical address on my online resume; they call information to get to my company and then work their way in from the front desk.
The product is a web based document management system, like Documentum in theory, but much easier to work with in practice. I've been using it since September 01, and it has grown on me.
Users interact via a web site or WebDAV (supposedly works on Linux) to view, add, check in, check out, or delete documents; to interact with workflows; to engage in discussions, and to do whatever else you have your server configured to do.
The web interface allows for use with any operating system, and the java widgets seem to run on our Linux, Sun, OSX, and Irix boxes. (and of course on Windows!) I can't speak for WebDAV, as I haven't used it. I spend a lot of time using their Office integration widgets, which allow me to interact with the repository directly from Windows or MS Office. (More menus appear in your apps.)
I like it because it exports XML over HTTP. I send it a URL and object number, and it sends back a pile of XML that I transform into a web page. It means that I rarely have to update web pages, as I just say a web page is made of objects of type Y, and those objects show up on the web page when a user checks them into the repository.
What's strange is that in the US, it is not too popular. Livelink consultants are impossible to find, and generally bill at around $100/hour. From my colleagues in the UK I understand that the billing is about the same, but that consultants are easier to come by.
It's strange that the product doesn't have much of a name because OpenText has been around since 1991. They're a Canadian company who do about 100 million in sales a year, so they're not small or new.
Anyway, I wasn't impressed up front, but have turned into a fan. And the user response, always important in IT projects, has been extremely positive. Just around 60% of the site's users have requested to participate in training sessions. (Which aren't cheap: Opentext bills $3000/day for onsite training!) And almost every user has been interacting with it daily, with almost no complaints.
So, it's definitely worth considering, especially if you have the money!
Starphire.com has a completely web-based content management system called "SiteSage". Called a few weeks ago and they stepped me through a demo online - that thing is sweet. All changes are in realtime - 2 seconds after I logged in i was changing images and text and font styles - NO CODE! Had search, security, ecom, boards, themes... bla bla bla... presented so you don't have to know code to change stuff. Everything's objects - the call them "Web Blocks". Hundreds of them. Slick stuff. I'm starting to sound like a commercial. I'll shut up now.
Some urls which may be of interest are:d r it y.html_ doubts-1 .html3 3424-60 1-2254701.htmlr ticle/jsp/sid/60 51470 122200 1.html#Headline7482
http://philip.greenspun.com/wtr/vignette-ol
http://www.puffinry.freeserve.co.uk/sss/SS-secu
http://www.shorewalker.com/pages/vignette
http://techupdate.cnet.com/enterprise/0-61
http://australia.internet.com/r/a
http://www.siliconalleydaily.com/issues/sar
Hi, we have a GPLd content management and web administrator application development toolkit written in Perl and running under Apache/MySQL, try that. Visit the site at www.changingpages.org and contact us for a download.
We're looking for volunteers to work on this project as we never seem to find the time to get the next release out. Hopefully a CVS will be available very soon. Anyone interested? Again visit the site and let us know.
oh?... 400m distribution doesn't cut it?
You'd have to ask what SWF brings to the party... maybe if it was openly searchable, transformable, interoperable or portable... but...??
Since V5 the Acrobat Reader install includes SVG viewer. SVG is an XML format.See the PowerPoint examples at SVGmaker:
PowerPoint SVG example from a bankers presentation
-- SVG Evangelist
If you mean "web content management" to mean organizing all your content, versioning, publishing from templates, automatically generating tables of contents, keeping track of the action items, controlling who on your web team can publish on the public web, and server-based content management, you'll want to look at WebSite Director from CyberTeams (http://www.cyberteams.com/). I'm a very happy customer for WebSite Director. My web teams use all the flavors of WebSite Director to manage some very large web sites, ranging from a database of stuff about lunar development (http://www.asi.org/) to sites about the computer game, The Sims (http://www.simgoddesses.com/). Our internal committees use WebSite Director Express for their work in progress. Frankly, we couldn't do what we do without WSD. We're building a new system called SimsHost (http://www.simshost.com/) that will end up spanning multiple servers with more than 100 webmasters. I haven't found any way to do this without the WebSite Director suite short of starting from scratch writing our own CMS. Greg Bennett, President Artemis Society International
Ignore the oficial 1.3.1 release and go for the dailly snapshot archive.
It can handle anything from the simple site to a slashmydot. It's new. It's all PHP. It has a simple interface for content management. It's my pick on CMS systems.
You get what is behind sites like:
The download site itself
Masterplan
Pratica
I'm in the WEB CMS business, so I resign from moderating though possible in this discussion.
:0)
:) is not supported.
Should anyone want some Danish jibberish, the direct URL to a CMS is Sophistic CMS and it contain a few screenshot cuts and some explanations. Go fish translate it, yeah right, no Danish support
The advantages of a CMS, are that deploying a full featured website is incredible fast assuming the right tool is available. If this is so, you can concider it as a toolbox. You may have plenty of ideas of what your website should do for you, but developing everything from bottom-up can be an expensive task, with a CMS with selectable components, you design as you please and can implement components as required, and it is up running in a heart beat. Furthermore you have a CMS development crew as your backing to ensure that the components are matching the needs of tomorrows tasks. To finish it off a thorough CMS also provides you with the tool to maintain and develop your deployed site even further, with ease, with proper access management to distribute specific tasks.
A sample is BLUNT a website for a new rockband featured on Danish TV3 through 10 shows generating lots of interest. Universal, their record company, had sponsored a small amount to support the band and their online promotion (We are currently preparing negotions with Universal to allow mp3 sales). The general photoshop/gimp+html design was provided by the design crew(among it were one of the band members)Monday, March 25th at around 13pm and was up and running less than 6 hours thereafter, supporting both IE and Mozilla/Netscape, Opera, etc.. I hope. Please dump me a mail at the address below if your browser (besides lynx, sorry
A CMS is a broad topic I give you all that. What matters is that it is about managing content, whether it be an intranet(office documents and tools for the daily business..e-business..), an extranet (marketing, product, pricing material, etc..), or a website with promotion, store, discussion forums, information spreader, its all about content, and thats what a CMS can cover as a tool.
Basically its reusing code and sharing the development code of an tailored administration system, where a professional CMS consist of a lot of different components from which the integrators can implement and configure according to their needs and be up running in less time and with less waste of money. One of the benefits for non-techies is that the editing tool works just as a word processor and it can be delivered to support just about all platforms thanks to use of java or activex, depending of what is best for the users.
Basically its a tool.
Sophistic CMS, to which one of the above links is directed to, is pretty inexpensive and could be what you need, Danish Design, what do I know, IT IS YOUR CHOICE.
The pages are only available in Danish, but you can reach our Danish office at +45 86 13 73 15 between 9am - 17pm CET og by email contact at sophistic dot com
This may be concidered an advertizement, but it is still free content, so if it suits you, it is free for you to use.
Best regards,
Casper
http://demo.vp1.com, its called intraview, and its soon to have it own api for the IT folks who want to customize their setup
We are using it as a component in a J2EE system. TEAMS performance, functions, and flexibility are worse than bad. So we're tossing it.
But notice that in all of the file formats I've just mentioned the conventional tools (e.g. Dreamweaver, BBEdit, Photoshop, Word (bleagh!) or Xpress) a document is pretty much just open/read-only/ or closed. What QPS did do which no other system I've seen since could manage came down to two things:
1.) control at the feature level, as in controlling for each user whether or not they could use each individual capability of the program on a given type of document
2.) control by document type, both in terms of file type (text, image, layout, whatever) and in terms of the document category (feature, ad, sidebar, etc.)
It also did an excellent job of allowing systems to be built with any number and type of statuses desired. Most systems out there are very primitive on that front, limiting you to say, rough, live, final, and trash. In QPS, companies could and did have as many as fifteen different statuses, which makes sense if your standard workflow involves something like three passes by an editor, two just for design, and then several final states such as publish, reserve, or hold for next issue.
If you were crazy enough (and the senior editors didn't figure it out) you could theoretically create individual config info for each user, saying, for example, that user Zorch could not change text color (feature) in a sports story (section) for the weekly edition (publication) that was ready for press (status).
Now this, obviously worked best with XPress and Quark's text editor CopyDesk, but the system also worked with Photoshop, Illustrator, Word, and a bunch of other apps.
Since then QPS has been sold to a company called Modulo, the primary provider of XPress plug-in equivalents (called XTensions) has four times gone bankrupt and Adobe has spent the better part of a decade on their now kinda stillborn competing system long codenamed Stilton. But... I've checked out each system as it has come out (Vignette, Documentum, et bloody cetera) and nobody yet has come close to that degree of control.
My sense is that if you can wait about six months this should finally start to change as I found that at the most recent Seybold show some vendors were starting to get a clue and the many SQL solutions were starting to create a pretty good toolset. Personally I'm now starting my own publishing company and I'm planning on going with a mostly homebuilt system based on SQL.
Good luck and tell us what you decide,
Rustin H. Wright
founder, Reed&Wright
pubgeek@netscape.net
I know this might come out as a saleproposal but it's really not. Turnpike is a CMS written completely in C# for you who do use the MS plattform (or thinking of using .NET for either FreeBSD or DOTgnu).
It's made by a team of Swedes.
If at first you don't succeed, you're running about average. M. H. Alderson
I don't think this poll shows very much.
Most of the people probably is using Postnuke because it's the easiest, but slashdot clearly has much better features, and is much more complicated, it's only problem is that it's impossible for a non-geek(TM) to install.
So the poll is merely a reflection of what people use, none of the 6000 people have tried them all, and so they can't give a objective vote.
____________________________
Listen to my friend's Band Nectar
And support her new band The Rime
I'd wind up doing those changes myself whether I had a fancy content management system or not. If I had gotten one, I would have had to learn how to use it effectively, which probably would have taken a lot more time than banging out a few shell scripts.
;-). They're just happy because they got it for "free" (as in beer). So, it stops a WHOLE PILE of moronic requests, like...
But the BEST part: the execs understand that the CM is put together in shell, and that there is only so much it can do. (Well, that's what they think
..can you change the "alert" button to blue?
..can you write a cache of user data on the guy's harddrive?
..can't you just see if the guy's using netscape, and if he is, load the page in IE?
..can you change the brower's borders to match the site layout?
..can you make the confirm dialog box use Arial fonts?
..can you generate thumbnails for the mp3s?
(etc, ad nauseum)
Do daemons dream of electric sleep()?
For source management in house. Says plenty about VSS, I think. Piece of fucking shit. Argh, just thinking about makes me so angry.
Unless they have changed there pricing - they are way over charging for this BLOAT WARE. They wanted us to pay over 9 mil because we were a financial company.
We laughed in their face and purchased something else.
Another GPL content management system you should be aware of is the APC Action Apps. It was designed primarily to aid non-profit organizations in their content-management needs, thus it has excellent content sharing features - but it can be generally useful for anyone. It also has some nice user documentation.
PHP-Nuke is your best option, try it! you will love it
In my opinion, set-based file management systems would scale better:
http://geocities.com/tablizer/sets1.htm
Trees don't deal well with things that fit into more than one category, for example.
Table-ized A.I.
We have Vignette in house, and we started to looking at Interwoven as a slightly cheaper and better solution to our requirements than Vignette. I've read flaming emails full of hatred for both products, and I'm having trouble picking out what's fact and what's not. Anybody been successful in actually evaluating and choosing a CM tool to replace one in house? The other problem is that we already existing crappy homegrown cm tools that are poorly architected and store the content in the worst ways. I've been trapped on this project (replacing existing cm tools) for the last year and a half. talk about nightmare. The only thing i've learned is that there is no good solution for content management right now. Nobody knows what they really want out of it. and don't even think about trying to do content management for a large enterprise without upper management support. You'll end up screwed. Thus endeth my small rant. Anybody have similar experiences?
You asked if there are other solutions. I would look heavily at divine's CSEE. Many companies are replacing IWVN and VGNT with this product. It is native to a J2EE application server (unlike the others you mentioned) therefore leveraging all of the load balancing and failover capabilities. Just a suggestion.....
I admit that I work for Vignette, so my opinions will be somewhat biased. And, I also admit that I really don't know a whole lot about IWOV's technology. However, I can tell you that there is a lot of disinformation about Vignette out there. The thing that caught my eye about this post was the suggestion that VIGN is not native to a J2EE application server. It is (more precisely, it is native to the Servlet Engine). VIGN has come a long way since our TCL days. Our STRENGTH is our ability to comply with industry standards (ASP and/or JSP) and not require the use of proprietary scripting languages (e.g. IWOV's version or PERL).
Perl is proprietary!! since when?
Ok, I get the bias and all, but I'll have to diagree with the "does not require the use of propietary scripting languages". You're right, Tcl isn't propietary, but what about all those storyserver commands? And the funky SEARCH TABLE and FOREACH loops? And the content records seem to allow you no way to escape from Vignette easily - if you have a wish to do so. I'll admit that some of my irritation with Vignette is most likely due to a really bad implementation.
Look at http://www.nuomedia.com. It has all you are looking for, and although it is a private implementation, I am considering opening the code up if users can port it (Java servlets) to something like postnuke. Give me a mail if you are interested - rmiddleton@nuomedia.com