Intelligent Enterprise

As an "Enterprise-focused" content management analyst, I am asked two basic questions on a regular basis. The first is "what about SharePoint?" The second is, "what about imaging?"

At many conferences, and regularly via e-mail, people ask me about imaging in the context of ECM. Imaging is the major cost that most projects either forget about or dramatically under budget for. During the buying process it's all too easy to get caught up in the flurry of believing that every file will soon be digital, even though paper is clearly here to stay.

If there is one word I hate to hear used in this industry it's "compliance."

To me it's like fingernails down a blackboard, and frankly if I never hear it used again then I would be a happy man. Of course I have to endure the word in virtually every article and vendor press release I read. I don't like the word because it is a blanket term that used without context is totally meaningless, yet it's a word (much like governance) that sounds impressive and few people in the room will admit that they don't really understand it. Well let me be among the first to point out that the Compliance Emperor often has no clothes.

Oracle announced on Monday that it is entering the archiving market with the release of "Universal Online Archive." UOA positions Oracle to compete more directly against EMC and IBM in the e-mail and messaging archive space. It's interesting as only a year ago nobody was much interested in archiving, but in the past twelve months we have seen everyone from Dell to Google try to gain a foothold, and the market shows no signs of slowing down. It remains a chaotic and confusing sector risking a consolidation (which of course might not happen, or at least not soon).

UOA is built on top of Oracle 11g with technology acquired from Stellent, as well as from e-mail capture experts ZL Technologies. Why would Oracle be interested in archiving e-mails you may ask? Well the answer is simple: because there is an awful lot of it. And by archiving it, the messages will move out of Microsoft's servers and into Oracle databases. Remember in most firms e-mail is by far the single largest type of "data."

The world does seem to love an XML fight. Last week, Microsoft scored a goal by getting its OOXML standard ratified as an international standard through ISO (International Organization for Standards) — a definite point score, since there were many other parties fighting tooth and nail to prevent this happening. OOXML is an important standard, with critical implications for the industry as a whole, and therefore represents a standard that we need to look at dispassionately to assess its true value and potential impact.

So HP finally made a move into the world of Enterprise Content Management by acquiring Tower Software of Australia. On the surface it's an unusual match for HP, as many had expected them to buy one of the top tier players such as Interwoven, Vignetteor even Open Text, but on closer consideration it's a move that makes sense. Revealingly, HP does not call this an "ECM" deal and focuses on the e-discovery and compliance benefits from Tower's addition, so it's possible HP has further moves to make if it wants to get serious about offering broader ECM services à la IBM.

So Google has launched a product called Google Sites to compete with Microsoft's SharePoint. Even at first glance it is no SharePoint killer and is (as is normal for Google) more of a Beta product then anything that shows real maturity.

The real discussion we need to be having is whether these tools are the "productivity" applications they claim to be. Both Google and Microsoft have the means and depth to produce impressive tools, but just because something is usable and quick to deploy does not mean that by definition it is a good thing, particularly when it comes to managing confidential information. This is where things can go badly wrong. As we have discussed elsewhere, one of the major problems with SharePoint is not the technology, but its viral growth. What seems good to one end user can represent a compliance and auditing nightmare to their employer. With Google we have the added complication that by using their collaboration tools we are also entrusting our information to their care. Google has a strong and valid approach to confidentiality, security, and privacy issues, but it's not one that fully assures everyone.

My thanks to our friend James Robertson for pointing to an important UK study that debunks many of the "Web 2.0" and "Google Generation" myths that currently abound. I have bit of a reputation as a cynic, but the Google Generation is something of which I have simply seen no real evidence, despite vendors and fellow analysts arguing loudly about its importance in today's workplace.

My colleague Jarrod forwarded a link to a news story of how one person deliberately destroyed seven years' worth of corporate content/data with ease. Meanwhile, I had another tab open, regarding the White House's inability/refusal to archive e-mail messages, and had just finished reading about the loss of a laptop containing the personal details of 600,000 people — quite a busy day for data destruction.

So EMC (read: Documentum) acquired Document Sciences. The announcement came over the holiday period and has been the topic of chatter in the blogosphere. It's an acquisition that makes perfect sense for EMC as it continues to reposition Documentum away from the traditional complex document management activities that established the firm — a market that is under attack from Microsoft and Open Text — and more into high-value, transactional document management and archiving.

It was easy to get lost in the maelstrom that was Oracle OpenWorld. With more than 45,000 visitors, it takes over San Francisco each year. Its "appreciation" parties are legendary — this year three simultaneous stages featuring Billy Joel, Stevie Nicks, and Lenny Kravits — and made it clear (as if it were necessary) that Oracle is an industry giant.

In the wake of Oracle's acquisition of Stellent, enterprise content management (ECM) was firmly on the agenda. Or rather Fusion Middleware was firmly on the agenda, and ECM was discussed as an integral part of it. The Stellent acquisition was very different from the (admittedly much larger) acquisitions of Siebel and PeopleSoft, as they have continued as separate stand-alone business groups and products (albeit now running on the same middleware and platforms). Stellent the brand is no longer. The acquired functionality has already been repositioned and in some cases rearchitected as a set of services that coexist with other standard functions such identity management, security, integration, and BPEL.

Has enterprise search finally reached the tipping point? Yes, seemed to be the conclusion of many of the experts at the recent Enterprise Search Summit West in San Jose, CA. Of course these are people with a vested interest in believing this to be the case, but there was palpably more energy, commitment, and enthusiasm for enterprise search than in previous years. I chatted with Sue Feldman of IDC and she confirmed my feelings that something has changed, that enterprise search is entering the mainstream, and that there is no turning back.

It would be interesting to know how many failed enterprise content management (ECM) projects stemmed from the wrong deployment methodology. I was pondering this after a discussion with Liz Ure recently in London. Liz is the Head of Information Strategy for the Scottish Government and she talked about the inappropriateness of methodologies that emphasize implementation, rather than change.

There are any number of methods being touted for ECM, from Agile to Prince2 through any numerous of (AA-style) "step" approaches. I have long argued that any methodology is better than no methodology, and these are all fine in their way. But to Liz's point, they all emphasize successful system deployment, with a focus on "going live."

At every talk or seminar I give on enterprise content management (ECM) technology, I stress to attendees that they have many different options -- including software as a service (SaaS). It's still early days for SaaS ECM, but the approach is now joining open source as a viable alternative to traditional software licensing models. So I have no problem telling buyers that SaaS may be worth considering, but actually recommending that they pursue a SaaS option is still something of a stretch for me.

There are some very strong selling points for SaaS, including the relative ease of deployment and potentially much lower operating costs, not to mention the obvious appeal of obviating IT burdens such as patch management. Although ECM SaaS providers remain few in number, the scope of their offerings is widening. For example Xythos sells a dedicated SaaS option that seems to be building out a decent customer base for its basic-but-proven document collaboration services. SaaS ECM market leader Spring CM has a quite full offering — the equivalent of many traditional ECM vendors — with a wide range of productized applications, ranging from mortgage processing to hospital bill reconciliation.

One of the great divides in the ECM world is the gulf between (and different understanding of) the needs of records management, on the one hand, versus IT storage on the other.

Archiving, storage and retention all sound like similar disciplines, and to hear some IT folk speak, you could be excused for thinking they are one and the same thing. All too often very expensive electronic storage hardware and software systems operate in ignorance of, and non-compliance with, legal and regulatory demands.

Inclusion in the Gartner Magic Quadrant (MQ) is believed by vendors to have a very positive impact on sales. In the 2007 MQ for Enterprise Content Management, published late last month, it's clear that little (in Gartner's view) has changed in the ECM world. Well, we beg to differ: 2007 has been a period of major change! And so rather than harping on perceived weaknesses in this highly influential document, let's point out where the analysis in the CMS Watch ECM Suites Report differs from Gartner's.

Enterprise Content Management (ECM) vendor Open Text recently announced that it will deliver a major upgrade to the acquired (ex-Hummingbird) eDocs technology for the Legal sector. Not earth-shattering, but important news nonetheless.

For starters it will come as a big relief to the very substantial customer base in the Legal sector that Hummingbird had built up before its late 2006 acquisition by Open Text. Secondly, it reaffirms Open Text's commitment to building on other repositories where sensible — in this case Microsoft — the platform that dominates Legal.

Microsoft SharePoint continues to grow apace. In a presentation to financial analysts earlier this week, Microsoft stated that in the past year it has seen 35-percent year-over-year growth and revenues of a staggering $800 Million. The company claims is has shipped 85 million seat licenses to 17,000 customers since the beginning of SharePoint time (in 2001).

Just a year or two ago, a major upgrade to the Documentum ECM platform would have been dominant news in the industry, but things change, and quickly. D6, the latest version of EMC's flagship enterprise content management platform, is undertaking a gradual roll-out through Q3 2007 to muted fanfare. EMC and its investors have high expectations for this new version, as the company's Documentum Content and Archiving division has shown only modest growth of 5 percent year-over-year, lower than most competitors.

This week, Oracle announced 11g, the latest upgrade to its flagship database. The announcement brooks great interest within the ECM community because, as we detail in the ECM Suites Report, so many ECM tools (including all the leading players) utilize the Oracle database.

Of particular interest is enhanced support for "LOBs" (Large Objects), such as documents, drawings, images, and so forth. Oracle says 11g can now provide:
• Comparable performance to regular file servers for access to large files
• Greater compression capabilities
• The ability to encrypt LOBs within the database environment

This year's big Drug Information Association (DIA) conference in Atlanta concluded with something of a shock for enterprise content management (ECM) vendors. Microsoft SharePoint will now compete directly with established Pharmaceutical and Life Sciences-focused vendors EMC|Documentum and Open Text. Pharma was supposed to be sacred ground for the big ECM vendors -- one area of turf where few thought Microsoft would tread.

In the content management world, I sense something of a backlash brewing against SOA (Service Oriented Architecture), but I wonder how real or or even practical this is. With most Fortune 2000 firms already way down the SOA path, there seems to be no turning back. At the enterprise architecture level, there is no Plan B.
So the issue for me is not whether SOA is the way forward for ECM, but rather how seriously some of the ECM vendors are embracing it.

The story begins with CA (Computer Associates), which spun off its Ingres line into a separate, open source project. Ingres is now teaming up with open source enterprise content management (ECM) provider Alfresco. The Ingres "Icebreaker" product (linux + database stack) will offer an ECM option provided via Alfresco.

Another datapoint to contribute to my growing belief that large enterprises are now reembracing reengineering. It seems that there is only so much streamlining you can do until you reach a point where you need to completely rethink a situation. That point is being reached by more and more large organizations, and radical change is now on the agenda for banks, insurance companies and manufacturing firms globally.

Tony Byrne has provided some advice to vendors regarding product demos. Those ten points make essential reading for vendors and customers alike, but there is another perspective. Since I have personally sat in on those demos both as a buyers' advisor and as a vendor (system integrator), I need to add three points that customers should keep in mind when asking vendors to demonstrate their products:

Transactional document management (high-volume throughput of relatively static documents) has long been dominated by IBM and FileNet. EMC this week announced that it, too, wants to compete in this lucrative market. Hence as part of its forthcoming Documentum D6 release, they have announced "TCM" (Transactional Content Management). Currently, TCM is essentially a user interface module for high-volume scenarios, to be complimented at a later date by changes to the core D6 platform as well as better application of EMC's Captiva technology.

"When Search and ECM Collide" was the title of a tutorial I ran in New York this week. It's an important topic to explore as both Search vendors and ECM vendors both seem to believe that either

A. They can do without the other,
B. They understand the other fully and see the opposing technology as simply a minor supporting or interfacing toolset.

At the recent AIIM Expo I had the privilege to lead a panel, "When E Means Big," where customers revealed lessons learned in truly huge ECM deployments, as opposed to the "regular" world of ECM. What struck me was how the discussion barely touched on technology, but instead gravitated toward such issues as governance, strategy, funding and ownership. A strong consensus emerged on the need for a detailed mid- to long-term strategy for ECM, that business cases needed to take into account cross-departmental processes and concerns, and that nearly all ECM at this scale is underpinned in some way by BPM (Business Process Management).

And most surprising: none of the panelists saw compliance as a key reason to deploy ECM tools. In their minds the only grounds for procuring such systems was good old-fashioned cost reduction and process improvement. This viewpoint appears to fly in the face of surveys and polls suggesting the opposite, but in line with anecdotal evidence for consultants and SI's who contributed to our recent ECM Suites Report research. Compliance is important, but improving the bottom line is essential...

Are SaaS (Software as a Service) options viable for enterprise content management (ECM)? I hear that question with increasing frequency, and frankly, it's a difficult one to answer. There are plenty of vendor options out there - from pure plays like Spring CM, to hybrids like Xythos and Treno. But just because there are plenty of options doesn't means it's a particularly good idea.

I can see the logic of Basic Content Services (BCS) being delivered through the SaaS model, but full-blown ECM deployments seem much more of a stretch. In the world of ECM we are typically looking at complex processes with integration into legacy systems -- it's hard to see the match up with SaaS.

Yesterday, Salesforce.com, the hosted CRM giant, announced it will enter the ECM sector with Web 2.0-style collaboration software it acquired earlier this year from a start-up called Koral. It's a bold announcement that boasts the Salesforce Platform will "manage all enterprise information on demand - structured and unstructured." In fact it goes even further to state that Salesforce Content will "liberate customers from....software like EMC|Documentum..." so said charismatic CEO Marc Benioff.

Currently on Intelligent Enterprise you can read my recent review of Xythos Document Manager Version 6. Though I have a short fuse with those who say that "basic content services" (BCS) are all that any enterprise needs for their ECM requirements, I remain nonetheless supportive of "lite" offerings.

Last week, IBM clarified its ECM intentions via a series of analyst briefings. In short the recently acquired FileNet P8 product will become Big Blue's main enterprise content management (ECM) offering, although legacy IBM products will remain supported for the foreseeable future. The latest version of P8 (version 4) is certainly a powerful solution for buyers of large-scale ECM systems and will benefit over time with the absorption of other IBM products, for example in enterprise search.

1. You are running your enterprise content management (ECM) system on old hardware and operating systems that need upgrading

2. The average size of electronic documents has grown to a point where current network bandwidth is insufficient to deliver documents to the user in a sensible time frame

3. You are delivering native format files to end users rather than making use of rendering functionality

4. You have allowed your users to dump content into your repository without concern for process, rules or structure, and now it's a humungous mess

5. You are running a centralized system but really should have a distributed one

James Governer has prompted an important discussion on his popular blog regarding ECM and Security. He raises some very good questions while lobbying enterprise buyers to team with him to pressure ECM vendors to respond. I'm sure many ECM vendors will be secretly annoyed about this, for they pride themselves on their security capabilities. But it points to two different perspectives around security. The Architect views security as stopping bad guys from getting in (the Firewall Syndrome). The Document Management view casts security as assigning permissions (the ACL syndrome).