QlikView: A Return to Simple

I've been hearing more and more from my old friends in the business intelligence (BI) world about a product called QlikView, from a Swedish company, QlikTech, that was founded as a consultancy in 1993 and launched its product in 1997.

QlikTech's war cry is one word: simplicity. And I like it.

Most mainstream enterprise software packages have been around roughly 15 to 25 years. Over those years two things invariably happen:

• The usual cycle of feature creep which, while well intentioned, results in hard-to-use bloatware.
  

• Massive, multi-level market consolidation

When you take these two factors in combination, it's scary.

Business Objects, for example, was founded in 1990, starting as a PC tools company. Then the web came along and we created a fairly separate web version of the product (WebIntelligence). Then enterprise reporting became important and, lacking the ability to build a nice enterprise reporting product, we bought our way out of the problem with the $1.2B acquisition of Crystal Decisions. Then software consolidation started big time and the company (I left in 8/04) bought I'd guess a dozen or so additional companies/products such as Acta, SRC, eXcelsius, and Cartesys.

Thus, when SAP bought Business Objects last year, they didn't buy one company / product; they bought literally dozens, each presumably with its own cycle of feature creep and none of them terribly well integrated. And if you think that's scary, then think about the state of affairs at Oracle, who's been much, much more acquisitive.

As I've ranted in this blog, I think the industry is starting to resemble the conglomerates of the 1960s. That leaves room for innovators of several types:

• Business model disruptors. People like Salesforce.com and the SaaS crowd. Or people like MySQL, Alfresco, and the open source types.
  

• Focus disruptors. People who want to do one thing very well. QlikTech is here, and I'd say that Mark Logic is as well.

The focus disruptors offer a different value proposition than the modern enterprise software "value proposition" which, sadly, has devolved to something like:

We have a suite of stuff that we've accumulated over the years and across the board it does a pretty good job of covering everything you want, most of it not too well, and in many cases we actually have 2-3 different things covering the same space and our consultants can figure out which you need, and yes there's a lot of redundant infrastructure that we're slowly eliminating and no, no one will be left behind, so we're going to maintain all those code lines for years, and because of that innovation will be pretty slow, but that's OK with us because we bought all our competitors, and you shouldn't worry because everyone's using our stuff so you will neither get, nor generally lose, competitive advantage by working with us.

I sometimes wonder on the personal front: "when did we become our parents?" At work, it's: "when did our industry become the mainframe business of the 1980s?"

I'm a big believer that much as a backlash erupted from the 1980s-era mainframe business so will a backlash (continue to) erupt from the current enterprise software business. Open source is part of it. So is SaaS. And I think focused best-of-breed product vendors will increasingly be part of it as well.

It's working for QlikTech. In 2007, they say they grew 80% to $80M in revenues and have 7,306 customers in 82 countries. And it's working for Mark Logic as well.

For more on QlikTech's thoughts on simplicity, check out the blog on SandHill.com that prompted me to finally write about them -- Simplicity: What's Next in Business Software.

Honey I Shrunk The Company, II: Convera's 4Q08 Results

Convera reported its 4Q08 and 2008 fiscal results (for the period ending 1/31/08). Highlights:

• 4Q08 revenues of $0.28M
• 4Q08 net loss of $7.0M
  

• 2008 revenues of $1.1M
• 2008 operating loss of $27.0M
• 2008 net loss of $9.0M (net of a $17.9M gain on sales of discontinued operations)

Cash and equivalents of $36.6M on 1/31/08, down from $47.5M on 1/31/07, amounting to an $11M burn over the year, including $18.1M of cash they received from the sale of RetrievalWare -- or, if I did the math correctly since they didn't present a statement of cashflows -- a pre-sale $29M burn rate on the year, roughing out to a cash burn rate of about $7.5M/quarter. That translates to 4.8 "quarters of cash" -- i.e., how long your cash lasts on your current burn rate.

So in the next few quarters, the new strategy better start generating more revenue, or they'll need to cut expenses or raise (yet) more cash.

The good news in here (and you have to look hard to find it) is growth. 4Q08 revenues were up 139% compared to $117K in 4Q07, and 2008 revenues were up 316% compared to $269K in 2007. It's nice growth, but it's off a minuscule base.

The bad news on growth is the 139% instantaneous growth rate (i.e., 4Q to 4Q) is lower than the 316% average one (i.e., 2008 over 2007) suggesting they're experiencing deceleration. Were they gaining momentum, the instantaneous growth rate would be higher than the average one.

See my original Honey, I Shrunk The Company post here.

New MarkLogic.Com Site

Please check out our new corporate website at www.marklogic.com. Thanks to the entire marketing team, those many who supported them, and particularly to Danielle "Can I Buy a Vowel" Guinebertiere for her driving and oversight of the project.

Send feedback to me at the ceo-at-marklogic-point-com address or give us feedback through the various feedback channels listed on the website, here.

Thanks!

Mark Logic: The Forgiving XML People?

I was at a conference the other day and was talking to a senior IT executive who said something along the lines of: "... but, wait a minute, XML is very inflexible and unforgiving ... it requires that all content be, for example, in a single schema ..."

This provoked a few insights:

• It's not XML that unforgiving and inflexible. It's the conventional tools (that weren't designed to manage XML content from the start) that you are trying to use to manage it that are giving you this impression. Au contraire, XML lets you basically mark-up anything in any way that you desire.

• A stark reminder that in the minds of people responsible for building and delivering systems that differentiating between "XML is unforgiving" and "conventional tools are unforgiving when dealing with XML" is difficult and may appear pedantic.
  

• That perhaps we should position ourselves as the forgiving XML people. Fact is, MarkLogic ingests XML as it is and then either (1) lets you run queries as powerful as the mark-up you've given us, or (2) lets you use XQuery to transform / clean-up your XML over time. (See this post on lazy XML enrichment.)
  

I Admit It. I Tried FriendFeed.

One of the reasons I spend so much time keeping abreast of Web 2.0 applications is that technology builds on itself. How can you understand Twitter as a micro-blogging utility initially aimed at SMS if you don't know what Blogger is? How can you understand FunnyOrDie if you don't understand YouTube? How can you understand why Facebook is cool, LinkedIn is professional, and MySpace is lame if you haven't tried them?

I kept hearing about FriendFeed, a kind of meta-web-2.0 application that ties together all your other web 2.0 applications. I was reluctant to try it because I had my standard, skeptical, "does the world really need this" reaction. But, based on using many web 2.0 apps, I think I saw the need. So, I tried it.

What does it do? Simply put, it integrates all your web 2.0 "exhaust" into a single feed. For example, it integrates my blogger posts, my Facebook feed, my Yelp reviews, my LinkedIn changes, and all my Twitter tweets into a single feed, available, for the time being, as a public feed, here, or via RSS/Atom here.

Who, other than my mother and possibly my wife would want this? It's unclear.

But for the new generation who are using these apps and about 20 others (e.g., Flickr, Delicious, Digg, Google Reader, iLike, Last.fm, Reddit, Magnolia, Pownce), it becomes, dare I say, essential.

Why? Because there are simply too many different services, too un-integrated. That is why the Facebook "as a platform" announcement was so interesting. Could Facebook effectively integrate Flickr, Twitter, YouTube, Blogger, Gmail, and a bunch of other apps/services into a single platform?

In that world, you don't need FriendFeed because everything's a Facebook app. But in the real world, here, today, FriendFeed's pretty useful -- of course, only if you're using the underlying applications; try selling a tire pump to someone who doesn't have a bicycle.

Soccer Fans: Please Support The Thunder

This is not a regular, business post, but since I know many old friends and colleagues read this blog, and I know that some of you are soccer fans, and I know that some of you would be eager to help support a youth soccer team, I thought I'd do a brief donation solicitation on behalf of my son's soccer team, the San Carlos Thunder. (Hint: you'll never find him in the picture unless you know what my wife looks like.)

The team is fund-raising to support a dream trip to England this summer. On the trip they will get to tour classic English football (i.e., soccer) stadiums like Stamford Bridge and Old Trafford, the homes of Chelsea and Manchester United, respectively. In addition, they will play local and or academy teams. In ~9 days, they'll play ~5 games, tour ~4 stadiums, and have a day or so in London.

As it stands right now, not everyone who wants to attend will be able to do so unless we raise more money. The boys have been working hard doing the standard stuff (e.g., running the Snack Shack at the local field, recycling cans, selling chocolates) to raise money. Thus far, in my estimation, they've been pretty successful; I think they've raised north of $6,000. That's a lot of hot dogs and cans. But all in, the costs run more than $2,500/person. So what they're planning on raising helps defray maybe 20% of the cost, and that's good.

But, simply put, not everyone who wants to go will be able to do so unless they raise more. If you're interested in helping out, every $25, $50, $100, or more will help a lot. (And yes, I've donated myself.)

So if you want to help out, please print this post and drop a check in the mail before 4/30/08 to:

San Carlos Thunder
Attn: MLB
860 Cedar Street
San Carlos, CA 94070

__$25 __$50 __$100 __$250 __$500 ___ other

If you prefer PayPal, you can send funds to me at dave-point-kellogg -at-marklogic-point-com (writing it out to avoid spam email harvesters) and I will pass them onto the team.

Thanks again. It will make a difference.

Bill Trippe on XML Blogs to Read

I was honored to see that the illustrious Bill Trippe of Gilbane Group cited the Mark Logic CEO Blog as one the blogs he regularly reads about XML technology.

Other blogs on Bill's list include:

• Trendwatch by CMSWatch
• Conquer Information Chaos by AIIM president John Mancini
• The DITA Blog at XML.org
• TeleRead on e-Books
• Brian Jones of Microsoft on Open XML

Anyone interested in the blogs I read can go see my public blogroll on bloglines, here. Need I say that Bill's blog on XML is amongst them.

TechCrunch Highlights Sequoia's Success Elements

Check out this post by Michael Arrington on TechCrunch, entitled Sequoia's Gospel of Startups More True than Ever. You can see what Sequoia calls the Elements of Sustainable Companies on Sequoia's own website, here. A TechCrunch intern wrote an article about them, here.

Here there are in summarized form:

• Clarity of purpose
• Large markets
• Rich customers
• Focus
• Pain killers
• Think differently
  
• Team DNA
• Agility
  
• Frugality
• Inferno

I think it's an excellent list of principles for startups to follow and I think at Mark Logic we are doing a great job at staying true to them. We know why we're here. We're targeting a huge opportunity (80% of an enterprise's information is, after all, unstructured). We target customers "in pain" with problems on which they're willing to spend money to solve. We are ruthlessly focused in our execution. We, by nature, challenge the conventional wisdom. We have a strong team and are seeing increasing returns from having one -- great people attract more great people; it's self-reinforcing in my opinion. And we're frugal. Heck, I don't pay a nickel to publish this blog or get the analytics on it. In fact, the blog makes money, though I'm sure I could make more money pan-handling than I have from my AdSense advertising experiement. But let's not digress.

Ultimately, it's not my self-assessment of Mark Logic that matters. The guys who wrote the principles, Sequoia Capital, have invested in Mark Logic in each of our three venture rounds. So, evidently, they think we're doing a good job on them, too.

My net message: if you're a startup, read the Sequoia principles and live by them. And be honest with yourself in assessing your adherence.

10 Things You Need to Know about Facebook

Check out this article in Information Today entitled Facebook 101: Ten Things You Need to Know about Facebook.

If you're not in the primary Facebook demographic and you're not spending time using it despite that, then you really need to read this article. You've got to understand Facebook, and more importantly the concept of the social graph, going forward.

The social graph will simply be another piece of context that applications will leverage to help people do things. And the question then becomes (1) how many times do you want to store your social graph and (2) how do you avoid social graph train wrecks. On (1), do you really want to tell Yelp, Flickr, Spock, LinkedIn, Facebook, Plaxo, YouTube, BlueCollarOrDie, Classmates, Match, SlideShare, Scribd, Twitter, and DocStoc that you're friends with Sara Jones? Can't you just do it once or twice? On (2), how do you prevent your LinkedIn associates from meeting your stays-in-Vegas friends from Fling? How can you prevent a Facebook app from asking you to compare a board member's body with a customer's?

I realized in chatting with a young Mark Logician that the day Facebook announced its application platform strategy there were two types of people:

• Type 1: Those who immediately went wow, this is going to change the world
• Type 2: Those who went, huh, what does that mean?
  

Sadly, I was type 2 but I was intrigued enough to try and figure out what it meant and am happy that I did. Here are the 10 things you should know:

1. Who is using Facebook?
2. What can you find on Facebook?
3. Why are people using Facebook?
4. What kinds of 3rd party apps can you add?
5. What are advertisers doing on Facebook?
6. Who else is joining?
7. What groups are now on Facebook?
8. Why is Facebook so popular for photo sharing?
9. How do you find old friends?
10. What about privacy?

I'd be remiss if I didn't note that I get quoted at the end of the story talking about KickIt, our (rather experimental) Facebook app. Here's the excerpt:

Mark Logic, Inc. recently joined Facebook with its Kick It application, according to Dave Kellogg, president and CEO. “We launched it because we saw an opportunity to build a nice, simple example of the power of XQuery and XML search,” he says. But the app was created quite by accident (who said “Necessity is the mother of invention”?) by David Amusin, a new staffer at Mark Logic and a recent engineering graduate from the University of California–Berkeley. Amusin had an extra ticket to see the Dave Matthews Band and wanted to find a friend to invite to the concert. Facebook’s existing search wasn’t very helpful in searching for friends by interest category, so Amusin built the Kick It (aka “hang out”) app with the Facebook API. The result was a new way to help find people to “kick it” with and learn more about their friends.

Kellogg reports that bloggers who have found the “neat little app” have responded quite positively to it. But the company isn’t making a big push to drive traffic and doesn’t plan on making money on it. “In the midterm, I think more and more publishers are going to want to link with the social graph and associated information in building their products,” says Kellogg. “They will want to use content platforms that show demonstrable Facebook integration and to work with suppliers who understand how to leverage the Facebook API.”

SAS Acquires Teragram

In a not terribly surprising move SAS announced that it is acquiring Teragram on Monday. Seth Grimes of Intelligent Enterprise covers the announcement here. CMS Watch discusses it here.

This is basically a replay of the $76M Business Objects / Inxight deal last May.

At a trends level, this is about the BI vendors (can you say that anymore? perhaps I need to say BI vendor since SAS is the only independent left) wanting to perform analytics against both structured and unstructured data. For most of the evolution of the BI category, unstructured data was not part of the picture. Only in the past few years, has it really hit radar in the BI community.

The approach these folks typically take is use text mining engines to structure the unstructured data. For example, identify which documents talk about which products. Identify which documents have which tone. Then load that data into the data warehouse so you can run a report that shows sales by week with two more columns added: number of emails to customer support and % negative in tone.

Put differently, when you have a multi-billion dollar data infrastructure and you're faced with content what do you want to do with it? Turn it into data. This is not a bad start and it does enable more and better dashboards and reports. However, to build really powerful content analytic applications, I believe that you need a specialized server (i.e., an XML server like MarkLogic) to handle the documents.

In many ways, it's like OLAP. If you wanted basic slice and dice you could build that into a reporting tool. But if you wanted real OLAP -- big databases, instant performance, complex analysis -- then you needed specialized DBMS -- an OLAP server -- to do it.

Scholr 2.0: Interactive White Paper on the Future of Research

Check out this white paper, entitled Scholr 2.0, which discusses the future of academic libraries, scholarly research, and scholarly portals. While there is a PDF version available for old-schoolers, the white paper itself is presented using CommentPress, an open source theme for WordPress (a blogging engine) that allows readers to comment at a paragraph level in the margins of the text, greatly helping the collaborative creation of documents.

While it's not new -- the piece has been out since August, 2007 -- I still think it's a great read and it accomplishes its goal to provoke thinking and comment on the requirements for the next-generation of scholarly portals. If librarians are going to remain relevant in an increasingly full-text-indexed, folksonomied world, then how are they going to do it? It's not an easy question and I think the white paper does a good job of answering it.

MarkLogic gets mentioned in section 3.0 for our work in supporting the Scholars Portal. Excerpt:

With the migration of Scholars Portal to Mark Logic, an important opportunity has been created to incorporate Web 2.0 technologies into the discovery layer of this service. With strong support for XML, XQuery, relevance-based searching, facet-based browsing, thesaurus expansion, language-based stemming and collations, automatic classification, web services, AJAX and other agile programming approaches, Mark Logic gives us the kinds of tools we need to incorporate the best of current Web technologies into a new Scholars Portal interface.

My favorite section heading in the report is: XML + XQuery = data and content happiness. I couldn't have said it better myself.

Or-Die Networks: Verticalizing Vertical Search

I've tracked FunnyOrDie.com since its inception, largely because our Sequoia partner / board member, Mark Kvamme, was heavily involved in its creation. Mark's son, Michael, an aspiring stand-up comedian, came up with the idea. The site launched with a now famous video entitled The Landlord featuring Will Ferrel, below, which has been viewed more than 50M times. If you're so inclined, here it is.



Frankly, it's not my favorite sketch, but let's not miss the point. The beauty of FunnyOrDie is that it becomes not only a vertical comedy-focused YouTube, but it also becomes the center of a community of comedians. I have a friend or two in that community and they're all aware of the site and use it frequently. Plus, I'd say the mixing of uploaded user-generated and professional (e.g., Will Ferrel) content was another clever innovation.

The reason for my catchy headline is that some people would say that video search itself is a form of vertical search. Ergo, YouTube is a form of vertical search. Now, when you generalize the idea of FunnyOrDie to other market segments, you then end up verticalizing vertical search.

And, as I learned the other day, that's precisely what they're doing. They've transformed the concept from FunnyOrDie.com to Or-Die Networks and have launched two sister sites: ShredOrDie with Tony Hawk and BlueCollarOrDie with Jeff Foxworthy, Bill Engvall, and (my favorite) Larry the Cable Guy.

The formula, it appears is:

• Identify a category
• Get a relevant celebrity to lend branding and provide some content
  
• Leverage the technology platform

Clever stuff. I wish them luck.

Yahoo Investor Deck: Future Value and the Microsoft Deal

I found this interesting Yahoo investor deck in this post on Valleywag. It's basically 35 slides on their future, the future of the Internet advertising, and why the proposed Microsoft deal undervalues the company. Among other things it projects a 72% growth in revenue from $5.5B to $8.1B through 2010 and a doubling of operating cash flow from $1.9B to $3.8B in the same period.

Here it is. Enjoy. (Press >> to see the full menu and then press the rotate button to see the deck properly.)


Yahoo Investor Presentation - Get more free documents

The Blissful Ignorance Effect

I've always been a fan of studying how we, as people, make decisions whether they're big ones (e.g., Smart Choices, Decision Traps) or day-to-day consumer ones (e.g., Why We Buy). On the latter, I'd always felt that the Internet was a marketer's boon because Why We Buy shows the pains that merchandisers must endure in the physical world (using cameras and observers) that get replaced by a nice friendly clickstream in the virtual one. For example, market basket analysis can tell you what customers purchased in a supermarket and in which combinations; but only human observation can tell you each product they considered, which they looked at, which they picked up, why they tried to bend over to reach but stopped, and which they read the label on and then hastily put back.

All this is a long way of introducing an interesting article in today's New York Times, entitled Some Blissful Ignorance Can Cure Chronic Buyer's Remorse. The conclusion, named "The Blissful Ignorance Effect," is that people who have more ambiguous information about a product expect to be happier with their purchases than those who have bought with more specific details.

Excerpts:

... there is a shift in buyers’ goals before and after purchasing something. Professor Nayakankuppam says these are called “accuracy goals” versus “directional goals.”

Before the purchase, “people want clear, objective information — they want to make sure they get it right,” he said. “After a purchase, they want to reach a particular decision — they have a directional goal.”

[...]

Elliot Aronson, ... co-author most recently of “Mistakes Were Made (but Not by Me)” [dk: great title!] ... is not surprised by these findings. He has spent a lifetime looking at how people justify their decisions. “If you have just a little information and then you’re a little disappointed, you can convince yourself that it wasn’t your fault, or that the item may be better than you thought for other reasons,” he said.

In general, he says, people are “cognitive misers” — they do not want to do a lot of thinking and research. That is one reason that brands and slogans are attractive; they are a shortcut to information.

[...]

Max Kalehoff, vice president for marketing for Clickable, [...] says “A product that really requires a lot of detail or work to understand creates implicit barriers,” he said. “Marketers are throwing things at us with increasing complexity. It would be stupid to assume they should hide information from consumers, but they need to focus — from a product or service standpoint — on prioritizing the most important information and presenting it with the greatest efficiency.”

The article ends with some advice:

Research big decisions thoroughly, but don’t worry so much about the small ones. Don’t overanalyze why you like things — and in the end, you can probably convince yourself that whatever you have done is the right thing.

Go Green With MarkLogic

Check out this article in EContent Magazine entitled It Ain't Easy Being Green. It's about the green angle on moving from print to digital publishing. Excerpt:

Everyone from publishers to e-doc technology providers and sellers sense an opportunity here, and for many a common thread is the use of XML: Create the content once, and deliver in many alternative ways.

The article even includes a quote from MarkLogic's increasingly ubiquitous John Kreisa (that guy's everywhere; is there more than one of them?)

Kreisa described how U.S. Army soldiers ... can now specify custom battle books (think IED defenses) using [an application] ... based on Mark Logic technology. There’s no more ripping out tons of paper pages from manuals they don’t need, must ship, and then destroy.

While there are many reasons to deliver information products from XML respositories using MarkLogic Server, being more green is certainly one of them. With MarkLogic, we can pinpoint-locate exactly the information users seek -- and only the information users seek -- and then return it for the device and format that they desire. The pinpoint location comes courtesy of MarkLogic's full-text extensions to XQuery that enable search using text as well as structural and semantic markup constraints.

Aside
By the way, this looks like a welcome policy change from EContent Magazine. For years they put nearly all of their content behind the firewall which disincented me from blogging about it. A quick surf now gives me the impression that everything's now un-firewalled. EContent makes great content so I'm glad it's now easier to access. Example: I found this story through a Google Alert, which won't happen when everything's behind the firewall.

Documentum Post DeWalt: One Year Later

I've never met Dave DeWalt, but I've met plenty of folks who have, and they universally say good things about him. So I figured it wasn't great news for the Documentum group at EMC when DeWalt left about a year ago to become CEO of McAfee.

Today I found an excellent post on CMSwire entitled Documentum: One Year After Dave DeWalt. Among other things it points to a superb post by John Newton, co-founder of Documentum and now co-founder and CTO of Alfresco, entitled The Departed, which goes into great depth about what DeWalt accomplished at Documentum and John's suspicions as to why he left. If nothing else, read Newton's post; I don't know how I somehow missed it a year ago.

Here's an excerpt from the CMSwire post by Marko Sillanpaa:

But gone is the passion and energy Dave and his team brought to content management. While some may disagree with the idea that content management is cool, I doubt few felt that way after seeing Dave’s keynotes. Rappelling from the ceiling or entry on motorcycles or horseback (even with diapers) woke you up in the morning and got you listening to the rest of the presentation, no matter how late you stayed at the table in Vegas.

In contrast, the EMC World 2007 keynotes were given with all the enthusiasm of a tenured professor in a second rate junior college. You could really see the difference between the west coast software and the east coast hardware marketing.

Overall, the post starts with a pretty grim impression of the post-DeWalt world, but then shows signs of hope, starting with the un-retirement of Documentum's other co-founder, Howard Shao:

Documentum had been a tight knit family. And fortunately, in mid-year Howard Shao came out of retirement to hold the family together. It was disappointing though that while he left with a roar there was not even a peep when he returned. Howard’s return did what it intended. It settled folks down and even brought a few people back.

Joining Howard to take the reigns of CM&A was Mark Lewis, who had held several roles inside EMC including CTO. He’s only been in the role for six months so there’s been little time for change but EMC World is coming up. We’ll see if this long time EMC leader finally looks across all of the EMC products. It still baffles me that after three years few of the product lines talk to each other (EMC’s Newest Competitor EMC?). The other question, can he motivate the troops?

Marko ends his post on an hopeful note for Documentum's future. I'm slightly less optimistic than he is because of one word: SharePoint. Acutally, two words: SharePoint and Alfresco.

I think a likely future for the ECM category is SharePoint attacking from the left with Microsoft's standard iterative-improvement approach and Alfresco attacking from the right as the alternative to SharePoint. First-generation ECM vendors end up as the IBM mainframes in that scenario (i.e., they're expensive and everybody has one, but they aren't deploying new apps on them). I've blogged before on the similarities between ECM and BI, and I believe that while BI jelled as an integrated category that ECM never did.

But then again, I do have a bone to pick, because EMC acquired x-Hive a while back and while there is a high degree complementarity between MarkLogic and Documentum, there is a fair degree of functional overlap with x-Hive.

However, I believe an XML content server is strategic infrastructure for the customers we're targeting and they won't just take what comes in the box with a CMS. So while I expect the vendor relationship to be more complex than in the past, I do believe that plenty of customers will use Documentum for content management and MarkLogic for their XML repository.

That said, looking to the future, I do believe that SharePoint will put a squeeze on the classical ECM vendors and become ubiquitous, so we're increasing our investment in SharePoint and Microsoft Office integration. And we're thinking about an Alfresco relationship as my spider sense says there's a good chance they will end up successfully positioning as the SharePoint alternative.

Speaking of MarkMail: Update

I thought I'd provide a quick progress update on MarkMail.

In a word things are going, well, great.

In the approximately four months since its launch, we've done quite a lot. For example, we started with about 4M emails loaded and indexed; today that's more than doubled to about 8.8M. From memory, I think we had no more than a couple of hundred K pages indexed in Google when we started. Today, we're over 2M.

Traffic continues to grow impressively as well. I've anonymized the attached chart in terms of time frame and scale, but it pretty much shows the trend we're seeing: high traffic growth.

A major driver for deciding which content we load is the open source community itself. For example, over the past few months we've loaded mailing lists about the following projects:

• Xen: 115K messages across 12 lists
  
• Perl: 530K messages across 75 lists
  
• PostgreSQL: 640K messages 30 lists
  
• CodeHaus: 420K messages across 390 lists
• PHP: 700K messages across 23 lists

The above represents a lot of knowledge about these projects that's now about 10x more accessible than it used to be. By the way, in most cases we load these lists either because a community manager or someone in the community asks us to.

Heck, we even loaded some smaller community lists, like eXist, with 24K messages. Somewhere there's some irony in using a MarkLogic-based application to get information on eXist, but we won't go there.

We've even launched an excellent blog entitled, The Making of MarkMail, written by Jason Hunter and Ryan Grimm, where they provide both updates on MarkMail itself (e.g., announcements of new lists and new features) and discuss the lessons they've learned in building an Internet-scale service using XML, XQuery, and MarkLogic Server.

MarkMail gets better because of you. If you've got feedback about the service, a question, and/or want us to add a new list, let us know. Please go here to do so.

Tim O'Reilly Likes MarkMail ... A Bit Too Much?

I loved most of this post, Perl Mailing Lists Added to MarkMail Archive, from none other than Tim 2.0 (Tim O'Reilly). Excerpt:

This is awesome. Markmail also has mailing list history for php, ruby, mysql, postgresql, and all of the Apache projects. (Coverage of Python and Linux is very limited.) And as I written previously on Radar, the markmail search and trend graphing interface is really slick.

I almost choked on my donut on the next line:

Google ought to buy these guys and offer this product as an upgrade to Google Trends! Not to mention applying some of their interface insights into other Google search products.

Well, Google did buy our next-door neighbors (Postini) and I can say with amazement that "the food" stories really are true. Seemingly moments after the Postini announcement trucks started pulling up every day, bringing the famous Google (free) food from Mountain View. That all stopped a few months later when the deal closed and they moved the people to the food instead of the food to the people.

In any case, Tim's views on MarkMail ownership aside, we appreciate the kind words and thoughts on the service. We like it, too!

John Kreisa to Speak at DoDIIS

I'll be at the DoDIIS show in sunny San Diego next week and am looking forward to hearing keynotes from Lieutenant General Michael D. Maples, United States Army Director, Defense Intelligence Agency; The Honorable Dale W. Meyerrose, Associate Director of National Intelligence and Chief Information Officer; and Major General John D. Custer, Commanding General US Army Intelligence Center and Fort Huachuca.

Other highlights include the "Big 6 CIO Panel," a panel with the CIOs of FBI, DIA, NGA, NSA, CIA, and NRO.

Mark Logic's director of product marketing, John Kreisa, will be speaking Wednesday, March 19th at 230 PM in room 9. Here is the session abstract for his talk entitled: Using an XML Server to deliver Knowledge Management.

XML is used as a format to define, store, and deliver information in the government. Being able to natively understand XML as a format enables agencies to better manage, share and deliver knowledge across applications and systems. Increasingly, agencies are using XML to tie systems together to create applications for more fine-grained sharing of knowledge, incorporating user-generated content to enrich their existing knowledge base and make content more relevant. XML servers are platforms for building critical components for maximizing the investment in XML technologies. This session will focus on how XML servers are being used within the Department of Defense in organizations like the Army to deliver next generation knowledge management applications.

I'd like to reward foodies who read through the whole abstract with this tidbit. I'm also planning to eat at least once at the Studio Diner, a 24x7 diner featured on the Food Network's program Diners, Drive-Ins, and Dives. I've never been there, but it looks spectacular.

Marketing Lessons from Hillary

While this is a business blog primarily about enterprise software, marketing, and content technologies, I do always like to keep the blinders off and look outside the immediate space for examples from which we can learn. For public relations (PR), there's really one place to go: politics. I'm not here to take political positions; I'm here to study PR.

I thought I'd look at a quick, fun example from the Democratic presidential campaign.

I am a big believer that marketing always be logical and truthful. I assume the audience is intelligent and thus they will either explicitly identify contradictions in the story or, when such inconsistencies exist, simply walk away with a queasy feeling of confusion. Neither is what you want. So my rule on PR messaging is always be clear, direct, and logical.

I think Barack Obama's team did a good job of identifying an in-hindsight-obvious inconsistency in Hillary Clinton's recent Obama-for-VP messaging. The logic goes something like this:

• You say I'm not ready to be president.
• You say you might make me your vice presidential running mate.
• But as vice president, I could become president in an instant (should anything happen to the president).
• So how is it that I would be a good vice president again?
• You then top this sound logic with a good sound bite -- "I don't know how somebody who's in second place is offering the vice presidency to somebody who's in first place" -- and you've effectively jujistu-ed your opponent, using the energy in their attack against them.
  

In messaging, you need to take a position and stick to it. And you can't build the position piecemeal, based on what sounds good at the time. You can't logically argue that person A is not ready to be president, but argue that they should be vice president when the vice president can become president at any moment. As Larry the Cable Guy might say: that dog don't hunt.

To avoid problems like this, I recommend a few things:

• Build your messaging in blueprint form first -- i.e., distill the key messages to a few message trees. It's much easier to spot inconsistencies when looking at the message holistically and in distilled form.
  

• While it's still in a fairly early draft, show your messaging to your engineers. As a rule, engineers are great at identifying logical problems; they probably even enjoy it. This will cost you 6-12 pizzas depending on the size of your company and will be the best money you ever spent on marketing consulting.
  

• If you can't get the need for logic through some people's heads, go find the smartest, most logical analysts in your space and hire them to tear apart your product marketing manager's (PMM) presentations a few months before their launch. At BusinessObjects we hired Curt Monash to shred a few of the cockier folks at one point and I'm sure they'd attest there's nothing quite like being shredded by someone who earned a PhD in game theory from Harvard at age 19. Better to be destroyed months before the launch in private than the week after in the press.
  

Kawaski Interviews Balmer at Mix08

Check out this post with notes from the Guy Kawaski keynote interview with Steve Balmer at the Mix08 conference this week in Las Vegas.

In the interview Balmer talks about Google, the Yahoo! deal, Apple, his three types of day [see below], Silverlight, the Facebook investment, Fast Search & Transfer [see below], the number of emails he gets per day (~60), and he even gives out his email address: steveb@microsoft.com.

On his three types of day:

1. With customers. From 730 AM to 800 PM and then get on [private] plane to next city.
2. Doctor in office. Wall to wall meetings all day. "Exhausting."
   
3. Think, write, and research.

On Fast Search & Transfer:

Fast is company had internet and website/corporate products. Sold off web search. They have great for high end search on enterprise and engines that can search web sites. Tech fantastic and team is great. Anxious to build both ways. Love company/people. Great integration plan - more to say.

This is consistent with my thesis for why Microsoft bought Fast (to fend off the Google Appliance in high-end enterprise search, aka, the best defense is a good offense). However, I'd not previously heard the message that they want to build Fast out "both ways" -- i.e., in enterprise search and in their Internet search offerings.

The only part of the acquisition that continues to amaze me is the ~8x revenue-run-rate price. That kind of multiple is in-line for high flyers, i.e., for healthy, high growth enterprise software companies. But Fast was in the midst of unwinding a world-class accounting mess, complete with lots of AR write-offs and a revenue restatement. I'd think companies in that situation are usually lucky to trade for 1-2x revenues.

Much as the the price SAP paid for Business Objects wasn't surprising until you noticed that Business Objects was about to announce a quarterly miss, nor is Microsoft's price for Fast surprising until you consider the not so easy to overlook financial mess. Personally, I would have guessed a sale in the $300M to $500M price range, proving that I'm not always right.

My current speculation is that there must have been a bidding war for the price to get so high. The fun question then becomes who else was bidding, why did they want it so bad, and what are they going to do now that they've lost?

Buxton Interview on Microsoft Office Open XML (OOXML)

Check out this two-minute interview with Mark Logic's Stephen Buxton (author of Querying XML) where he discusses Microsoft Office Open XML format (OOXML), which Microsoft evidently now refers to as Microsoft Open XML format.

This is one of a series of videos on Open XML, which has its own landing page on YouTube, here.

Visualizing Warren Buffet's 2007 Shareholder Letter

Fans of Berkshire Hathaway and Warren Buffet have come to love his lengthy, poignant, wide-ranging, well-written, and often funny annual letters to the shareholders of the company. His recently released 2007 letter is no exception and I highly recommend reading it, here.

Among other topics, he talks about the real estate slump, the expected return on stocks over the long term, their management and acquisition philosophy, the worst mistake he ever made (Dexter shoes), their insurance businesses, the state of the dollar, how public companies "juice" earnings, and the fun details around the upcoming shareholder meeting in May.

But the purpose of this post is to point to a post on the wonderful Infectious Greed blog where Paul Kedrosky uploads the letter to IBM's many eyes for visualization. Check out Paul's post here.

Personally, I'm not in love with what the tool lets you do, but I always think it's good fun to try new tools on fun content.

SAP / Business Objects 101

It seems to me that SAP hasn't spent energy much getting the Business Objects story out, particularly when you consider that they spent $6.8B to buy them. What I do hear about the strategy, the integration, the overall story, I get through my BOBJ alumni backchannel.

Thus I was pleased to find that Business Objects employee #8, Timo Elliott, has recently posted a video to his blog that does a great job of covering what (I think is) the SAP / Business Objects story at the 101 (i.e., introductory) level. It's 23 minutes long, and it feels a bit like a training session, but if you want a comprehensive walk through the story, I don't know a better place to get it.

Here is a link to the video.