Well, it’s that time of the year: prediction season. As we traverse it, I’ll highlight some of the better predictions lists that I encounter. The first one I’ll pick comes from CMS Watch.
Here are their predictions (the bolded points are their predictions, quoted excerpts of the copy beneath each prediction are theirs, and my commentary appears prefaced by DK):
1. Open source ECM players get an initial boost
DK: I agree. I think Alfresco is leading the way here, bringing an elegant and simple approach to a market that’s overloaded with complexity. It’s telling that the Wikipedia entry for ECM is a disorganized, kitchen-sink list of features, which includes BPM which in turn includes BI (suggesting absurdly that BI is part of ECM), and is prefaced by: “All or part of this article may be confusing or unclear.”
2. Office14 casts long shadow on SharePoint
The build up to the next major release of SharePoint — hitting beta perhaps as early as the end of 2009 — will rightly focus on its deep relationship with Office 14, but will also yield a collective pause among customers as the year progresses. Those that have run wild with SharePoint as a potential ECM replacement throughout the enterprise during 2007 and 2008 will be reassessing their strategies.
DK: I’m more bullish on the replacement of classical ECM systems, such as EMC Documentum, with cheap and cheerful alternatives like SharePoint. To me, ECM is a classic bloatware market, with too much functionality, too much complexity, too many failed projects, and too many unhappy customers. I think it’s begging for a return to simplicity and a focus on what matters: dynamic delivery of content to users (or, to cliche it, delivering the right content to the right person at the right time).
3. “Taxonomies are dead. Long live metadata!”
You should be able to get to information the way you want, which may be different from your colleague’s approach. We still need controlled vocabularies. We still need to tag content. Text mining and auto-tagging software is gradually improving, and extracted terms can be applied as metadata. But that metadata needs to be a lot more fluid, cloud-like, and by no means fixed in a single hierarchy.
DK: we love metadata and we think XML is the right way to represent it. And we think that rich XML content is stored in no better place than MarkLogic Server. By the way, aren’t taxonomies just one form of metadata?
4. Regulatory-compliance concerns reignited
5. Renewed interest in pro-active e-discovery
6. SaaS vendors expand offerings
Many Software-as-a-Service (SaaS) customers across a variety of content technology segments — from WCM to social software to e-mail archiving and document management — are asking for more than just technical services from their SaaS providers. Customers will continue to view SaaS vendors as extensions of their internal teams and will increasingly look to them more as one-stop consulting services, rather than just the system used to manage their content and visitor interaction.
DK: I like SaaS and I think both SaaS and perpetual software play a role in the future. Using Geoffrey Moore’s terminlogy, SaaS is for “context” and custom apps built on perpetually-licensed platforms are for “core.” See more of my thoughts in this post: To SaaS or Not To SaaS.
7. Oracle falls behind in battle for knowledge workers
But 2009 will expose Oracle’s weakness with front-office applications at a time when Microsoft, IBM, and many smaller players are fighting for the hearts and minds of knowledge workers.
Customers are already feeling indigestion, as different Oracle teams market overlapping and often incomplete solutions. For example, Oracle is struggling to combine four different enterprise portal offerings, and many customers are chafing at the financial and architectural challenges of aligning with the putative winner, Oracle WebCenter Suite (OWS).
DK: In the spirt of one of my favorite quotes, “when did we become our parents,” let me ask: “when did Oracle become CA?” It’s happened over the last 5 years, almost without noticing. First, they put bankers in charge. Then, what we saw as a new era of software consolidation was in reality little more than a CA replay: Oracle did for the client/server era companies what CA did for the mainframe era ones.
In the end, I think “indigestion” greatly understates the massive integration problem that Oracle has signed up for. A friend once quipped: “I met the chief architect at CA, and that must be one hell of a job, …”
Today, you could say the same of Oracle. In reality, I don’t think they’ll ever integrate. They’ll simply ride the maintenance revenue streams into the sunset and, once a given stream becomes unprofitable, they’ll force migration to a related one.
8. New emphasis on application search
In a world where system users increasingly evaluate the efficacy of their applications by the quality of the underlying search mechanism, vendors have to make adjustments.
“Faceted search” will become an RFP checkbox feature and this, among other things, will cause Web CMS, Portal, and ECM vendors to reassess their bundled search solutions. Expect to see OEM partner relationships get juggled around. For example, some existing OEM deals with FAST and Autonomy might not get renewed.
DK: I agree, and I’d note that Mark Logic’s OEM business is booming. MarkLogic easily enables the features described above (e.g., faceted navigation) and, particularly where the content is stored in XML, MarkLogic can be an ideal enterprise search alternative.
So I’d add MarkLogic on the XML side, and Lucene on the open-source side, to the list of competitors vying for replacement of traditional enterprise search engines, such as Verity, in OEM — and for matter, SaaS –applications.
9. Social computing dif
fuses into the enterprise
There is a growing debate about the relevance and advisability of enterprise investments in social software during an economic downturn that is supposed to focus everyone back on “the basics.” Yet, social software will not fade away, adoption will continue regardless of enterprise policies.
DK: I think social softare is here to stay. While Mark Logic is not currently focused on Enterprise 2.0 applications, we do most certainly make it easy to build web 2.0 style apps (e.g., tagging, annotation) which most of our Information/Media and Government customers do.
Simply put, public websites now set the expectations for functionality and user interface for most classes of software.
10. Long-awaited consolidation comes to the WCM space
11. Mobile and multimedia web analytics become key requirements…and disrupters
DK: Sitemeter, Feedburner, and Google Analytics work just fine for me.
12. Buyers remain in driver’s seat
In the current economic conditions, this leverage of the buyer will remain stronger than ever in 2009. Buyers will be placing more emphasis on researching the vendors and products. There will be a renewed emphasis on identifying a vendor and product’s associated levels of risk.
DK: I have mixed feelings on this one. While the economy suggests reduced IT spending which in turn should increase buyer leverage, let’s not forget the other side of the equation: consolidation greatly reduces buyer leverage. Let’s not forget that both effects are at work.
For example, if you’re a pure Oracle shop and you want ECM, then you are either (1) going to use Oracle offerings or (2) use someone else’s (e.g., IBM, Microsoft) but increase entropy by bringing in another supplier. Since mega-sellers know that most CIOs are (perhaps wrongly) trying to reduce entropy by cutting the number of their suppliers, the net effect might well be to increase seller leverage.
I’ve been noodling on this idea for a while because the herd mentality of IT tends to reduce buyer leverage in the aggregate over time. For example, early on, buyers had quite some leverage on Salesforce. But now, if Salesforce came along and wanted to increase rates 30%, what exactly would you do about it? Other than hold your breath and get angry, you don’t have many alternatives. And the reason you don’t is that everyone else did what you did.
Similarly, if you’re a pure Documentum shop and EMC wants to give you xDB “for free” as part of a big enterprise license, have you (1) gotten a good deal on an piece of technology or (2) reduced your buyer leverage with EMC by increasing your dependency on them and condemned yourself to use an inferior XML Server at the same time?
I think I’ll do a future post on this topic.