Entries from July 2009 ↓

Now that the deal between Microsoft and Yahoo is done, it got me thinking more about Yahoo and it’s position as a pioneer of the web. That it was one is not in doubt; its directory was indeed a first and useful. But how much else did it actually pioneer? Have a look at a list of Yahoo’s acquisitions.

• Search – surely the cornerstone of the business, yet its very first search engine was licensed from Open Text in 1995. Sure, it went on to build its own and do some great work, but now with the deal done with Microsoft, Yahoo has exited the search business.
• Webmail – I recall getting a great Yahoo email address when that came out in 1997; it was a pioneer, but it got there by purchasing Four11.com and its RocketMail service . Now my account is overrun with spam and unusable (yet the company has the temerity to regularly threaten to cut me off for having too much mail and not using it enough, all of which is spam that its own spam filters can’t control. Hmmm.)
• Personal publishing – it was an early major player in personal publishing now called blogging, but again it got there largely by purchasing GeoCities for an enormous amount of Yahoo stock.
• Advertising – it pioneered banner ads on the web but as we all know but caught out by keyword search innovation from Google (which it built, rather than bought). And it had to buy Overture from Idealab.
• Rich media – the Broadcast.com deal unleashed Mark Cuban on an unsuspecting world.
• Photos – to its credit and bought Flickr and then largely left it alone
• Social bookmarking – same goes for De.licio.us

But if the rest that can be said for it is that Yahoo knew when to leave its acquired companies alone, to give them space to grow and continue innovating then all that’s left is a dwindling brand and company with some choice assets left intact for others to pick up over time. As this article pointed out recently, many of these acquired assets have been closed down, some only after a few years.

There is still innovation happening within Yahoo, plenty of it and I wish the folks at Yahoo working in the labs on some great semantic technology, among other things the best of luck. But touting a new home page just last week doesn’t give me much hope that Yahoo really gets the distributed, read/write Web. Who cares about home pages?

And of course Yahoo isn’t going anywhere soon, it has plenty of cash in the bank and a new revenue stream courtesy of Microsoft. But I doubt it will last out the 10 years of this deal.

One of things all this demonstrates is that M&A is different in different parts of the tech industry. In enterprise software quite often a company is buying a customer base and its ongoing maintenance stream – this is how Oracle has grown. But on the web with people not paying you directly and with very low (to nil) switching costs to another search engine, serving up a different set of ads, things are very different and you have to focus on core competences, not run after each new fad just as it’s peaking and buy your way into it.

Interesting news from Ingres today that it is teaming up with VectorWise, a database engine spin-off from Amsterdam’s Centrum Wiskunde & Informatica (CWI) scientific research establishment, to collaborate on a new database kernel project.

The Ingres VectorWise project will create a new open source storage engine for the Ingres Database that will better enable it to be positioned as a platform for data warehouse and analytic workloads, although Ingres does not have detailed plans for the productization of the technology at this stage. The starting point for the project is the theory that modern multi-core parallel processors now look like, and behave like, symmetrical multi processing (SMP) servers, and that on-chip memory is taking the place of RAM, but that database software has not been updated to take advantage of process developments.

In order to do so Ingres and VectorWise will be collaborating on vectorized execution, which sees multiple instructions processed simultaneously, and in-cache processing, through which the execution occurs within the CPU cache and main memory is effectively treated like disk. The result, according to Ingres, is to reduce the I/O bottleneck for query processing. Additionally, the VectorWise engine enables on the fly decompression and operation handling in memory and includes a compressed column store.

It is claimed that the Ingres VectorWise project will deliver 10x performance increases over the current Ingres database.

VectorWise span off from CWI in 2008 to commercialize the the X100 system previously created by its database architecture research group. Development of X100, now also known as VectorWise, has been led by respected research scientists Peter Boncz and Marcin Zukowski.

Ingres maintains that by working with the CWI research scientists it has proven that their theories are technically feasible in a commercial product. Bringing such a commercial product to general availability is the next step, and history has proven that can be easier said than done. With that caveat we are impressed with the vision and ambition that Ingres is demonstrating.

Quick thoughts on the deal. We will have a full report for clients tonight. This is mainly thoughts about the text analytics part and I haven’t had a chance to speak with either company at the time of writing, so bear that in mind.

• This is long-predicted, by us and many others. I recall a chat with SAS founder and CEO Jim Goodnight a couple of years ago and he said it me – and I’m slightly paraphrasing –  in so many words, “why doesn’t IBM just buy them, I don’t understand why they haven’t already?” Well IBM finally has, or at least has made the initial move. And for $50 per share or almost $1.2bn.
• Of course like almost every IBM deal in recent years, the two are partners, IBM signed an OEM deal for the SPSS’ PASW statistics software in Q2 and has had other deals with it going back many years.
• IBM has text anlaytics tools, of course but they really are just that; tools. It is not a major player in text analytics applications at this juncture. The vast majority of its engagements tend to be very large, custom-based ones and are still few and far between, as far as we can gather, mostly in financial services and telecommunications.
• SPSS, on the other hand has tools, workbenches and applications and has found  some hot spots in this area, including analyzing customer feedback surveys, in particular the open-ended questions that can provide some of the richest material in such surveys but are often ignored because they’re too manual-intensive to analyze by hand.
• SAS Institute now has a much bigger analytics competitor. Goodnight didn’t rate SPSS much as a competitor, but IBM? That’s a bit different.
• SAP-Business Objects must be thinking of making a move too.

More considered thoughts from myself and my fellow 451 analysts later on today.

451 clients will have seen my report of Autonomy’s Q2 results last night, so I’m not talking too much out of school here, but one of the more interesting things for the longer term from its conference call was the identification of three phases of evolution from basic eDiscovery through information governance.

The spot in the call where this was examined was given over to COO Andrew Kanter, who is a lawyer. He didn’t elaborate on it as we has clearly reading from a script (so much so that he said “click,” at the end of each slide ;)), but nevertheless I though it was interesting to note and pass on.

The three phases, which the company believes will encompass roughly five years at most large organizations are:

1. Archiving and basic e-discovery as companies deal with litigation or are not in compliance
2. Legal hold and early case assessment – part of what it calls advanced e-discovery – when companies come to the conclusion that manual methods of legal hold – sending emails out to the employees saying not to delete things – don’t work.
3. The third phase is information governance, i.e. the policies and technologies meant to dictate and manage what corporate information is retained, where and for how long. 
   

At the moment, the company is seeing ongoing work in phase one and the start of work in phase two. It has one unnamed client doing phase-two work – a Wall Street institution – with 70,000 desktops and 490TB of data to manage across six geographies. Autonomy says the number of potential deals in its pipeline for phase two has increased in the last quarter, but its timelines are still a bit fuzzy. But it seems like Autonomy is not seeing any phase three, i.e. full-on, enterprise-wide information governance work at the moment.

We have seen this movement from e-discovery to information governance in our own research, but we’ve also noticed how early we are in that process. In fact Kathleen Reidy is about to publish our report on information governance that picks up directly from where our December 2008 report on e-Discovery and e-Disclosure left off. In this new report we will examine various approaches to information governance and how it will impact the market for archiving, content management, search and e-Discovery going forward. Kathleen or I can provide more detail should you require it.