Improving usability with Oracle
At this meeting Oracle and founding members will:
- get to know each other
- collect usability issues
- determine common issues to work on and
- set goals and direction to improve usability of enterprise applications.
The usability issues I will be raising are:
- Out of the box usability must be high
- Better usability does not mean more features. It means features must be more usable.
- Think of users when offering help.
- Don’t focus on making error messages better, aim to prevent users making errors.
I’m sharing some slides I plan to use at the meeting. Anyone want to add anything?
Oracle announced that the premier support for Oracle UCM 10gR3 ends dec 2012.
|Oracle Universal Content Management 10gR3||May 2007||Dec 2012||Dec 2015||Indefinite|
|Oracle Universal Records Management 10gR3||May 2007||Dec 2012||Dec 2015||Indefinite|
Note that Oracle UCM 10gR3 is the current release! (See http://www.oracle.com/technology/products/content-management/downloads.html)
Overview for all Oracle Enterprise Content Management Releases
Maybe somebody should revise the Oracle® Universal Content Management guide, “Managing Security and User Access,” where it says on page 7-4: “If you enable accounts and use them, you cannot disable them without losing data. DO NOT enable accounts unless you are certain that you want to use them.” Either the documentation is wrong, or you lose data. It says nothing about the “appearance” of having lost data.
Page 3-3 of the same piece of documentation says: “The number of security groups should be kept at a minimum to provide optimum search performance and user administration performance. If your security model requires more than 50 security classifications, you should enable accounts and use them to control user permissions.” I take this to mean that the performance degrades noticeably (or can degrade noticeably) after you scale beyond 50 security classifications. Later, the documentation cites an example where changing a single permission can take 10 seconds. Not to be a pain in the ass, Bex, but how does this support your statement “This model scales very well”?? (I take it back. I am being a pain in the ass.)
One last carp. You say that “ACLs are horribly slow and impossible to administer.” For this particular CMS application, that may be true, I don’t know. All I know is that ACLs are the de facto industry standard way of doing this sort of thing. When you choose Door No. 3 and invent a nonstandard approach to solving a problem for which the wheel has already been invented, you only end up needlessly confusing and scaring analysts — and making customers read documentation, something they hate doing.
At any rate, I did learn a lot from your excellent writeup. Thanks for doing it. I feel better now.
Don’t know why that’s there… the content you checked in is still in the repository, and the metadata is still safe and sound in the database. Users will lose access to these documents, until you either update all your users, or update all of the values for “account” in the database to blank. You can do batch metadata changes with the Archiver tool… which should be done prior to turning off accounts anyway.
I take this to mean that the performance degrades noticeably (or can degrade noticeably) after you scale beyond 50 security classifications.
In general, performance degradation is due to the complexity of the security model, and not the number of groups or accounts. For example, if you have 100 classifications, but most users can only access one or two classifications, you won’t see many problems. The “security clause” I mentioned above would be pretty small… However, if every user gets access to 50 classifications in different ways, then you’re likely to see performance to degrade a bit because of the increased complexity of the SQL in the security clauses. This can be fixed with some database tuning, however. Some of the admin applets — like User Admin — load more slowly depending on the number of security groups, but that’s rarely a big deal.
All I know is that ACLs are the de facto industry standard way of doing this sort of thing.
Slow searches are also the de facto industry standard
ACLs are easy, which is why everybody does it that way. We took a look at how everybody else did it, and knew that they were doing it in a way that would require a ton of hardware in order to function, a ton of maintenance, and a ton of risk. We didn’t want to go that route… and what we came up with was pretty close to how LDAP does things. Seems to me like a good gamble that paid off…
I cannot name names, but I encourage you to talk with enterprise architects in industries with serious security concerns — like financial and government — to ask them what they think of ACLs in general. As I said, you still can do ACLs with Oracle UCM, but you’ll need beefier hardware.
August 31st, 2009 by Janus Boye
Oracle is among the largest global enterprise software vendors and like IBM and Microsoft, Oracle entered the CMS marketplace via an acquisition (Stellent in 2007). Oracle Universal Content Management (UCM) is based on the original Stellent product now fully rebranded, much improved and leading the market according to IT analyst Gartner. Does this make Oracle an obvious and safe candidate on your Web CMS shortlist?
We find that Oracle UCM does not come up often in standalone Web CMS selections, which is why it did not appear on our 2009 CMS Shortlist. According to Oracle sales pitches, the product has experienced increased adoption in recent years. As the Oracle customer list is very long and Oracle is known for upselling to the install base and for including UCM in larger deals, this sounds plausible.
Depending on your specific requirements, there are several reasons which might make Oracle a meaningful inclusion on your shortlist.
- Oracle has continued to invest engineering resources in the product and made several recent improvements to the WCM part of UCM including usability, personalisation and accessibility.
- As a large software vendor, you may already have a strong existing relationship with Oracle. If this the case, your stakeholders will probably appreciate getting a proposal from Oracle.
- If you have a strong requirement to manage non-web content, eg. documents, this will play well with the product’s strengths.
Before you go ahead and add Oracle UCM to the shortlist here’s a few bullets for your consideration:
- License and implementation cost will require a serious budget. The starting price is either US $115k per-CPU or $2,300 per system user. Moreover, Oracle implementation partners are not known for attractive hourly rates.
- Usability might have been improved, but still existing customers on the newest version of the product are so frustrated with poor usability that they publish commentaries like Oracle, can you improve your poor usability please? by Mark Morrell at BT.
- You will need to learn the proprietary “Idoc Script” language for Site Studio until 11g release comes out.
- UCM is a complex product and will be overkill for many scenarios.
Oracle is planning to release the much-anticipated 11g version of Oracle UCM later this year, which we look forward to studying closer. In the mean time, consider talking to Oracle on getting more information about what’s coming.
Comment on this article by Kas Thomas August 31st, 2009 21:49, Source: http://www.jboye.com/blogpost/should-oracle-be-on-your-web-cms-shortlist/
I would add another precautionary bullet point, having to do with the rights model. Study the UCM roles and rights model carefully and compare it against your requirements; that’s my advice. Maybe @bex or someone with deep UCM experience can educate me here, but I find the UCM rights model a tad unconventional. It defines a security group as a collection of files (not users). It maps rights to roles, then users to roles. Each security group is accessible to appropriately privileged roles.
If you create more than 50 security groups, system performance (initially at the admin level, but eventually at the user level) begins to take a hit, at which point Oracle suggests you turn on a feature called Accounts, which is a more granular, hierarchical permissions model. But if you choose to enable “Accounts,” you can’t go back to a non-accounts-enabled model without losing data (according to Oracle’s own documentation).
The whole thing seems a bit scary to me, but maybe that’s because I don’t understand it, which is not infrequently the case with things that scare me.
SharePoint 2007 Workflow (Visual Studio 2008) Livelink Workflow (Livelink Server 9.7) Workflow Platform Type Framework Engine Electronic Forms Support Yes (InfoPath) Yes (LL Forms Module) Interoperability with Desktop Applications Yes (Microsoft Office Suite Only) No Interoperability with Email Yes Yes Integration with Line of Business Applications Yes No (Actually maybe with some insanely complicated level of integration) Electronic Signatures Yes (WRMS, a bit complicated) Yes (eSign) Supports Attachments Yes Yes Expression builder (WF Logic) Yes Yes Dynamically assigning WF steps to Users With customisation Yes Easy MIS Needs to be built Yes (OOB) Logging Needs to be created in WF Yes (OOB) Interoperability with custom services and web services Yes No Development Required Yes No (except for customisation) Development Language .NET OScript
First of all, it was only since Visual Studio 2008 came out that I decided to start comparing elements of SharePoint 2007 such as Workflow and Business Process Management to other leading Enterprise Content Management systems. Before that exercise would be pointless.
Livelink in this game comes as a seasoned veteran and SharePoint 2007 and WWF are playing the catch-up game here, so I’ll focus more on how much WWF caught up so far.
Windows Workflow Foundation’s (in SharePoint context) biggest weakness is in fact that it’s a framework that requires skilled developers to design workflows where other workflow solution are in a form of an engine, with very easy, user friendly design studio that can easily be mastered by less skilled people (basically non-developers).
That, being a weakness, in fact is also the biggest strength of WWF, as in my personal career, the percentage of clients that were happy with OOB (out of the box) workflow (such as Livelink) functionality is less than 30%.
On the other side customisation in WWF now is almost flawless, it provides variety of highly functional Workflow steps that make use of the latest technology available for .NET, and is provided in known strongly supported languages (such as C#). Whereas in Livelink for example, the whole solution is written in a proprietary language called OScript (written especially for Livelink) which is object orientated (mostly similar to C, has elements like assocs and stuff) with a very good object model but also very outdated making the customisation less then easy (in today’s terms), and since it is proprietary the skills are very scarce (there’s only very few of us in the country that are good with it) making it not very profitable for the clients and service providers.
Also, to add a custom step to Livelink workflow requires creating whole module, installing it and in some cases having to even re-do the existing workflow (if it existed before), they don’t have the code activity type steps (although other WF engines like K2.NET and FYI have it).
Windows Workflow Foundation for MOSS2007 also wins “BIG TIME” with InfoPath 2007 and Forms server with SharePoint 2007 against Livelink Forms. Being able to add various controls from the InfoPath UI to your SharePoint 2007 Electronic forms is a good start, and then controls such as: Repeating Sections, Repeating Tables, Optional Sections, leave Livelink Forms far behind.
Livelink still leads with the tight integration of workflow with rest of the Livelink suite, consisting of much more refined products to better provide various Enterprise Content Management Solutions.
All in all the race has become tight and I see MS SharePoint 2007 Workflow solution leading in near future if the OpenText Livelink crowd doesn’t realise how important it is to keep your solution on the latest Development Platform.
The talk has been for some time that version 10 of Livelink will be completely re-written in Java, we’re yet to see if that will increase the stakes in the game (if it comes any time soon).
SharePoint 2007 Workflow (Business Process Management) Solutions are solid enough to play with the big boys now, even without the help from the mighty (yet expensive) BizTalk server.
LinkedIN question dated May 12, 2009:
I would like your feedback ‘technically’ speaking on this question…
Should we migrate all Oracle UCM content to Microsoft SharePoint and why?
Email responses May 12 – July 14th, 2009:
I think we should migration all UCM content to SharePoint.
From an end-user’s perspective:
1) Searching for content in UCM is painful. The indexing process is slow, and the search results are terrible.
2) The UI in UCM is dated, not very modular, configurable or extensible.
3) It is harder to expose content in UCM to other consumers.
From a Developers perspective:
1) SharePoint is built on top of Microsoft .NET.
2) SharePoint is integrated Microsoft Visual Studio, one of the best development tool in the industry.
3) Extensions on SharePoint can be written in any .NET languages.
4) SharePoint uses modern technologies/tools.
5) It is much easier to write and debug code in SharePoint.
That said, the learning curve is definitely much steeper in SharePoint since you would have to know the entire web related technologies including Windows Server, IIS, MS SQL, and .NET.
I would suggest moving non-web content to SharePoint but only if we can use with the KnowledgeLake Imaging Server. The maturity of web content management features in UCM beat SharePoint. Having said that there are a host of reasons why SharePoint should be considered for library content services.
This would get all technical and non-technical staff more familiar with SharePoint.
We have already found benefits of locating multi-user spreadsheet access in SharePoint. We inevitably will find other organizations that may want help to migrate UCM to SharePoint. This would be good experience for our technical staff.
I still have difficulty locating information in UCM; however w/o KnowledgeLake search we would be in the same boat with SharePoint. Overall I find the SharePoint GUI to be far more intuitive and we would likely get better user adoption.
SharePoint seems to be a good choice for document collaboration.
However, to answer the question in a larger context of where SharePoint might “fit” in our business and what our peers, customers and competitors are articulating in the marketplace is also a good question.
SharePoint shouldn’t be seen as a replacement for Oracle (or IBM, EMC). So far, here are the following reasons:
1.) Records management – although there is some records management capability in SharePoint, the feature set it a bit “underwhelming”.
2.) Workflow – Oracle has a superior workflow capability.
3.) Storage – there is no native support (in SharePoint) for comprehensive storage technologies (like WORM) or any other “write once read many” technologies. (and the marketplace seems a little wary about Microsoft controlling all aspects of their document lifecycle).
These three things basically relegate SharePoint to the “tip of the spear” as far as document management, business process and storage is concerned. Anyof these could change rapidly in the future, but it seems there is more integration development between SharePoint and Exchange and MS Office Suite than for an expanded feature set along these lines. If this is true, it will push SharePoint even closer to the document origination function, and away from a document storage function (ie… greater than 5 year retention and storage).
I think I finally understand that this is what people mean by collaboration.
It seems that asking a company to standardize on any one document management system is like expecting them to buy just one type of file cabinet, and then storing all documents in one single place. For large fortune 1000 companies this isn’t reality. For small companies, this may be just fine…SharePoint may be all they need (so I don’t see much of a future for products like DocuShare)
Maybe we should see the product suite as a document technology “stack” that begins with document origination in SharePoint and ends with Oracle.
For many of our purposes, the shortcoming(s) of our existing UCM system focus on unsatisfactory search results. I assume this could be resolved if we were to rely on UCM for dynamic (changeable) content (and we should do this if we retain UCM). However, from a strictly content management standpoint, I see no dramatic difference between UCM and SharePoint’s potential using the current versions (MOSS 2007 & UCM 7.5.2) of both products. Anytime we rely only on text string queries, we are going to frequently get more in the result set than what is desired.
If we use SharePoint, we need to be able to use metadata in our queries and I understand that the current SharePoint is inadequate unless fortified by something like the KnowledgeLake enhancement for query.
As important as content management is our need to have effective collaborative tools which are substantial even though incomplete in SharePoint (these can be augmented by products such as Oracle Autovue, WebEx, etc.). The development staff seems to prefer the SharePoint characteristics.
It is highly probable that long term product growth and enhancement will be more active in the SharePoint product line vs. Oracle UCM. Likely both products will advance in capabilities over time, but SharePoint seems to have the momentum. In addition, add on enhancement product selection looks to provide a wealth of focused vertical market solutions in the SharePoint environment. SharePoint looks to be more pervasive in the marketplace than Oracle UCM.
If we have the resources to do so, I suggest that our long term internal requirements and our product marketing/sales objectives would likely be more thoroughly served by a migration of the information in UCM to a well governed SharePoint system.
We use SharePoint & UCM in IT help desk support for many of the reasons listed we have gravitated to SharePoint for our KB, Support Portal and Content Repository. Because it is very easy to post information in SharePoint our usefulness of UCM has been declining rapidly. I have not seen or researched 10gR3 UCM and it may be worth looking at before we pull the plug on UCM. As one of the others said the searching for records in UCM is very difficult but SharePoint brings you back a lot of information you don’t want or need.
Technically they do a lot of the same functions but I think that our teams and departments needs more collaboration based application software for staff working remotely more than we they need Information Lifecycle Management. I would like to see an examination of how we currently access content and what types of information we commonly use. This would help us to decide if we migrate to SharePoint or upgrade/repair UCM and/or use a product like the ILINX connector to SharePoint for a combination of the two.
There are couple interesting links regarding the two:
This is the somewhat Oracle-centric article referenced in the above:
I personally would bet on SharePoint simply because I prefer, and am more experienced in c# and .net versus java, and also find Microsoft’s interfaces much cleaner and intuitive.
I felt the following was an informative article on this subject (link to actual document follows at bottom)
Does Oracle UCM Standup to SharePoint?
By Barb Mosher | Aug 8. 2008
Gilbane recently produced a research paper entitled: Information Workplace Platforms: Oracle vs SharePoint which has some interesting findings on which of these two ECM platforms is the best choice for an organizations information lifecycle.
The report findings are based on research on a number of companies who reviewed both platforms for their Information Management needs and what their final decisions turned out to be.
The report, authored by Tony White, Lead Analyst, Web Content Management at the Gilbane Group isn’t a long read, but takes us into the world of information workplace platforms and the pain points they address.
Defining Information Workplaces
This paper is more about educating us on what Information Workplace Platforms are and why they are needed then about the actual technical solution that Oracle or SharePoint plays. Which is a little disappointing because the title would lead one to expect something totally different.
As for a definition then, the paper defines an Information Workplace Platform to include: federated search, portal, content management, records management, document management, rights management, retention management, imaging, Web 2.0, e-discovery, and workflow.
Choosing Between Oracle UCM and SharePoint The research covers three specific customer examples where both Oracle and SharePoint were considered as Information Workplace Solutions. In most cases, Oracle was selected as the solution of choice. Of course, that solution didn’t just include Oracle UCM — it also included the Oracle Fusion Middleware of which UCM is only a part. Does that make it a fair comparison to SharePoint? Not sure.
One point that was discussed was the lack of SharePoint federated search. This is something that has since been resolved with SharePoint. Microsoft also received their U.S. Department of Defense 5015.2 Certification in May of last year, so the report is a little out of date in some respects.
CMSWire asked Tony Byrne of CMS Watch what his take was on Oracle UCM vs SharePoint. “The contrast between the two is a bit more nuanced than departmental vs. enterprise and collaboration vs. ILM. Generally speaking, yes, SharePoint is better suited to the former and Oracle the latter, but the story gets murkier when you dig into the details of each product.”
And it’s the details that we lack in this report.
Alan Pelze-Sharpe of CMS Watch agrees saying, “the thing is to see UCM as part of the Fusion platform, in that context it is a true ECM linking into BPM and a whole array of web services (identity management etc) and importantly supports ILM thoroughly.”
If that is the case, then should we not also include BizTalk Server — Microsoft’s middleware solution — and other MS solutions in the context of a decision on which platform is the better choice.
Aside from that, Microsoft has acknowledged it needs to do some work with SharePoint and associated solutions to provide true ILM.
The report concludes with Gilbane’s own views on which platform is the better solution and it’s obvious they are very pro Oracle — both UCM and Fusion Middleware combined. They do acknowledge that both solutions have their strengths and weaknesses and that they have based their conclusions on the platforms value for ILM.
That being said, SharePoint should not be considered only a “point-solution” for collaboration and basic team workspaces based on this research alone. In the right context it can be the right solution for the right organization. It’s all in the requirements and a full review of the platform’s capabilities — maybe even — shock — piloting or prototyping.
THE ABOVE MENTIONED DOCUMENT CAN BE FOUND AT:
Right now we are using UCM for two purposes, to store some of our active content and host our public facing websites.
I don’t have extensive experience with MOSS Content Management, but what I have seen is that it is in practice not much different from UCM. Technically I don’t see any glaring problems other than getting over the MOSS learning curve. Both technologies have the problem of needing a well defined metadata model to support searching of our content. The one UCM feature we might miss is the conversion of documents from Word, Excel, PowerPoint, etc to PDF.
I have never seen MOSS do web content management, but I have read that the feature exists. We need to do more research into this feature to make sure it meets our requirements.
As a developer I would much rather develop in MOSS. If we are going to be doing more MOSS products and projects in the future, and then migrating our own systems to MOSS would provide us with much needed experience.
It would depend on what our future considerations are for the use of the products.
If the intent is clearly just to be used for collaboration purposes then I think SharePoint is the better product. Why?
I have recently been using SharePoint for multi-user spreadsheet access and used it more extensively on a recent project because that is where they stored all the project information that we needed to access. I find the user interface in SharePoint to be much easier to use and more pleasurable to use. UCM is clunky and outdated. In SharePoint, searching for documents is easier, although the volume of documents was low in both my use cases. In UCM I can search using more of a “google” search for known items that exist in the system and still not get a hit back when I search for it. I have to sometimes use just capital letters, lower case for others, or combinations of both for others to get hits back. So, on paper it is supposed to have better search capabilities but they really stink. I find that I either get nothing back or too much back with UCM when I search.
If our plans are to grow the system and get into more sophisticated workflows, records management, etc. then we should either stay and “suffer” along with UCM (and its clunky UI) or look at using SharePoint with an ILINX connector that could promote the content from SharePoint into IPM, UCM or other document management systems for archiving, BPM/workflow, records management, etc. Use the power of the document management system we have to provide the true ECM capabilities but utilize the easier to use interface of the SharePoint product.
I don’t believe that Microsoft will ever get the product to be an enterprise level product. They have gotten it out there but I don’t think it will evolve into a true ECM product. It will remain a good department level collaboration tool but not go much beyond that. They seem to be able to get products out in high volume to the lower and mid level clients but never seem to really break into the larger higher end customers like an Oracle does. They have tried to do this in the database space and the ERP space but never seem to be able to overcome the hurdles to really get the largest of the customers. Except for those who hate Oracle!
In the case of the project I referenced earlier. Now that the project is over they should be moving the important project documentation to an archive. The project team has been disbanded but the information needs to be stored for long term and then records and retention managed yet easily retrievable. They need a document management system for that. Not SharePoint.
There are many valid points that are raised in these responses. I think one important consideration for this dialogue is ‘context’. The individuals responding to this query all have unique perspectives that are largely shaped on their functional role in the company and their direct experience with clients. At times, we tend to look at things only through our own perspective and not in the broader context. Industry trends and technological limitations can’t be ignored, but again, let’s consider context.
First, regarding the challenges that we face surrounding searching in UCM. We have to consider that the original metadata model that was developed for our Content Server was developed as a prototype when the platform was relatively new. It was then put into production and remained there. The metadata architecture model for Content Server is fundamentally different than the IPM platform which is where our core competency and extensive experience base was rooted. At this point in time, we have a much more mature understanding of the differences in those metadata models, why they matter and how that would play into a design and implementation. It is likely that if we were implementing that whole solution today, the metadata model might look very different, which in turn could significantly impact the quality of the user experience. Secondly, some of the metadata (e.g. author) was complicated because of the content review process that was put in place. If we tried to search for content based on who we naturally thought the author would be (i.e. a Program Manager for project management, etc…), we couldn’t necessarily do that because someone else (admin staff) ultimately would up being the ‘author of record’ because they approved and did the final checkin. Again, all of these things impact the ability to search, and the end users are unaware of the details, only that they are having a hard time finding their content. The question this poses is this; “Is that a function of the system, or a metadata model and review process that could have been refined to make for more effective search capability?”
Second, the ways that we use Content Server (and all of its associated modules) is extremely limited. We’re only using basic document storage and WCM functionality for our websites. There are myriad components used by other customers (i.e. Collaboration Management, Digital Asset Management, Records Management, customized applications) that significantly enhance the value of the platform to any given organization relative to their needs. We’re only going just below the surface. When you combine the ability to customize the interface, leverage the content repository, and include additional database support, and creative design, some very valuable and meaningful applications can be produced that add significant value to a business. The training application that we developed for medical sector client in 2007 is a perfect example of that.
Third, from a WCM perspective, as mentioned above, there is no contest that Content Server is still more mature than SharePoint.
Fourth, all of the comments above about the flexibility and ease of development (for skilled developers) in the SharePoint environment is true.
In Summary, SharePoint will likely go the way that Windows did back in the late 80’s and early 90’s. It won’t be able to be ignored forever. When Windows first came out, it was going head to head with IBM’s OS/2 operating system, a clearly technically superior product. Through some very brilliant marketing strategies, the Windows OS started showing up on the “new computers” (IBM clones). People started using the clones because of the lower cost and Windows came with it. Subsequently, applications started being developed for it, and over time, it became a defacto standard. Over the years, it matured and eventually became a stable, meaningful and valuable player in the market place and technical space. That was then, this is now and Microsoft is now the largest software company in the world with vast resources for development. SharePoint is not going away. It will grow, mature and a large application base will be developed for it. We have embraced it, learned it, and are developing in the environment to push the envelope for it the same way we do for other platforms to fully realize the possibilities. That being said, it should be considered in the broader context of content management, all content management. It is a tool that is part of a larger tool bag. Our model and methodology has always been to use the ‘best of breed’ for any given situation or solution. I don’t think this discussion is any different. We should leverage the right tools for the right job(s). We’re starting to do that as our knowledge and skills with SharePoint evolve, I think we just need to be thoughtful when approaching decisions to eliminate entire platforms and ensure that the decisions we are making are informed decisions and that we are considering things in the broader context.
The information contained in this document does not necessarily reflect the opinions of the company. Individual names have been removed. The intent is to share the thought process that may be going on in many organizations that are grappling with the same decision being discussed in this document.
It’s been 7 years since the last major release of Vignette Content Management (7.0). If you wondered about the fate of the promised last year VCM V8 in the light of Vignette’s (news, site) various tribulations and the Open Text acquisition, worry not. It is officially here.
Using descriptives like “ergonomic design,” “engaging web experience” and “ultra modern personalized workspaces,” Open Text and Vignette are hoping it would have a much better installed base response than VCM 7 seemed to have when it came out.
What Changed From VCM 7 to 8
The looks is one of them. As Vignette told us before they became part of OTEX, UI has undergone a major rework in this version.
GUI Changes and New Functionalities
There are two main views in VCM 8:
- Preview environment for page-centric, in-context management approach
- Work spaces for content-centric and bulk operations
Content workspaces in VCM 8 where content is managed independently from the presentation.
Personalized floating menu on the right to manipulate content at the page level. Sample sites are shipped with out-of-the-box content type, workflow and presentation assets.
Users will see multiple productivity enhancements like smart lists, last used, quick actions, search and filter, shortcuts, AJAX-enabled no full page refreshes, minimization of pop-ups, etc.
In VCM V8, drag and drop is enabled in multiple places. The interface got more user-friendly action label names (e.g., save and close) and easier way of getting to most common tasks via tabs and menus, with easier access to system metadata.
Re-order and add items, browse content in contextual, multidimensional workspaces by site, content type, folder, category or explorer views.
For marketers, a new concept of vanity URLs allows to explicitly add a new URL to any page for marketing campaigns and landing pages.
In-context management is one of major improvements with the ability to add new content in the preview environment and create new pages based on default layouts with default content that can be kept as is or changed.
Publishing and workflow have turned into more of a couple of clicks operations without users being exposed to multiple technical concepts.
Site Cloning and Presentation
On the presentation and site management side, more functionality like site creation and site cloning has been exposed to non-technical users allowing for shorter time to market and (after it’s all set up) not having to go to IT. The new presentation management capabilities include theme management, page by page, or template by template management.
Changing page layout from in-context view
This approach for layout management is based on default templates or preregistered CSS and XHTML.
VCM 8’s themes concept allows to keep the content the same, but change the look and feel.
New sites can be created (or existing sites can be cloned) using one of the existing implementations as a template.
Upgrading From VCM 7 to 8
Knowing that migrations and upgrades can be cumbersome and learning from Microsoft’s Office ribbon failure, Open Text decided to take a different approach this time. The vendor introduced a compatibility mode, where customers can run V8 in either the new or the v7 GUI, while keeping their old extensions with the old V7 look and feel.
VCM 8 compatibility mode
When installing the new product, customers can choose to upgrade to the new GUI, or completely skip it for now and introduce it at a later stage in the upgrade. Might be a bit confusing at times to see different portions of the system looking different, but makes sense for those looking to do a progressive upgrade and adapt to the new interface at their own pace, without much investment in retraining everyone.
As far as upgrade efforts go, Open Text says the most optimistic scenario is that it may take a day max to run the upgrade from backup to install to launch. Depending on the number of customizations and the level of implementation complexity, it may take months to upgrade to V8.
All is Well That Ends Well
So, despite some concern, Open Text Vignette Content Management V8 (with a video and all) is live and brings considerable improvements and welcome additions. One wonders though whether VCM V8 will be the turning point for Vignette as a long-standing, veteran product on its own, or as part of the Open Text ECM suite. Will we see magical revival, or milking of the installed base cash cow? Is it the very last release of Vignette Content Management in the days of our lives?
My Oracle UCM improvement wish list
Am I missing any of your issues?
1. Publisher control
Oracle UCM has poor ‘granularity’ of permissions and no obvious back end to see who has access.
The permissions only allow two (publisher) levels, a limited ‘only edit what’s already there’ and a far too powerful ‘does lots of complicated stuff with a very complex interface’. This suits organisations with a few powerful people in a central group, but not BT’s intranet governance model which has decentralised publishers.
A hard coded menu item called ’switch region context’ has no place on an interface I expect a large number of users to use, and that’s only one part of a convoluted process to add a new page.
The user ids do not integrate with other user ID systems and it is difficult to integrate this with other processes (e.g. make sure that only people who have done training have access).
2. Quality of web pages
Oracle UCM should never be able to generate invalid code.
The conversion from MS Word is very poor ‘out of the box’, producing inaccessible and invalid code. With a lot of work – BT was able to improve this but never approached an acceptable level. Oracle UCM needs only to allow ‘well written’ MS Word documents (i.e. only accept well formed documents) and to reject (with explanations) documents it cannot convert to valid, accessible pages.
The site studio interface is poor and difficult to apply standards to. The browser version support is difficult and you have to rely on using admin permissions to install a clunky java applet. I don’t know if the applet is usable to people with disabilities. If it must use ‘rich’ interface elements then accessibility must be considered.
3. Template creation and management
There seem to be few well written components to use in the templates. Additional features (e.g. embedded video, RSS) need to be custom written and template specific. That’s a maintenance headache waiting to happen.
Management of templates appears to be awful. This encourages poor re-use of template development resource. It’s hard to quantify the effort required creating a template, but it seems to be excessive compared to other competitors. BT’s aim is to reduce, not increase, costs.
Please help me to help you with Oracle applications’ usability by commenting.
There is no best web content management system. But there is probably a good Web CMS for what you’re trying to do. With that, we kicked-off a heated debate on Inconvenient truths and unsolved WCM industry challenges at the J.Boye 2009 conference in chilly Aarhus, DK.
And despite strenuous protestations by Aarhus mayor, Nicolai Wammen, who addressed the delegates at the opening ceremony today, the weather has not improved. Thankfully plenty of hot air warmth was radiating from the conference discussions, including the one on how to fix the WCM industry with all its problems and challenges. But why fix it, I ask, if it ain’t broken?
It Ain’t Broken, a.k.a. The Bright Future of Web CMS
CMS Watch’s Jarrod Gingras was certain there’s nothing to fix. The competition is healthy, the buyers have many options and that’s good for the marketplace. Clearly, there’s some frustration, but it is tied directly to the disconnect with end-users.
Gingras continued on to provided a rather positive outlook into what’s coming for the WCM industry in 2010:
- Consolidation: There will be some consolidation, but nothing earth shattering or landscape changing.
- Innovation and usability: Complex enterprise vendors will not concede to smaller vendors.
- Platforms and products will continue to co-exist.
- SharePoint 2010 is better, but will still fighting for deals.
- Open source will see continued interest, but will have to improve usability.
- SaaS: The hype may decrease a bit, acquisitions are expected.// //
The Big Questions
Janus Boye, the conference organizer, and Jon Marks, lead of development at LBi (both were also presenting at the session), didn’t seem to agree with Gingras and posed the “big question” — How do we fix WCM? One way is to pay more attention to user experience and the needs of end-users by involving them more throughout the implementation. Change, as well as content, needs to be managed, said Gingras.
Marks, also known as McBoof in the Twitterverse, addressed several WCM issues:
Whose fault is it when a WCM project fails?
The ecosystem is sometimes broken, and we see a lot of finger pointing from customer to vendor to implementor. The products are not perfect, but they are what they are. As implementers, we do the above all the time, but we should know better, even if the customers do silly things.
Does the term WCM make any sense?
WCM as a term might be too big and too vague. WCM means too many things to too many people. The term needs to be sub-divided and we all need to take into account all its sides with respect to blurring boundaries among WCM, portals and SoCo.
Why are RFPs so bad?
Everyone is still doing checklists and matrices. This process encourages vendors to do what they hate. As a result, RFPs ask questions that achieve nothing.
The current ecosystem does not foster standards well. Let’s pray for CMIS and those that follow. Web CMS standards should not start with ‘J’.
They can be manageable with correct tools and processes. The migration of the delivery side is almost always a rewrite. And don’t believe vendor marketing bollocks.
How to Fix WCM?
In order to fix this mess, Marks suggested that one should:
- Align goals for success when starting a WCM project
- Plan for disaster
- Don’t make monoliths
Boye brought up the fear and intimidation tactics used in the industry. If you’re waiting for WCM vendors to wake up and fix the problems, it is unrealistic. Vendors will not change until you make them change. The big problem here is that buyers are not talking to each other as much as they should. But Boye sees the buyers acting as catalysts for change by:
- Openly sharing lessons learned
- Stepping forward as references
- Forcing vendors to take responsibility for WCM implementations
- Identifying and adopting industry standards.
Quotes of the Day
- Not every project fails, there are some happy customers (Gingras)
- Too many projects fail (from the audience)
- So many systems look and work very much alike (CMS Watch analyst Theresa Regli)
- Q: I am asking a simple question… We made the wrong choice when selected our last WCM vendor. That set us back 2 years. What do you suggest we do instead of using RFPs and vendor interviews? A: Pay 2-3 vendors to build something and pick from the best. (from the audience, answer by Marks)
- The most successful projects paid for the POC, people cringe at the bake-off idea, but what is the cost of a failed implementation? (Gingras)
- The terminology is flawed. We should focus on the dialog. (Umbraco’s founder Niels Hartvig)
- Don’t be afraid you will lose your job if you share knowledge. There are so many educational prigrams and books about the industry. (Boye)
- Some of the problems arise from the sheer diversity of people using WCM products. (from the audience)
- Standards are there, but most people don’t even do the basic things like [compliant] HTML. (Boye)
- WCM vendors are having a hard time keeping up with the content.. and customers cannot integrate a Twitter widget into the CMS they bought for managing press releases. (Regli)
- CMIS is still thought in the general audience as the document management-specific standard, even though DM and WCM are both subsets of ECM and the hope is that after version 1.0 of the standard it will be extended to WCM as well.
- Buyers should think about where they want to go, not which “car” (=tool) to use to get there. (Boye)
We’re live from the City of Smiles, where the J. Boye conference kicked off today in Aarhus, Denmark. The first day is the Tutorial Day with topics ranging from content strategy and governance to persuading people with digital content.
One of the tutorials — CMS selection: the process, the pitfalls and the best practices — was presented by analysts Jarrod Gingras of CMS Watch and Peter Sejersen of J. Boye.
The session was prefaced with a statement that there’s no perfect system. Making your CMS requirements too special may make it nearly possible to find a system that will fit them.
Dos and Don’ts in Choosing a CMS
When looking for a CMS, look for both a software vendor and a system integrator, if they’re not part of the same package. Make your project attractive for bidders. Make it easy for them.
Here are some additional dos and don’ts that were offered:
- Do not employ a detailed scoring methodology
- Do not develop a full list of requirements
- Buy a scoping exercise before buying a full-blown
- Control the dialog with vendors
- Do not look beyond 3 years
What to consider:// //
- Project timelines
- Creation of a business case
In the best cases 3-7 months, worst case a year or more, according to Sejersen, who quoted J. Boye’s report on Selecting a CMS. Your level of complexity will affect the timeline, added Gingras. This is not a trivial process, but you need to move fast. In EU, the open tender process may make the process even longer, when vendors have around 40 days to respond to your RFP.
Who Needs to be Involved
Marketing folks, developers, a project manager, system admins, application admins, website manager, etc. This mix of groups depends on your organization. Engaging the project champion and the project manager are absolute musts, said Gingras. It’s a political battlefield you want to navigate carefully, when many (or too many) people are involved and you need to make a decision. You need to communicate and make it clear what their role is. They’re not making the decision, but provide input.
Don’t select a CMS before developing your requirements. Gather the requirements only after the business case is completed. How do you justify acquiring a Web CMS system? There are 3 dimensions to a WCM Business Case:
- Greater efficiency and reducing costs and frustration largely through automation
- Reduce risks (a system can provide insurance, but also introduce new risks)
- Enhance value (look to get more value from your content, improve quality by implementing standards)
What Should be Included in the RFP?
Understand internally what you’re trying to achieve. The short and effective RFP is not more than 10 pages. It can start with a brief intro to the project and your organization and include project scope, goals and characteristics. As well as:
- Specific needs and business scenarios
- Technical architecture and standards
- Structure and design of the platform
- Role of the vendor
- Schedule and evaluation criteria
Do not issue checklist RFPs – unless it is absolutely required in your organization. When vendors get it they throw it to the most junior person in the organization to fill out based on a previous example. In many cases not a lot of time get put into these answers, and the possibility of cut and paste from a previous checklist is higher. A checklist alternative is providing specific, testable scenarios. But don’t go overboard – 5 to 7 scenarios should be enough. Entering a CMS vendor contract, is like entering a marriage. You have to like the people you work with.
Key vendor differentiators:
- Vendor intangibles (chemistry, company focus, financials, trust, quality and experience)
You don’t want too many vendors on your shortlist. Somewhere between 5 and 7 vendors is ideal, and more than 10 is too much. Do your research and get input from blogs, peers, network and external analysts. Do not assume that there’s a best vendor/product. The analysts acknowledged that many reports and analyses exist but noted that they are useful as sources of information, but will not tell you which the best option is for your specific context.
Reading and Evaluating Proposals
After you get your RFPs back, look for indications of heavy copy and paste activities. More importantly, check if vendors understood your scenarios. Check their references. Many are willing to share their experiences (the good and the bad) and can provide candid feedback. And, of course, look for pricing and license models. Finally, develop an evaluation matrix, so that you don’t have to waste time on the losers.
When running a demo, make sure your team is well represented, but keep your team size limited, as 100 plus people is probably an overkill. Beware of canned demos. They often look good but fail to address your specific needs.
Give vendors the rules up front and tell them what you want them to show. This way they know to keep focused on your scenarios. Ask questions about pricing and be sure to understand which modules are part of the core system and which ones are add-ons which might cost more.
If you’re thinking of involving a system integrator, do include them in your demo processes. Keep in mind that licensing cost is usually the smallest part of the total cost of ownership of the CMS. Usually, it is recommended to have a scorecard to evaluate each demo. Formats may vary, but look for pros and cons in two areas:
- product fit
- vendor fit.
With open source CMS, you have to look at the vibrancy of the community. Make sure to debrief immediately after the demo, so you document the impressions properly and don’t forget who said and showed what. One of the ideas in the audience came from Martin White of Intranet Focus who suggested to videotape the demos (even though vendors may not like this) and compare them later.
Bake-off and POC
When you’re down to 2 or so vendors, you can do a bake-off using real scenarios and real content to get the real feel of the product and work with real people from the vendor or integrator side. It can be expensive, but what’s the cost of a failed implementation?
Do a proof of concept before you dive in completely. One of the benefits is to get your users to try the CMS.
Making the Final Decision on a CMS
How do you make the final decision? Blindfold, roll the dice, flip a coin? Do the due diligence and evaluate the vendors from the point of view of:
- Viability and stability
- Support and community
- Services and channels
- Strategy and roadmap
You can also attend group meetings and user conferences, which would be perfect opportunities to check references and get honest opinions. You’re not just buying software — make sure that you meet the people you’re going to work with.
It is very challenging to decipher the licensing models, as they can be CPU-based, user-based, server- and domain-based, etc. etc. Don’t take this exercise lightly and don’t underestimate the costs. The advice is to budget 2 to 4 times the license cost to allow for integration, consulting and customization expenses. Open source systems are not entirely free, keep that in mind when making your vendor selection and deciding on the budget.
Finally, yet importantly, don’t just sign the contract before you negotiate.