Friday, May 26, 2006

Hummingbird Acquired

So Hummingbird has been acquired, and I doubt this is the last ECM acquistion this year. The liklihood is that private equity will move in now to consolidate many of the mid tier players. Speculation is rife around OpenText the other key Canadian ECM vendor.

Hummingbird has had a fairly torid time of things since PC Docs first merged with Fulcrum in 99 (I think). Fulcrum was at the time the leading Enterprise Search tool - the merger turned into a serious mistake and the firm seemed to spend the next 12-18 months battling to decide who was boss. In the process upstarts iManage moved into their Legal space and began quickly to first challenge then replace them. Then connectivity firm Hummingbird stepped in and acquired the PC Docs Group as it was then known. It's been downhill - slowly from that point on.

Yet its worth remembering that PC DOCs was:

  • One of the first EDM (enterprise document management) tools on the market
  • Dominated the Legal sector for years
  • Fulcrum was the finest enterprise search engine in its day.
  • Hummingbird was once one of the top 50 software firms in the world
  • They were pioneers of the enterprise portal

They also recognized very early that the major infrastructure vendors would ultimately take over this space, and were pretty much the first of the larger players to align themselves with one (Microsoft).

So its really rather sad to see them snapped up by private equity (Symphony Technologies), but without this their future looked bleak indeed. The past week or so I have started to feel old as I watch the industry morph into something new. For ECM never really was anything new, it was always more a branding concept for lots of document and web content management technologies bundled together than anything more. WCM was the last real mini revolution, and I was fortunate to ride that wave. But it is still sad to see old timers go, and go largely unreported as people forget how great they once were.

Of course Hummingbird continues - (Symphony did buy them, not close them down!) but the chapter on PC Docs a pioneer in this industry is closed.

Tuesday, May 23, 2006

Sun Setting on ECM?

ECM isn't dead, but it's certainly on the way out - or is it? Frankly I can't make my mind up. There was a telling picture at the AIIM Expo and Conference in Philadelphia, as the imaging side of things continues to grow and grow, yet ECM seems to shrink. Yet ECM in the enterprise is clearly growing at a pace. Its all a fuddled mush.

Managing unstructured data is clearly high on the agenda, and I have no doubt it will continue to maintain this status over the coming years. Yet the role of ECM specific vendors seems to be retreating. I wonder if it is a little like Records Management. A couple of years ago in the wake of Sarbanes Oxley and a slew of new regulations, Records Management vendors were being acquired at a pace, and all looked set for huge growth. But it never happened. Or rather RM took it's rightful place on the Boards Agenda, but this didn't translate into increased sales of RM technology.

Seems the same is happening in ECM. Some long time vendors are losing market share to the likes of ECM Documentum, who like FileNet seem to keep on growing briskly. But its more like the same size pie, with a few players taking ever larger slices at the expense of tier 2 and 3 vendors.

There can be no doubt now that the infrastructure vendors will eventually own this space - and it makes sense that they do. Technology is technology, data is data. After the event I received a wonderful and thoughtful email from George Parapadakis of FileNet (ex CapGemini - and founder of ocument Avenue - an excellent early source of DM related commentary on the internet) in which he states (far more eloquently than I) that we have been fooling ourselves for years that ECM is in anyway different to regular data management. He is of course right.

But ECM and ERM are still vitally important topics - even if we buy into the fact that most major ECM deployments could be just as effectively built using App Server and Database Technologies as they can with dedicated ECM Suites - ECM remains very important.

And this is what I worry about most - for even if data is data, and IT is IT - managing documents in the enterprise is a highly specialised and mission critical skill set. Understanding how people interact in complex business processes with documents (be they digital or paper), enabling that process, and yet also understanding underlying regulatory, file structure, retention and destruction requirements - with a profound understanding of how to architect information is not for everyone. Those who can build an ECM system, are not neccesarily (seldom in my experience) the same as those who can design or deploy an ECM program.

In my speech at AIIM (huge personal relief at the big turnout at what was the graveyard slot on thursday!) I emphasised again and again that these 'soft' skills are few and far between - and yet essential to the success of any ECM or WCM deployment.

Somehow we need to stimulate more of a discussion and more activity around the business implications of ECM & WCM, encourage more to learn how to analyse business processes, design solutions and learn to architect information in creative and innovative ways that have a major impact. Without this work we will continue to see ECM & WCM deployments fall way short of expectations and requirements.

It was wonderful to see so many old friends at AIIM, I see ECM as my second career and it was not one of choice but at the time of financial neccesity (I was a photographer and writer in my first career). It is now my career of choice, I remain because I have developed a passion for the topics that ECM, ERM & WCM embrace. The challenge now is to forcibly separate the 'hard' tech work from the 'soft' business analysis and solution requirements in this area.

I would love to publish Georges email in full here - but believe it deserves a bigger audience (Editors take note!). From now on I will personally lead my practice with this mind set. IT is IT - managing unstructured data has its unique issues, but at the end of the day Data is Data - and the technology of ECM should be in the Architecture and Infrastructure practice. The business of analysing document centric processes also has its own unique issues, and to date they have not been separated out cleanly enough from the demands of the technology. I suspect I have the Worlds best job at times, I lead a small ECM business consulting practice, that is supported by the Worlds largest team of ECM & Portals consultants (approx 2000). That brings with it a responsibility to understand the technology and its limitations, but to recognize that the value we bring is in understanding core business needs and designing processes and systems that will meet those needs. Common sense really - but as I say ad nauseum, common sense is not as common as we like to think.

So not sure where this is all leading, or what clear conclusions to draw, beyond this years AIIM was personally a bitter sweet affair. I saw the sun setting on an industry I have lived with for many many years, yet I can see the sun rising elsewhere. Leading our clients in the right direction will be a major challenge.

Thursday, May 18, 2006

Google and the Enterprise

I attended this mornings keynote at AIIM which was given by Dave Giraourd of Google Enterprise. Though I learned little new about Google I was left with lots to think about.

The Enterprise (large businesses) is a very different concept through Google's eyes than mine. And I don't want to be so presumtuous as to say that I am right and they are wrong, but I suspect I am right and they are wrong :-)

One of the key statements of the keynote that was emphasised many times is that enterprise software is not designed for users and it should be. Well I don't think it is that simple. Should it be usable and efficient? Yes, but business applications are designed for businesses by businesses
to meet specific business needs.

Then the speaker posited a couple of others things that made me scratch my bald pate. He said that search is not useful enough for end users - and then later said that if a search takes longer than 1 second people just don't bother with it and abandon the search very quickly...If this is the case, then how come search is such a mission critical tool?

He also went on to say that it is not realistic to standarize methods organizing content in a large organization. Yet this is precisely what forward thinking companies are doing, and with great effect..

Then.."Employees and Consumers are the same people" - well not really I am a husband, a consultant, a father, a writer and a rotten golfer. I have quite differing demands and needs to satisfy in each of those roles.

I realize that Google has a good search engine - and it may well have one that is good for enterprises also, but based on todays speech and gathering feedback from others who listened, the enterprises that Google are pitching too are few and far between. Google for consumers is a very different proposition, there they have targeted perfectly the masses. But in an enterprise things are filed and there is a perfectly logical (if funky at times) order, most people know where things are and simply go there (no wonder they give up bothering if a search tool takes more than 1 sec to retrieve an answer). In short Enterprise Search and Consumer Search are very different beasts indeed.

Friday, May 12, 2006

Service Oriented Content

I had a great conversation with my colleague and Senior Architect Sam Leung this week regarding the concept of SOC (Service Oriented Content) and COS (Content Oriented Services).

I declare here loudly and boldly that this is the first time these acronyms have been coined! All and future arguements regarding the origin of these brilliant acronyms will be settled by referal to the web archives :-)

I will touch on this topic next week at the AIIM conference and plan to write an article outlining this topic further, but in short:-

SOA (Service Oriented Architecture) is now past the very early adopter phase, and is proven as a logical and valid concept. Many of Wipro's clients are embarking (or are already well down the road) to building and designing SOA's that are enterprise wide. Yet many have till now largelly ignored the issue of unstructured data. The leading vendors, though shouting SOA loudly, have in fact been very slow to respond and find a place in the architectural layer. Why this is, could be for many reasons too complex to go into here. Nevertheless ECM has been a bit of a SOA wallflower todate.

Yet in strategy discussion with large enterprises - even though they don't neccesarily talk about ECM as such, unstructured data is moving rapidly up the agenda. For if we look five or ten years down the road, and recognize the layered approach to computing that SOA promises, then we need not just usable webservices, but content designed and managed specifically to feed these services (Services Oriented Content), likewise we will build Content Oriented Services, and arguably are already doing so in some basic ways.

Its a complex topic that deserves more than my cursory mention here - and as I say I and my colleagues will write more on this in the future. But the concept is one I believe worth considering, for it raises questions of - how we should build and structure object models and repositories that can layer seamlessly into the infrastructure - provides thought for how we can build more exciting and usable pull as well as push applications - what are the opportunities and challenges for SUN, Oracle, Microsoft, IBM - how does this fit into the convergence of Telco and IT we envision in the future etc etc

Monday, May 08, 2006

IT vs Business Disconnect - again...

In this weeks Economist (6th - 12th May Edition) there is a very interesting article on the state of Outsourcing in China, for those interested in the topic it's well worth a read. But a particular sentence or two really caught my attention:

Describing Chinese engineering graduates "It is as if they can describe a hacksaw and how it works perfectly.......but have no idea how to build a door with it"

I find the same thing with many IT only led projects - endless time and cosideration is given to the exact configuration of application services, pre-processors and database arrays, but little to mapping this to the needs of the business. Or rather they are under the impression that they are building a system that maps to the needs of the business, but the disjoint is so absolute that they cannot see it.

A very common illustration of this in my world of ECM is the depth of consideration IT gives to the performance issues of the 'system' but how little time to considering a logical mapping of repositories, cabinets, folders and documents to the granual operations level.

Or the same problem from a different angle...The alarm bell's always go off on a project when I hear somebody say that " the users need instant access to this data ". 'Instant, can mean many things it can mean sub-second, or it can mean a second or it can mean within a day of the data being captured (as due to the actual business process I can do anything with this data till the following day anyway.....

Other alarm triggers are people who look blankly at me when I speak of stakeholders. If you have not even stopped to consider, group and prioritize various stakeholders needs and requirements then we need to talk.

Selling the concept of spending far more time gathering and analysing requirements to customers though is hard - and I personally love the idea of joint risk and reward projects. Were I ever to run my own firm this would be the model of charging I would use. For the reality is that too many people cut corners with business consulting and yet lavish money at boxes and software, later to moan that the deployment failed or fell short of expectations. With a risk and reward model, I would be able to demand the prep work of discovery and stakeholder analysis be done properly :-)

On a separate note or two: I read through the edited version of the article for Intelligent Enterprise - once again I owe a huge debt of thanks to an editor, this time Doug Henschen. I think the piece now looks great and am looking forward to publication. There is also a small spin off article that I think will go live this week on Open Source ECM that I shall post a link to.

Looking forward to AIIM next week - George Parapadakis is getting his MIT award, and I have a huge number of people I want to say hello to. If you are to be there too say hello to me, don't be shy! My speaking slot is the track keynote at 10am on Thursday, fingers crossed - I have given this a huge amount of thought, and had some wonderful input and feedback from so many. So this time its not just Alan on stage spouting off, its Wipro, and this blog - the response on the blog (though as usual most seem to want to communicate off blog - and that's ok) has been heartwarming. I realise that having a passion for ECM and RM and more importantly Doing IT Right (better) is a pretty lonely furrow to plough, but it fires me up to know I am not the only one.