The EMC World 2007 conference, while under EMC World’s wing, still felt like the session count wasn’t that far off.  I remember a few grumblings, but nothing documented, so I’ll just use that as a baseline with the understanding that the previous year was at least as good.  I believe 2007 was the year that the long-standing pre-conference tutorials vanished, so calling it even is a throwing EMC a bone.

Let’s See Some Numbers

• 2007: “Momentum Sessions”=120 Sessions, Content and Information Management Tracks (includes Momentum and lots of Developer)=178 sessions.
• 2008: No “Momentum” grouping, just a Content and Information Management similar to above…104.
• 2009: Counting the sessions in the Momentum book, which includes “Developer” sessions…88.  Using the previous methodology from 07 and 08, you get 96.

2007 to 2008, WOW!!! That was impressive. We spent so much time running around that we were just happy to find our session, much less worry about an alternate session to attend in 2008. The 178-104 is a fair comparison as those are both identical EMC categorizations.  That is a 41% drop!  No recession talk a year ago, much less the 6+ months before when they planned it.

The drop from 2008-2009 is less severe and I’ll even call that a recession downturn with a drop of 7.6%.

So let’s get to the meat. 120 Momentum, non-Software Developer Conference, sessions in 2007 and a recession adjusted 95 sessions (88*1.08) is a drop of 21% over 2 years. WAIT A SECOND!!! Those “95″ sessions INCLUDE all the CMA developer sessions (these 17 Documentum ones) that weren’t included in the 2007 Momentum count of 120.  Taking out the 17, and keeping the recession adjustment consistent, that is now 120-77 which is a 36% drop!!!  Over a third of the Momentum sessions gone!!! (Non-recession adjusted 120-71 gives 41%!!!)

Now, anyone with better numbers (*cough* EMC *cough*), please contribute. I am not counting double-sessions or Birds-of-a-Feather sessions.

Where Did The Sessions Go?

If anything, there should be more sessions.  EMC has acquired more technology under the CMA umbrella.  The X-Hive acquisition by itself adds a few sessions.  So with the same number of sessions being offered, there would still be a hit to the core sessions.  Even with the advent of SourceOne, which is home grown, we still see a decrease.

I’ll tell you right now that there are more options and technical details for a developer to learn than there were just two years ago.  DFC and WDK are still here and not going away, but we now have TaskSpace.  There is xCP and DFS.  Composer isn’t easy yet as it is still maturing.  Developer sessions, just in themselves, should be growing, yet they aren’t either.  Shoot, there were 15 breakout session slots filled, plus 2 un-filled, by the CMA.  If it wasn’t for a few long sessions, 2 of them I think, you could single track and hit all of the CMA Developer sessions!!!

Anyone miss the “Meet the Speaker” socials or the “Meet the Experts” gatherings?  How about those pre-conference tutorials?

What gives!?!?

For one thing, there is no longer a Call For Topics.  For the past two years, EMC has not allowed users to submit topics to present.  We, the attendees, got a lot of case studies and real-life examples.  There were more business users telling their stories, and TRADING their stories with each other.  I learned a lot from those people over meals and between sessions. I might only see them at Momentum, so I valued that time.

I see other consultants all the time and only the ones that you know well share their mistakes with you. We are in competition, so it pays to play it close sometimes. Users share because they can.

Each year, there seems to be less business users attending, and they are the ones with the problems we are trying to solve.  Content Management is not an IT problem.  Documentum is not used by IT.  It is used by the everyday, to steal a term, Knowledge Worker.

Go to 10 companies with Documentum. In 8-9 of them you will find end-users that know the name “Documentum”.  Go to 10 non-IT companies owning primarily EMC storage, and no Documentum.  I bet that in 8-10 of them, the everyday end-user, ones whose only interaction with IT is the helpdesk, won’t know the name EMC.

Different products, different users, different stories.

Storage can be measured in tests.  There is a lot to it and I don’t want to take away from anyone on that side of the house. Content Management is about people. It is about managing change, defining processes, and getting people to buy into the technology. It is not just how to deploy the technology.

We don’t only need to know what the next features are in Documentum.  We need to know how people are successfully gaining user adoption.  How are they selling the ROI to the managers?  How they are getting the system working with legacy systems, and the political hurdles they overcome.  I know DFS can do things, but how did Uber Pharma make it work with their legacy testing system?  How did they upgrade from 4i to 5.3 to 6.5?  What were the lessons learned? How did they manage the risk?

Fill the gap. Bring back the Call for Topics.  Bring back the business users.

Make the conference about users again.

I was thinking about talking about my plans for EMC World, which looks pretty good, but the I realized something…This will be my 10th Documentum User Conference!  So I have decided to flash back a bit to the previous conferences.

• 2000, Momentum Vegas: Ah the Venetian. This was my first look at the wider Documentum world. I was with Infodata Systems and stuck in the booth.  We were showing Plumtree and Documentum together that year. The system was beautiful until it crashed, literally, an hour before the show, killing the system. I was so tired that I didn’t get to enjoy the show after-hours. I seem to recall feeling like my feet were about to fall off.  I do remember tigers though.
• 2001, Momentum Chicago: Don’t recall the venue, but it was a decent conference, even if I was stuck in a booth again plugging AnnoDoc.  It took place a few weeks after 9-11, so attendance was down. I actually spoke on Streamlining Content Management Using Portal Technologies (plus Enterprise Search, but that wasn’t in the title).  It was the last day and thus a sparsely attended session. I told the attendees that I couldn’t sell them the perfect solution as it didn’t exist yet (still doesn’t), but we could start the Journey. One of the best days of my life occurred on that trip, but you’ll have to buy me a beverage to hear about it in detail.
• 2002: Momentum Orlando: Like many mid-sized Orlando conferences, this was at the Dolphin.  You begin to hate pastels after a while at that place. This was my last booth tour as it was my last conference with Infodata and AnnoDoc. This also marked my one foray into Karaoke.  I think the world is a better place for not trying that again.
• 2003, Momentum New Orleans: Best. Conference. Ever. No booth so my feet didn’t hurt.  I attended more sessions and had fun with more people, than I have at any conference before or since. Winner of the best food and partner events, hands down. This was the second conference of the revolving name-tags (Impact Innovations Group). I will never, EVER, forget this conference.
• 2004, Momentum Montreal: Disappointing. Everything was spread out and there weren’t a lot of people that I knew in attendance.  Learned a lot and enjoyed it, but this has to rank last. Following New Orleans didn’t help. This year I was with DRC, the buyers of my company. The Red Sox winning the World Series didn’t add to my enjoyment.
• 2005, Momentum Las Vegas: First year with DeepBridge, and the 4th year in a row with a different company name. (1 acquisition, 2 job changes) During the conference, the possibility of merging with EMC World was discussed.  Everyone complained and we got one more year in 2006. Interesting party at the racetrack.  Good times.
• 2006, Momentum Anaheim: The last U.S. Momentum. Great time. I met Jed that year sitting at a bar, watching American Football.  I spend the weekend before at a condo overlooking the beach and got some free babysitting. This trip was also with DeepBridge. I believe we were named Partner of the Year that year. Train performed at the final party and they put on a great live show. Their Led Zeppelin covers even impressed those that had actually seen the original in concert.
• 2007, EMC World Orlando: Much grumbling about the loss of the spirit of Momentum, which still continues in the States today (the grumbling that is). This was the motivation for starting this blog, so I guess it wasn’t a total loss. Everyone tried to say that their were separate Developer and Momentum conferences, but it wasn’t.  This was my first year at my current company, Washington Consulting, Inc.  They brought me in to run their ECM solution.  I still have the job, so I guess I’m doing okay.
• 2008, EMC World Las Vegas: This was the year that I knew everyone. I’d run into an old friend and my boss would take a snapshot. You’d think I was a social butterfly. I’m not. I just know a lot of people at this conference. I got to meet Craig this year. I had probably met him before, but this was the year that it was more than me being one of a few hundred random attendees.  I also conducted a roundtable discussion that for some reason, Alan Z has never let see the light of day, God bless him.

That’s about all the time I have to write today.  I do want you to consider attending if you use, or are getting ready to use, any EMC product.  There is fun and education to be had by all.

Promise.

Yet Another Mark Lewis Keynote

I was unable to track Mark down before his keynote. The keynote was disappointing, but in different ways than at EMC World. The first was that it spent more time talking about EMC as a whole and less on the Content Management and Archiving group. I understand why that decisions was made, and it was probably a good marketing move. My only thought is that when I sit down to listen listen to the head of the CMA group speak, I want to hear about the vision and where things are going within that CMA group. Adding context about how it fits in at EMC is great, but that context shouldn’t be the bulk of the presentation.

I didn’t hear a lot of positives on his presentation skills at the forum. Mark presented better here than at EMC World, and I have a theory as to why. Before I get into that, let’s go over the content.

The Empowerment of Information

That sums up EMC’s vision. Sounds familiar…

Enterprise Content Management is the empowerment of all content within an organization. This is accomplished through the centralized management of content, allowing for people and systems to access and manage content from within any business context using platform agnostic standards.

I’m not saying that it was taken. I’m saying that it is a good vision. I think that my proposed ECM definition would be a great vision for how the CMA group is going to contribute to the execution of EMC’s vision. Instead of that though, Mark stayed focus on the word information.

A key point that has been hammered-out by the EMC people recently, including Mark, is Search and the ability to find information. It ties nicely into the buzzword eDiscovery. Search, and its variations, are becoming a common theme that people are starting to address more and more. The concepts aren’t new, just the importance.

Mark finally got to talking about the CMA group. It was very little new information presented since EMC World. It hasn’t been two months yet, so no problems there. The one thing that struck me was Mark’s statement that EMC is not an application company. A poor choice of words perhaps as Documentum is a software application, but I understand his point. Documentum is a platform. Their goal is not to build Content Applications but to make it easy for people to create their own Content Applications through configuration and a minimum of customization.

Mark went over the four areas of the CMA. The only point that I want to really share is that Mark said that Transactional Content Management is what ECM should do best. I disagree. Compliance is what it should do best. It does transactions well because that is what it has been doing for well over a decade. The problem here is that sometimes it makes sense to use outside applications to drive the business, but retention and RM is best applied in the repository directly.

In fact, I think basic retention support should be core to the product. Make basic RPS part of the Content Server and have an additional license for RM. They’ll never do it unless someone else does it first, but RPS is something that needs to become as basic to an ECM offering as full-text search.

Is it just me, but isn’t Magellan and the Interactive Content Management interfaces (WCM and DAM) Content Applications and not platform characteristics? They are important Content Applications for them to offer, but it runs counter to the stated strategy.

Maybe if the strategy was to separate the platform and the Content Applications and just do them both well. Use an ECM SOA standard to connect them and you are done.

At the end, Mark did offer a light vision (only a few minutes). It was the combination of Web 2.0, Mining and Analytics, Virtualization, and SOA with XML holding it together and SaaS as a new, auxiliary component. Hey WAIT!!! Isn’t that the vision from pre-Mark days? Add a little SaaS and good old fashioned XML and you have the “new” vision. There was one thing new, the phrase “Platform as a Service”. Nice concept and spin, but will people like it or take a PaaS? (I apologize to everyone for that.)

Meeting Mark

After Mark’s talk, Whitney led a roundtable. It was entertaining, especially as I knew one of the panel members. I was really waiting for the break to track down Mark. When the end came, I walked up to Mark and introduced myself and even confessed to my previous post, just in case he had missed it.

I confronted him on the lack of the work Content in his talks. He told me that it is a conscious decision that he and Whitney have made. That made me feel a little better. I strongly disagree with that approach, but at least it is a decision and not a deficiency. You have to know your audience, and the audience knows ECM. ECM is part of the Information Management world, but it still has its own identity. When most of the room has Documentum, content is not a dirty word.

I did tell Mark that I like where he seems to be taking the CMA group and that I hear good things from his product people. Mark was very engaging and genuine. I sensed a little awkwardness because of the “Content” issue, but he didn’t try to get away from me quickly.

I think Mark does better the more intimate the discussion. I saw him talking to a group of about 6 people and he was very animated. I’m not sure if it is the group size or the amount of interaction that makes the difference. I think it is the discussion. He seems to thrive on intelligent discourse and it shows. When you put him in front of a crowd, his communication skills seem to degrade.

The next year is key. We need to see the promises in SaaS, Magellan, and DAM delivered. I’d also like to see a keynote at EMC World that is heavy on vision. Something that will make the audience sit up and go Cool. I would like enthusiasm to ooze off of the stage when Mark shares that vision.

I’d also like an Official Red Ryder Carbine-Action Two-Hundred-Shot Range Model Air Rifle.

Problem?  Yes, problem.  The problem is that nobody knows what to do to make everyone play together.  I’ll tell you right now, playing together is required.

Partner for Your Thoughts?

In the expo area, every other partner had a SharePoint/Documentum solution.  The rest were either storage partners or ones that won’t have a booth next year.  SharePoint was that popular.  I had a heck of a time talking to colleagues and friends working booths with SharePoint solutions.

There are a lot of solutions out there.  A majority of them focus on creating Web Parts and using them to interact with Documentum.  These are basically using SharePoint as a portal.  I was able to throw several requirements at each partner that showed gaps in their solution.

That isn’t to say that the solutions were bad.  Far from it.  Many were innovative and implemented well.  I will be keeping several of them in mind for use at clients.  They are just solutions to aspects of the bigger problem.  They aren’t a true joining of the two technologies into something that can work for the Enterprise.

They are old-school solutions using new technologies to jazz them up.

EMC and SharePoint

EMC’s status was both disappointing and reassuring.  The disappointing part is that there is no architectural improvements to the solutions being offered.  They are basically the same as those last year.  The reassuring part?  Simply put, they looked at what they had been working on as a next step and realized that it wouldn’t solve the problem either.

A lot of what I learned was that the SharePoint team has been trying to determine where they can best add value for their customers.  Trying to do it all would equate to no solutions in the short-term, and people need help NOW.  There are a couple of levels to the help needed, which I, and others, shared with Andrew and Erin.

The biggest need is to take the content out of SQL Server and store it in the Content Server.  It can’t end there though.  If content is just going to be stored there, why not throw the content on a secure file share?  Easier and cheaper.  To store it in an ECM system, you need to add business value.  The act has to “enable” the content for use throughout the organization.

The added value is the crux of the problem.  If you are going to interact with the content through Documentum or in some other interface other than the original SharePoint site, you have to take Identity Management into account.  If someone secures a document in SharePoint, shouldn’t that be reflected in the Content Server?  If a records administrator locks a document down in the Content Server, shouldn’t that be reflected in SharePoint?  These permissions, and the users and groups involved, need to be everywhere.

My favorite challenge presented by Andrew on this thought, and he had an excellent point, was what happens if a user wants to remove it from SharePoint, but they don’t have rights to delete it in the Content Server?  Do you allow the user to delete the shortcut in SharePoint or fail the operation completely?  My answer: YES! It should be configurable by installation and site, and transparent to the end user.

ECM 2.0 is key here.  The SharePoint user shouldn’t know that the content isn’t stored in SharePoint natively.  It should appear as if they were only using SharePoint.

What about throwing Lifecycles and Workflows into the mix?  I can see EMC providing some actions on a document through actions to the document library, but leave the web parts alone.  The partners are ahead on that and they are doing quite well.

EMC should probably play the leading partner solutions up on their website, delineating what they do and emphasizing the value that they provide.  It should be very visible.  The partners and clients would LOVE it.  Thus, it probably won’t happen.

Did You See That?

Yes, I threw in the need for Identity Management up there.  Just because we want users, groups, and access control lists everywhere doesn’t mean I want to go everywhere to administer them.  I want to manage it all in one place.

I hit the DFS people up on this need and told Andrew why he needed it as part of his solution.  SharePoint and Documentum should work together using an ECM SOA standard.  A key aspect of this is the ability for the systems to share Authentication and Authorization sources.  This is a problem that everyone faces.  The first ECM vendor to solve it will have a leg up on the others when it comes to being the backbone of the Enterprise.

Right now, if you lock something down in SharePoint, you have to lock it down separately in the Content Server.  The exception is content exposed in Web Parts.  The content isn’t really in SharePoint, which limits the ability to truly collaborate.

Is web parts the answer?  Kill Document Libraries and just use the web parts?  That solves the problem on some levels, but it isn’t what I would call ideal.  It turns SharePoint into a portal and shifts the focus away from being a collaborative platform. That is its strength and why it has spread so widely.

Take that away and we may as well wait for Magellan. 

First Witness, Alan Pelz-Sharpe

Alan provides this view into the keynote and the thought-process that it started in his head:

Lewis’ keynote on the Tuesday was very poorly attended compared to the others, yet was one of the more revelatory I have seen. Not in terms of the product announcements — as we’ll see shortly those were pretty much all expected and par for the course. Rather it was his lack of understanding and topical depth for CMA software that showed throughout his hour on stage.

Bringing on VP of Worldwide Marketing (and Documentum veteran) Whitney Tidmarsh to run the demo portion of the presentation made his lack of passion and depth all the more obvious. As Tidmarsh shone on stage, and talked both eloquently and knowledgeably about Transactional Content Management, Compliance and Archiving, the Knowledge Worker, and Interactive Content Management initiatives — at each switch back to Lewis, the contrast became ever more stark. I think this tells a story about where CMA is headed.

It also begged the question (asked over drinks afterward by attendees) why Lewis — a man who had spent his entire career in storage — had been given the job at all, and why a veteran, and clearly very competent Documentum exec like Tidmarsh, had not. None of us had the inside dope and speculating about vendor motives is a guessing game. Still, the only answer anyone could come up with was that Lewis represented change, and repositioning of CMA into areas more conducive to lucrative storage deals.

Alan is not the first to ask these questions. I myself had similar thoughts last year when the appointment was made. I was told at the time by people in the know that Mark was much better suited than I thought. Thus I canned my original post for a simple, small one instead. After all, you have to give a guy a chance, right?

Second Witness, Mark Lewis

I admit it. I didn’t introduce myself to Mark Lewis at EMC World when I saw him chatting up some people at the CMA Party. I kept waiting for Mark to post something to his blog, and on Monday I was “rewarded”. I mean this that he posted a simple, small, recap of part of his keynote. Nothing specific about CMA. Nothing specific about Documentum. Just a description of the model where ECM becoming transparent.

Where is the leadership? Where is the insight? He talks about Information, but where did the word Content disappear? Look at his other posts. Sparse. He posted a series on Top picks for 2008. It started December 16, 2007. It finished May 1st. 2008 was a third of the way complete!

Witness for the Defense, the CMA Group

Things I heard about Mark at this year’s conference are counter to the picture already before my eyes and ears. Let’s start with the keynote.

I met up with one of my favorite EMC people and we decided to have lunch. He brought a couple of his colleagues along. They asked me what I thought of the conference. I mentioned the disjointed delivery of the keynote and suggested that it wasn’t rehearsed enough. I also pointed out that Mark and Whitney didn’t tie everything together between their sections. I mentioned the lack of a solid overall vision.

Instead of defending the contents of the keynote, which was my primary focus, he immediately started talking about Mark. He mentioned that several people had observed that Mark didn’t look well and that he wasn’t on his game. He said that the Mark that we saw wasn’t the Mark that they are used to working with on a daily basis. I believe that the person that shared this with me was genuine and wasn’t just covering.

At another juncture, I was talking roadmaps with a product manager. It was mentioned that the roadmap that was announced at the conference only existed because Mark had them accelerate the original schedule and then made sure that they had the resources to execute. This is a very good thing. I won’t tell you what product it was, but I will tell you that it wasn’t related to Archiving.

Finally, there are the words of Mark and the direction of the products. As I’ve mentioned, SaaS and Archiving are two of the things that were Hot Topics at the conference. The other two were SharePoint and Magellan. That gives us three topics where DFS and Web Services are important, and a fourth which is beefing-up the back-end to handle more content. That is ECM 2.0. Mark talks about it in his EMC World post, but without using the word Content. His keynote at Momentum Europe (wish I could get to that one) also addressed this vision.

I think Mark needs to change his communication a bit. He needs to start using the word CONTENT more than information. He is head of the CMA, not the IMA. I also want to see some excitement and enthusiasm. The best I can say that I’ve seen is cheerfulness.

Mark seems to be getting the job done and leading the CMA Group. I think he is putting the emphasis in the right places of the product line. That isn’t enough. We in the CMA community need visible leadership and excitement. If Mark can’t do that, then we need a new face to look up to when we want to know where the Documentum product line is going.

Just think on how Alan’s article for CMS Watch would have been read if it had been Dave Dewalt delivering the same keynote.

Oh Mark, it is spelled C-O-N-T-E-N-T.

Shall we set sail…

The Rigging

Before setting sail, let’s take a very quick look at what we know of the the rigging. It is using DFS for all of its communications with the Content Server. By default, it is using a local instantiation of DFS and communicating intra-JVM. This is because DFC is still faster and more efficient. All APIs are faster than the services that they are built upon. It is configurable to use a remote DFS service though, so I wouldn’t consider it an issue.

In the front, we have Ext JS (The new rich media user interface is Flex). I’ve just taken a quick gander, but I liked what I saw. The phrase 508 compliant does tend to float across my mind though. Hopefully it is in the rigging somewhere.

All this means is that there is no WDK framework. This is okay. They have stated that the WDK is not going away, at least not in D7. Webtop is going to remain as an interface for the power user and for provided heavily customized interfaces.

I like Webtop these days, when you keep the above focus and purpose in mind. Magellan is being provided as a collaborative CEVA built to work with an ECM system through services.

I think I’ve heard that architecture before.

The Easy Course

Okay, now I am now leaving port and I couldn’t be more excited. Like everyone else I think Magellan is incredibly cool and sexy, especially for an ECM-based interface. It offers a simple interface for all of the basic library services with the addition of the Workspace paradigm and the pervasive use of discussions and comments.

The vision for Magellan is quite simple:

• A Web 2.0 Client: Check. Had that. Very pretty.
• Information Intelligence: Saw the introduction of a Folksonomy with tagging and a nice search filter. They may be getting that correct, given current technology.
• Anywhere Access: There are plans for a mobile version that will be location aware, I think in the D7 release cycle. Don’t hold them to that as that isn’t written down and my memory could be off. With it built on DFS, I don’t see why it wouldn’t be achievable in that timeframe.
• Web 2.0 Platform: Personally, I think this is just an excuse to say Web 2.0 twice. It is important, but the entire application is Web 2.0. Another reason for the split? A four part Vision seems more robust than a three part vision.

It is a good vision and EMC seems well on its way to delivering it. Once the Enterprise version of the client is delivered in Q1 of 2009, if not sooner, it is planned that Magellan will surface all of the Collaboration components that are currently in the Content Server and part of the Collaboration Edition (DCE). I can almost feel the eRoom vibe from here.

Once you get away from the coast, you enter the part of the map labeled “Here there be Dragons!” For EMC, it more likely reads, “Here there be SharePoint!”

SharePoint vs. Magellan

This isn’t a blow-by-blow comparison. After all, I don’t have Magellan to play with and it isn’t even out in Beta. Once it is in Beta, I hope to be in the program and sworn to secrecy. I’m not very good at Beta applications, so we’ll see if that happens.

Until then, I did note the following features not in the Magellan Essentials, the “free” version of Magellan, that is in SharePoint Services, the “free” version of SharePoint: wikis and lists/data tables. Now I understand the data table as that could be considered an advanced eRoom feature. The EMC data table is, in many ways, more flexible and easier to use than the SharePoint list. No problems there with leaving that out.

As for the wiki, I think that this is a potential missed opportunity. A simple, non-moderated, wiki would be a great asset in the Essentials client and would help increase the Web 2.0 feel of the interface. Provide an advanced moderated version of the wiki in the Enterprise client for differentiation. I think that if it properly packaged, it will drive adoption of both clients.

The goal shouldn’t be to stop people from switching to SharePoint. The goal should be to reverse the trend.

Paradise Found

Just past the dragons, we come to the South Pacific and see beautiful islands. This place is the stuff of dreams and why we started in the first place. Of highest interest to me is the fact that Magellan represents the first new User Interface for collaboration in many an age (DCE doesn’t count). We now see the next generation of collaboration with a roadmap that feels tangible and a destination that is in sight.

The goal is that by D7, all the features of eRoom will be in Magellan Enterprise. Not all of them are in the Content Server yet, so there is still a little more work on that end.

What makes everything feel tangible? How does this feel more real than last year’s eRoom 8, aka Phoenix, announcements?

• How about the announcement that they have engaged Crown Partners to build a migration tool to go from eRoom to the Content Server-based Magellan?
• How about their desire to take those on eRoom maintenance and just port their licenses and give them the tool?

That is money out of their pockets. A lot can change, but it sounds like they are committing to saving their comrades in the eRoom world.

Magellan is a priority for the CMA group. This is obvious from talking to multiple people on multiple levels. This isn’t just a side project or just a token attempt to stop losing market share to SharePoint. This is a strategic commitment to creating a versatile, mobile, and flexible user experience in the Enterprise. There is a sense of urgency to get it out there and let people see it, feel it, and experience it. This is their Enterprise 2.0 interface into their ECM 2.0 platform.

Will it come in time? Will the SharePoint dragon become so bloated that it can’t handle the potentially nimble Magellan ship? Time will tell.

You can either click on the link in this post or you can look up at the header and click on the page link there.

Magellan

Okay, maybe one particular order. I loved it! However, it wasn’t nearly as cool as the new interface (the video doesn’t do it justice) for the Rich Media folks. I talked to several users of the current Digital Asset Manager product line and they loved it. For years, Documentum has been focusing on the back-end of the system, making it stronger. That is a good thing if people are going to be using Documentum as infrastructure in a transparent manner. It doesn’t make the end-user feel loved though.

Something that I liked about the new interfaces, they are all using Documentum Foundation Services (DFS). That’s correct. They use Web Services which means that they are part-way to being generic front-ends that can serve any ECM system. I’m sure that isn’t the intent, but it is one step down the evolutionary path. Next step, get them using an ECM SOA standard.

This delay on the evolution of the front-end has been a gamble. It has allowed SharePoint to make strides and hurt eRoom dramatically. Many Documentum people acknowledge that it was a risk. They feel that it was worth it as the confidence in the strength of the Content Server is very high. My observation was that EMC’s best hope is for SharePoint to become so widespread that it became unmanageable, leading people to look for collaboration solutions on true a true ECM backbone.

[Edit: Full Article here.]

SharePoint

I’ve learned that no matter what your desired approach to integrating SharePoint and Documentum, there is a partner that will help you out. Wingspan has a pretty cool collection of Web Parts and Schema Logic has a plan for managing metadata between SharePoint and Documentum, though not the security data. I made a feature request for that. Nobody, and I mean nobody, has a total solution. EMC is the only group even considering the full picture.

This is something that is going to bear watching. What are we watching for? Check back in a few days. Trust me, it won’t arrive in the meantime.

[Edit: Full Article here.]

Archiving

Documentum started as a document management system. There is a lot of leftover overhead in the database from that legacy. This has limited its effectiveness as an archiving solution. Say each document takes 3K of overhead in the database. Take 10 million objects and you get 30 gigabytes. It adds up quick when you just consider the emails that the average company ingests. Take the Lightweight Sysobject, throw in some ways to optimize code for batch imports, and you have yourself a system that works extremely well as an archiving system.

The Transactional Content Management requirements are really aiming for that type of volume. If Documentum is going to be part of the infrastructure, then these new features really need to work well, and soon. While I think that EMC has the lead on the other ECM vendors from a scalability side, they need to step up here if they are going to stay ahead of the specialized archive vendors.

Software as a Service (SaaS)

I kept hearing this in session after session. I finally went to a session on it and while I didn’t buy the entire pitch and angle, I saw enough to determine that EMC will be a player in that environment within the Enterprise. DFS was just a start and seems to be part of a bigger picture of Enterprise Content Services (ECS). This is something that will take a little reflection upon to get right. It is, I think, the way to get DFS to provide what is needed to support a full SOA environment.

Other Resources

Just in case you missed it, here are some of the other bloggers that attended and little snapshots into what they saw and thought. I’ll be addressing many of these topics myself, but if you hadn’t found the posts yet, you shouldn’t have to wait.

• Alexandra Larsson: Been reading her blog for a while now and I had the pleasure to meet her while in Vegas. I hadn’t added her to my blogroll only because her Documentum posts were few in number. Well, no longer. Alexandra took notes on several sessions that I did not attend, at typing speeds that I cannot match. Her Reflections post was very good and she shared a few additional insights into the conference that are dead-on. One of her comments made me think about something someone else had shared with me. Overview sessions on Monday with more detailed sessions as the week goes on.
• Big Men on Content: Both Lee and Marko were there and have shared their initial thoughts on the conference. I got to meet both of them and I was impressed by both of them. They each want a larger community for the CMA. EMC seems like they are trying to begin building a non-developer version. They are calling it Momentum and I have subscribed to see what will happen. In the meantime, I expect more posts from the Big Men in the next few weeks.
• Andrew Chapman: I spent some time with Andrew talking about SharePoint and what we need to make things work. His most recent post bails on the issue, saying we don’t know what we want. All kidding aside, many people do only know that they want Documentum and SharePoint to work together. In one session, Andrew and Erin Samuels, the Product Manager for SharePoint, worked hard to get myself and several other users to more clearly define our use cases so that they can get what is needed to the field. I don’t envy them their task.
• The Storage Anarchist: I know what your thinking…WHOA!!! A storage blogger. This is THE storage blogger of the EMC world. He has some nice picture of a trick pool shot from the exhibit hall. More importantly, if you care about anything besides content, he has a nice summary post that links to a lot of other coverage.
• Craig Randall: Craig was there and in addition to working with me to understand in more detail the issues we are facing around Identity Management in our SOA world, he is a pretty great guy. He directed me to some critical people with whom to share my use cases and as the show went I continued to chat with them about I am seeing on a growing basis. Craig has several good posts about some of the DFS sessions, with links to resources, on his site.

Oh, did I mention the YouTube collection of videos? You can either follow that link or go to EMC’s EMC World Online Newsroom

Sense of Perspective

I’ve been to every US Documentum user conference since 2000 and I’ve seen a lot of changes over time. I remember the exhibition being setup for the whole conference with almost no traffic (better now). I remember motorcycles zooming in to the stage at the keynote (fun, energetic, suffocating). I remember walking down Bourbon Street and having a great time as if it was just an extension of the conference (a great sense of community).

Some things have improved. One thing that hasn’t is that sense of fun and community. It was easier to find Documentum folks on Bourbon Street in 2003 than it is at time at EMC World. In 2006, there were over 2000 attendees at the last Momentum in Anaheim (thrown together on the quick) and more than 2500 the year before in Vegas. This year, the number given was 1500+ out of 9300+ attendees. It makes it hard to find people that have similar interest unless you already know them. You can find those people at sessions, but the sessions aren’t that well grouped together. The decrease in CMA attendance speaks volumes.

One reason that I think that less people are attending is that it is becoming a more technical conference. Documentum isn’t just about the technology. Ask any ECM expert and they’ll tell you that the business efforts (change management, business process re-engineering, etc.) are more important than the technology. All of the storage people, and developer conference attendees, really swing the demographics further to the geeky, nerd range and that doesn’t make the end-users comfortable. They want to talk to other users that experience their pain.

Billy Crystal

Billy was funny as hell. Loved the show. The final event sucked though. The final events are usually a touch-and-go quality to them. That is fine. What I enjoy best is getting together with everyone I met, or only see once a year, and having fun and chatting about things besides ECM. This year, it was a no-go.

For those that missed it, we were funneled into an arena and quickly served food and drinks. We were then ushered into our seats, for a long time. There was no real opportunity to wander around and talk. Then Billy came out and talked for over an hour (I love Billy C!). When he finished, it was over.

That’s it. Everyone going their own way, no place to really network and get together. It was horrible from that perspective. The previous night, there was a CMA party hosted by Crown Partners featuring The Groove. That was great! We could wander around and talk to people and enjoy the sense of community. The only problem with that event was the invitation requirement. It scared off some people that didn’t go grab an invitation, even though they were very liberal with them.

Things I’d Like to See Next Year

One, very smart, EMC employee told me that it would take three years for them to get it “right”. Well, we are two years into it and I actually saw a step back from last year in some areas. Here are some suggestions for next year.

• Group all the CMA presentations into one area of the conference: This will let me know if people walking by are CMA and I have a chance of running into people that I met two sessions ago. This includes the Software Developer tracks as well. Those would be a good transition area in the conference between the different technology tracks.
• Different colored badges for each community area and/or handing out community ribbons at registration: You could get ribbons this year IF you went to the community talks on Sunday and picked one up. Many didn’t realize they were there, and some weren’t even at the conference in time to participate.
• More room for the community meet-and-greets: The community meet and greets were a great idea. There needs to be a larger area for each one for people to sit and chat. There were no chairs so people just talked for a couple of minutes and moved on. We need to encourage people to talk.
• Group the sessions in the catalog by topic area: This year they were listed as one long alphabetical list which made it hard to scan to find good topics. Another benefit is that it also will help make sure that people go to the sessions in areas that they care about without having to decipher non-specific titles or flip through the book to find the detailed description.
• Less overlap of events: Take a couple of invite-only events and put them at the beginning or the end of the conference. I had to duck-out of some things in order to attend a Product Advisory Forum, which I needed to attend. It ate deeply into my time to participate in standard activities.
• A zero-conflict Meet the Experts time period: They had several meet the expert time periods where you could go and talk to product managers on their product specialities. Not only was it poorly advertised, it took place during sessions. Do I talk to the expert among a group of similar users or do I attend a critical session that isn’t repeated? If the exhibits are open, fine, but no sessions.
• Bring back the Meet the Speakers sessions: Along this same line, I miss the Meet the Speakers sessions as well. It was a great chance to network and talk. That should be at the end of each day after the sessions have ended.
• The booths should be open until 7:30 or 8 and not close when the sessions are ending: If I spend the whole afternoon in sessions, it makes it hard to get to the exhibits to visit all of the booths.
• Bring back the CMA content: Speaking of sessions, there were five CMA tracks this year, plus the Developer sessions. In previous years, there were 6+ well defined tracks. Why do we have less sessions when we are now talking about more things like Captiva and X-Hive? Where is the generic Introduction track? That was a great track for Project Managers and users that were new to Documentum.
• Schedule the broader, overview sessions for Day one: The Documentum 6.5 platform overview session was offered on the last day. Attending that used to alert me to cool things to check out later.
• Bring back the call for topics: It was removed this year which is how they got down to five tracks, less customers presented. Case studies are great. Watching end-users who are passionate about the technology that they just implemented is great. You can learn a lot. This year, only a few from sponsors which doesn’t really engender a sense of community.

I have some session specific thoughts as well, but I think I’ll save that for a more content focused review next week. This used to be a user-centric conference. Now, it is a technology-centric conference. It still has value, but it has shrunk.

• Starting off with a sales-pitch for scanning. Cost, productivity, compliance, and data (extraction)
• Enterprise Document Capture
  • Capture: scan/fax/email, moving to distributed scanning in many organizations which leads to shorter turn-around time
  • Identify: need to identify the document formats to reduce the level of human intervention
  • Extract: pull data from the documents and use it for processing, not just for search
  • Validate: Check against business rules and external data
  • Deliver: drop into repository for distribution in the right locations based upon data, send data to other business systems
• Distributed scanning reduces shipping and gets information in the system and actionable much more quickly
  • Faster processing, limit delays transporting paper to capture locations
  • Reduce risk of lost or misfiled documents
  • Enable knowledge experts to contribute directly into the system
• Multi-Function Devices (MFD) can be used to scan documents directly into a central repository
• Automatically Classifying documents
  • Able to identify page layouts that identify the starting page of a multiple page document without the need for bar-coding or a cover sheet
  • Rules can be more complex
  • Needs to be reviewed as the technology improves to improve efficiency
• Full back-file conversion may not be feasible
  • Scan-on-demand will make needed documents available
  • As old records are eliminated and new records continue to be scanned, the physical storage needs will decrease organically

That’s it. Expect more posts over the next few weeks with analysis and observations. In the meantime, I think I need to find Starbucks. Luckily the Luxor has 3 of them.

Disclaimer

All information in this post was gathered from the presenters and presentation. It does not reflect my opinion unless clearly indicated (Italics in parenthesis). Any errors are most likely from my misunderstanding a statement or imperfectly recording the information. Updates to correct information are reflected in red, but will not be otherwise indicated.

All statements about the future of EMC products and strategy are subject to change due to a large variety of factors.

• Key characteristics
  • Habitable software
  • Community governance
  • “Leveraging humanity”, David Freund
• EMC’s Innovation Network drives research collaboration with social media
  • Global Research Collaboration
  • EMC ONE, Online Network of EMCers
• Daoli Trusted Infrastructure Wiki, http://www.daoliproject.org/
  • Hosts research on trusted computing and virtualization for EMC Research China and several university partners
  • Open to the Public for Comment
  • Daoli: A research collaboration environment where everyone can make active contributions
  • The purpose is for people to learn, not to build product which would be private
• Web 2.0 Total Customer Experience (TCE)
  • Gives an overall view of a data center
  • Mashes together multiple support sources plus customer knowledge
• Experts Finder helps find people one given topic
  • Allows a rating to measure expertise
  • Mashes location together as well
• Five lessons learned
  • Social computing follows social connections, start with the existing ones
  • Problem solving drives participation
  • Corporate efforts need a common “space”
  • Information sharing must be intentional and a habit
  • Transformation takes time

Debating on attending another Social Computing presentation or getting some work done. I guess we will all know soon when we read the next post. One thing is certain, coffee.

Disclaimer

All information in this post was gathered from the presenters and presentation. It does not reflect my opinion unless clearly indicated (Italics in parenthesis). Any errors are most likely from my misunderstanding a statement or imperfectly recording the information. Updates to correct information are reflected in red, but will not be otherwise indicated.

All statements about the future of EMC products and strategy are subject to change due to a large variety of factors.

• DFS is a component of Enterprise Content Services, ECS (Great, a new term)
  • DFS is core package of ECS (SaaS services separate from DFS but part of ECS is my guess)
• DFS will work against 5.3, with some caveats with the workflow services
• Primary Interop Assembler is gone for .NET, so DFS is the way for .NET applications to talk to Documentum
• How do we solve the problem
  • Create Java Service, Plain old Java object (POJO) or Service-Based Object (SBO)
  • Annotate service, DfsPojoService or DfsBofService
  • User DFS toolsto generate artifacts
    • WSDL, proxies (SOAP, local), etc.
    • Optional Java and .NET productivity layers
  • (Way too fast on the slides. Luckily they are online)
  • Use generated Java client classes to build tests
    • In-process API first to simplify development
  • Use DFS tools to package the WAR/EAR application
    • Validate tests running remotely (same tests!)
  • Validate WSDL with non-Java clients
• DFS enables SOA but it cannot provide SOA (I would criticize this point, but there may be users here that don’t know it, they are doing well with this point and I’m a little glad it is in here)
  • Users must separately invest in governance
• DFS extends the Java standard JAX-WS tools
  • Thus Java 5+ environment (It is a D6+ application, so this is redundant)
• Requires JAX-WS 2.1.2 reference implementation from Sun
  • JAXB 2.1.3 reference implementation from Sun
  • JBoss 4.2 support is coming in D6.5 support in 6.5
  • Tomcat support coming in 6.5
• Performance appears to scale well, see the presentation for a chart and numbers.
• UCF versus MTOM file transfers
  • UCF was slighter faster in 50MB to 100MB range, similar below that
  • MTOM had issues around 100MB
• Best Practices (all in presentation)
  • When setting repeating string attributes on an object, use semi-colon delimited string
  • QueryMaxResultCount defaults to 100, not -1 as documented
  • When results are expected to be over 500, you must cache and cycle through the reults
  • When using UCF, you must set both the ServiceContext and the ObjectOperations Content Transfer modes to UCF, not just one or the other as the documenation states. We also found that when ContentTransferMode was set to UCF, custom services would return a ucf error (even if they were not doing content transfer) so I explicitly set the ServiceContext and ObjectOperations to base64 when not doing content transfer and to UCF when doing content transfer.
  • specify the attribute value “requires Authentication=’false’” to conduct services that don’t require authentication (docbase list…)
  • Ensure you explicitly specity your propertyProfile for querying objects
  • Use the Schema Service to validate attribute information before calling updates and creation calls
  • Use the DataPackage for services that return result sets
  • When validating a returned DataPackage, check to see if it is NULL, and if not, see if .DataObjects is NULL
  • For DataObject, r_object_id is always Identity, not as important in D6.5
  • JAVA_HOME environment variable needed for UCF data transfer on the client even though WDK applications don’t need it
  • Don’t run DFS on a D5.3 Content Server machine, security hole, not an issue in pure D6 environment, just put DFS on its own server in 5.3
  • Session affinity still required for clustered environments, DFS sessions are still tied to the host
• The examples/demos are available via Craig’s post. (They look pretty)
• People are building their own SharePoint integrations with DFS

Trying to determine if I want to go hear about the Social Computing in EMC or take some time to work. This is a 8:30-2:00 session heavy day with breaks in-between. No lunch break, though I suspect, and hope, that snacks will be out during the breaks.

Disclaimer

All information in this post was gathered from the presenters and presentation. It does not reflect my opinion unless clearly indicated (Italics in parenthesis). Any errors are most likely from my misunderstanding a statement or imperfectly recording the information. Updates to correct information are reflected in red, but will not be otherwise indicated.

All statements about the future of EMC products and strategy are subject to change due to a large variety of factors.

• The presentation seems to be acting on its own. Possessed even.
• Web 2.0 “aims to enhance creativity, information sharing, and most notably, collaboration among users.” (Too much quoting of wikipedia)
  • Rich user experience
  • User participation: one of the most important aspects (Agreed. I like the collective intelligence aspect of it)
  • Dynamic content
  • Metadata
  • Openness
  • Freedom
  • Collective intelligence by way of user participation
• More focus on Customer
  • Focus on introducing new brand experiences
  • Proliferation of creative content
• Internal drivers for Web 2.0 concepts in the Enterprise (Enterprise 2.0)
  • Support virtual teams
  • connect thought leaders
  • RSS feeds
• External drivers
  • Link partners and internal teams
  • Facilitate unstructured business processes
  • Promote self-service models
  • Allow partners/customers to help each other
  • Solicit feedback at all levels
  • Generate customer interest and loyalty
• Enterprise 2.0 (was beginning to wonder)
  • Will we get constructive feedback, or just rants?
    • Liability incurred from external posters
    • Compliance issues of the information posted (is it a record?)
  • Moderated wikis mitigate risks and control the message
  • Things should be driven by company objectives
  • The appearance of openness is critical, even if not 100% open (I like this presentation. He is sharing the reality of meeting company objectives by controlling the external interactions)
• Content Distribution Services is replacing Site Caching Services and will be bi-directional when desired. (This is important in order to capture the end-user content from the website and place it back in the repository for tracking, auditing, and compliance)
  • This information once collected can also be sent back to be thrown into a workflow and classified
• Dynamic Delivery Services delivers XML to a Web Server Runtime that will render the XML as HTML based upon the user interactions
• (Brian is refreshingly, brutally honest. He is now one of my favorite people. I wish there was time in this convention to talk to him more)
• Recommendations can help identify and reinforce an organization’s objectives
  • “Most popular downloads” could be filtered on the user profile
• Working on taking content collected on one web server and replicating it to other web servers
  • Discussion thread in US replicated to Asian server
  • Can have some review and approval before the stuff is replicated
• Use Content Transformation Services to “normalize” the format and metadata
• Tagging is taking prominence
  • In Web 2.0, the community decides what tags are best
  • in Enterprise 2.0, the organization needs to be able to prune some tags

Off to DFS Real World examples.

Disclaimer

All information in this post was gathered from the presenters and presentation. It does not reflect my opinion unless clearly indicated (Italics in parenthesis). Any errors are most likely from my misunderstanding a statement or imperfectly recording the information. Updates to correct information are reflected in red, but will not be otherwise indicated.

All statements about the future of EMC products and strategy are subject to change due to a large variety of factors.

• The vision
  • Web 2.0 Client
  • Information Intelligence
  • Anywhere Access
  • Web 2.0 Platform

• Web 2.0 Tools for collaboration
  • Author & Publish: Wikis, blogs, Collaboration, IRM, customized lists
  • Organize & Manage: Guided Navigation, Tagging, Classification, Personalized views
  • Network & Access: Enterprise Ready, Securde off-network, Access anywhere, retention
• Need to find the information and find related related information
• Magellan is Compliant, secure, intelligent Web 2.0
• Individual, Team, and Organizational Workspaces
• Trying to improve Ease of Use, hiding complexity of policies and content intelligence
• Ease of installation is a focus
• Future in 2009
  • Easy configuration
    • Partition deployments and configure features that are needed
    • Apply a common set of policies to all spaces in a site
  • Ease of Deployment (D7)
    • Package and license individual units of functionality
    • High Availability
  • Ease of Development
    • Tooling and RAD
    • Leverage the client side tech that best suits the business (Ajax, Flex)
• Magellan Essentials client: no cost client that works with Documentum, late August free beta
  • Team workspaces
  • Library services
  • Access control
  • Content templates
  • Guided navigation
  • Lifecycles
  • Desktop Integration
  • Event triggered collaboration through SOA support
  • Reecent Changes
• Advanced client has more (This is the one that would be worthwhile)
  • Wikis, Blogs, and RSS
  • Template and component base user interfaces
  • Personal spaces
  • Tagging
  • Data tables for managed collections of information
  • Federated search
  • Visualization
• Show the canned demo from the keynote (Essentials), slower pace
  • See recent documents and recent comments on start page
  • Comments and discussions are supported around documents
  • Includes thumbnails, assuming a thumbnail has been created
  • Everything can be directly accessed with an URL
  • Drag and drop in the UI
• Composed Pages in full client
  • Bring wikis, blogs, RSS Feeds, and other information into one page
  • Unify information into one Space
• Custom components in D7
• Data Tables from eRoom and Collaboration Edition will be surfaced
  • Will add validations down the road
• Q1 2009, Essentials and Magellan full will be released
• Magellan D7 in latter half of 2009
• Webtop D7 is in the works and future versions are planned
  • Showcases the full power of Documentum
  • Provides more specialized and flexible customizations
  • Client of choice for power users requiring advanced content management functionality (Virtual Documents)
• eRoom will be supported for many years to come
  • Maintenance releases will continue
  • All eRoom features in Magellan in D7 timeframe
  • Working with Crown Partners for free migration utility leveraging Bulldoser. Will be available for customers to migrate from eRoom to Magellan.

Time for dinner and the party. [Edit: Follow-up Article here.]

Disclaimer

All information in this post was gathered from the presenters and presentation. It does not reflect my opinion unless clearly indicated (Italics in parenthesis). Any errors are most likely from my misunderstanding a statement or imperfectly recording the information. Updates to correct information are reflected in red, but will not be otherwise indicated.

All statements about the future of EMC products and strategy are subject to change due to a large variety of factors.

• More on Lightweight sysobjects, but much better graphics to explain the concept.
• Default sysobject covers 6 tables before hitting any custom type tables.
• Lightweight sysobjects lose 3 of the tables, including both dm_sysobject tables, and any custom parent object types
• Seeing 25-50% size improvement on the database. It depends on the application design. (Use these numbers over others I may have elsewhere)
• Shared features characteristics include
  • Security
  • Retention
  • Folder Linking
  • Lifecycle
• Lightweight sysobjects cannot be versioned or replicated
• Will need to change folder structures in model as they all share (Keep low in the hierarchy to prevent accidental browsing)
• Can re-parent a lightweight object to change to a parent with different shared characteristics.
• Providing a migration script call MIGRATE_TO_LITE
  • Options to perform bulk of migration offline (via the oracle redefinition package) to minimize the application impact
  • Can selectively pull from object attributes to light object
  • Object id’s are reused
  • Initial tests show 1 million to 10 minutes
  • Will be more painful in SQL Server as it has less of the cool packages for redefining data models
  • Provide utility to remove orphaned parents once they have been move to Lightweight (slower than the migration)
• Adding range partitioning to database tables in D6.5
  • Can group objects by date as older tables are typically accessed less often
  • Partition Id will control what partition the data sits
  • When mixed with tablespace management, is basically Content Storage Services for the metadata
  • Parent objects and child objects, from lightweight model, don’t have to be in the same partition
• Adding high-speed insert (Use for migration or super high-volume processes…)
  • Ingest into off-line partition with no index
  • Once complete, index is applied and partition is placed online via a partition swap
  • Requires a content server bounce
  • Doesn’t work with business objects applied
  • went from 324K/hour to 20mil/hour
  • This is a database feature revealed through Documentum
  • Not good for 24 hour processes or quick availability
  • Meant for massive amounts in short time periods
  • Not into Centera
  • Should only use Lightweight objects
  • FileNet migrations as an example
• D7 things coming forward
  • Content transfer improvements: parallel uploads of large files
  • Scaling the XML Store up higher
  • Lucene indexing (More offline questions for that)

Off to learn more about Magellen. Everything that I hear is very good so far.

Disclaimer

All information in this post was gathered from the presenters and presentation. It does not reflect my opinion unless clearly indicated (Italics in parenthesis). Any errors are most likely from my misunderstanding a statement or imperfectly recording the information. Updates to correct information are reflected in red, but will not be otherwise indicated.

All statements about the future of EMC products and strategy are subject to change due to a large variety of factors.