Message from Mark Logic

Been a while since I posted anything. Life has been real busy, and continues to be busy. Aside from getting everything in order to head to EMC World for a week, I am working on the whole ECM/SOA world of ECM 2.0. Not really changing my view, but refining it so that I can get the point across more clearly and concisely.

Meanwhile, I found a follow-up posting by Dave Kellogg of Mark Logic commenting on my previous Mark Logic post and on the long name of EMC’s latest product, EMC Documentum XML Store OEM Edition. Now I don’t really want to enter into any debates, because

1. I don’t have enough hands-on experience with either X-Hive or Mark Logic.
2. As Dave points out, it turns into a “he said/she said” kind of argument.
3. I don’t really care who wins.

I really just want people who read my blog to be aware of Dave’s latest entry. It is worth a read and is about as balanced and fair as what I hear from X-Hive people. He has three comments which I’ll very quickly address.

My Non-Rebuttal

Dave says that they only compete with 2% (very rough estimate) of EMC’s product base, so they aren’t competitors. That is playing with numbers. Try this spin…EMC offers a counter product to everyone of Mark Logic’s products. Thus, Mark Logic competes with EMC in a big way. It doesn’t mean that there aren’t times where Mark Logic can work with parts of EMC, but as a rule, they are a competitor.

Enemy? That depends on how personal you take it. Remember, EMC never told me that Mark Logic was the enemy. That was my word. EMC did throw Mark Logic’s name around whenever competitors were mentioned.

Next comes a dig at X-Hive:

Indeed, I do view MarkLogic Server and x-Hive/DB — i.e., EMC Documentum XML Store (11 syllables) — as same-category competitors. But I believe they’re both XML content servers only in the same sense that Oracle and Microsoft Access are both relational databases.

WOW! That is a slam. I didn’t think they were enemies. I think that proves my point. Whatever Dave may have meant by that comment, that pretty much closes the book on the issue.

Benchmarking

Dave does say that trading benchmark data from private evaluations is a poor idea. He offers some reasons, which I agree with in general, though maybe not in some specifics. I will say this though…EVALUATE BOTH YOURSELF!!!

There are lots of different things that you can do with XML, so play it out and see what works best. Both will give you evaluation copies. Use them and go from there. Set a minimum performance level and see if they meet it. Load them up with sample data and make sure that they still work.

Don’t forget to read their sizing documentation. That always helps give you a good idea of the relative performance levels. If they don’t have any sizing documentation, that is a bad sign. If you can’t find it, ask for it.

And when your done, pick something else entirely just to drive everyone nuts.

See Pie at EMC World 2008

As if there was any doubt, I will be attending EMC World 2008 this year in Las Vegas (May 18-22). It is the best place to corner the product managers to see what is coming down the pike. The sessions are mostly useful, but the highlight is getting to talk to other users and the EMC staff. I strongly recommend attending if you have a Documentum installation.

Watch for the Word

I plan on blogging during/after every session that I attend. While I can’t share many of my offline conversations, some of what I hear is not for public dissemination, I will share everything from the presentations.

Tips: Deleting a Lot of Rows from a Database

Recently, I had to remove a large number of rows, almost 10 million, from a very active database table in a LIVE system. Forget why for now as that is the subject of another post. The basic problem was how to remove 99.8% of the rows without impacting users or removing the few rows that we actually wanted to keep. To make matters worse, the field that was determined to have all the answers didn’t have an index.

Adding an index to the field in question would only solve part of the problem. It would take resources just to apply the key to a table that large, especially one that already had several indices. There are also locking issues and let us not forget the Transaction Log usage. We are talking Gigabytes of space for any new index and for temporary Transaction Log space.

Mark Logic Wants to Play with X-Hive?

Dave Kellogg, the CEO at Mark Logic posted a response to my X-Hive and the Content Server post. His basic theme, is that Mark Logic is not the enemy of EMC. Maybe, maybe not. Personally, I don’t care as long as my clients are happy. I do want to comment briefly on it so as to clarify things as I understand them. Any EMC person that wants to add clarification, please do so for everyone’s benefit.

We have many common customers. They want the products to work together.

Is this Mark Logic with X-Hive or Documentum Content Server?

MarkLogic Server complements document management — we deliberately decided not to build a CMS precisely to avoid competing with ECM vendors. (Ironically, x-Hive built a competing CMS called Docato on top of x-Hive/DB.)

I’m not sure how this changes things. They complement the Documentum Content Server. So does X-Hive. The won’t compete with Content Server. It is the competition with X-Hive, the Documentum XML Store, that the competition centers upon.

Mark Logic is about doing one thing better than anyone in the world: high-performance XQuery on top of large XML document stores. I don’t believe that’s the mission statement for x-Hive/DB (now “EMC Documentum XML Store”) which I’d guess is more of “how can we get Oracle out from underneath all our implementations?”

MarkLogic is more than just a basic “store.” First, MarkLogic is a high-end product — it goes very fast and scales to contentbases in the hundreds of terabytes. Second, MarkLogic provides a new top-to-bottom XML way of building web applications.

When attending the X-Hive presentations, they claim to do what Mark Logic does, except better. This is usually where the impression of Mark Logic as the main competition, aka “the enemy”, crops up. I do want to point out that this comes from X-Hive acquired people, not from any of the acquirers.

I think the point that I have taken away is that X-Hive and Mark Logic compete, but Mark Logic can work with the Content Server. Personally, I’d be happy if I was Dave. They compete with X-Hive, but the EMC sales people may try and sell customers more that just the X-Hive product, making things easier for Mark’s people. Also, they have a fair shot of staying ahead of the performance curve as X-Hive is spending resources becoming part of the Documentum platform.

Enterprise Content Management 2.0, Still in Beta

The Big Men on Content, Lee to be precise, recently joined the ECM 2.0 discussion, stating that they are going to wait for EMC’s sp2 before they jump on-board. That could be a long wait. After all, we are still in Beta as far as I am concerned.

This was prompted by a reading OpenText’s Enterprise 2.0 Content Management strategy. Note the placement of the 2.0. We’ll be getting back to that.

Web-Centric Content Management

For the past few months, I have been popping in on the Infovark blog. For those that aren’t aware, they are creating an application for the Enterprise 2.0 world. From reading their various posts you can get a fair idea of what they are creating, at least conceptually. I’m not going to go into that, as I may be off and you should check the site out if you are that interested.

What made me decide to comment was their conceptual approach to managing content. It is refreshing and clean. I’m not sure how it will work in the wild in many corporate environments. There are a lot of details that I don’t know yet, so I’ll be optimistic.

ECM Design Patterns

Recently, the EMC Developer Network has started posting some “Design Patterns”. I use the term loosely to mirror their terminology. Each “pattern” is really just a quick description of the problem and two approaches to solve the problem. It is all very high level.

Before I get any further, kudos to them for actually taking the time to begin developing these “patterns”, starting last fall. There is a definite need, and their choices for the first two are ones that are encountered quite frequently, at least by myself. All I am doing here is offering some feedback, most of which I have already shared.