James Governor's Monkchips

Tivoli and Lotus acquisitions weren’t failures. More thoughts on why IBM should buy Amazon

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

I was reading this post from IBM Eye (thanks very much for the props Greg) which took my IBM should buy Amazon post seriously (it was meant seriously).

According to Greg, Amazon’s CEO Jeff

“Bezos would take on a man of the people / savior-rescuing-his-company-from-the-big-bad-business-people role”

Dan Farber agrees that Bezos wouldn’t be interested:

“In the end though, it’s just too alien to think of Big Blue and Jeff Bezos together…that is unless IBM has a radical change in how it views its future business, which is unlikely, and Bezos wants to give up control.”

Would becoming IBM’s Internet Services Division really be a step down? We’re talking about a major transformation, bringing IBM into the core web infrastructure game, redefining an American Icon. If Palmisano gave Bezos the space to succeed (Bezos definitely wouldn’t agree to a deal otherwise).

Now its time to talk some history, which many of you will disagree with. You see, its my contention that Lotus and Tivoli were not failed acquisitions. These deals changed the IBM culture and business model fundamentally, and charismatic leaders came in to drive change (though they later left of course).

Soon after IBM acquired Tivoli in 1996 Frank Moss, Tivoli’s CEO, humourously described the deal as a reverse takeover by Tivoli. He did have a point. IBM didn’t have a culture of platform independence at that point. When IBM purchased Tivoli it brought a culture of dedication to heterogeniety. The IBM bean counters were initially horrified when Tivoli made its first purchases of Sun hardware as development and test machines. IBM – buying hardware from Sun? Use a frigging IBM RS/6000 instead came the order. Moss and his team had to make it clear that they weren’t there to support IBM hardware, they were there to build software that ran on all key platforms, and managed all key platforms. Moss won. IBM Software Group was arguably born at that moment.

So what about Lotus? Naysayers, many of them Microsoft heads, argue that Lotus brought IBM nothing. But Lotus kept IBM in the human-facing software business. Lotus is today seen as a SWG engine for growth. But there was another seminal technical decision that gave Lotus a lot of its technical power (you may think its clunky but Lotus Notes was definitely a powerful tool for building apps that touched unstructured information) – which was supporting Microsoft OLE. You can bet some IBM voices argued “supporting Microsoft object models on the desktop? What a terrible idea, lets push out that Microsoft code and build our own.” But the Lotus team maintained its focus on the market, which was voting with its feet, steaming towards Windows and COM.

So Lotus and Tivoli were transformative acquisitions which brought charismatic leaders into the mothership and encouraged change on a grand scale. These two companies were the basis for IBM’s software group. DB2, for example, was still effectively a mainframe franchise (arguably that only changed when Janet Perna handed over the reins to Ambuj Goyal and the division rethunk its role as Information Management). Without Lotus and Tivoli’s cultural changes WebSphere could never have become the success it has. In the 2000 timeframe Sun Solaris was SWG’s biggest market… SWG really was an independent software vendor (ISV) by that point. IBM makes a lot of money selling software to run on Windows. Failed acquisitions? I don’t think so.

IBM learned how to do software acquisitions from 1995-2001. Now its executing on a working model called Blue Washing. IBM understands the balance of hands off to hands on far better than it used to. But most of its recent acquisitions are not transformative so much as incremental. Its time for IBM to get bursty.

Before signing off I would like to add another couple of reasons why an IBM Amazon merger might make sense.

1. Amazon Web Services doesn’t compete with the companies that build on its infrastructure. In IBM’s parlance it “doesn’t sell applications”. Unlike Google and Yahoo, which keep buying companies to fill gaps, but creating also rans in the process, Amazon seems fairly purist from a infrastructure vs application perspective. That’s a nice fit for IBM.

2. Doing a deal would remove the need for stinky legal arguments about ecommerce patents.

3. Its all about location, location, location. The location in question being Second Life. IBM is the computer company investing most heavily in Second Life and immersive environments. Amazon is the web company investing most significantly in Second Life. Both companies have a great chance to get ahead of the curve when it comes to Second Life meets First Life API and service mashups. So why not join forces. Call it the first major acquisition of the Metaverse age. Or maybe that will be Linden Labs (duh!) Real Life shopping is coming to Second Life. Imagine the growth opportunities in more immersive online shopping. Then there is the ability to print in 3D and bring SL objects into RL.

disclosure: IBM SWG is a client. Amazon’s A9 subsidiary is a client.


7 comments

  1. I’ve never been a big Notes/Domino fan, although I thought they were potentially more powerful than what MS had in the same space, but recent activity under the Lotus brand, notably Quickr and Connections, and their good fit with the Websphere products and the Eclipse RCP suggests that IBM will see greater value from the division in the future.

  2. Interesting article.

    The Tivoli and Lotus acquisition folklore is not something I can comment on (it was well before my time) but are you sure about the anecdote about buying Sun hardware? I could have sworn MQ ran on multiple platforms way before 1996. It must have been tested on *something* 🙂

  3. Cheers Ric. nicely expanded. You really are getting more Lotus interested, aren’t you?

    Richard – I am sure about the Sun hardware anecdote. Perhaps the Tivoli execs were exaggerating but… I accept that MQ was multiplatform, but that doesn’t make it any less likely IBM bean counters would have pushed back about the “new boys” buying competitive hardware. Bear in mind SWG only announced support for HP-UX as a top tier platform in the late 90s.

  4. “but that doesn’t make it any less likely IBM bean counters would have pushed back about the ‘new boys'”

    Fair point.

    “Bear in mind SWG only announced support for HP-UX as a top tier platform in the late 90s.”

    One of my first jobs in IBM was running the stress test suite for Message Broker on HP-UX. My thoughts about that operating system are probably best left unshared 😉

  5. and the word is that ALL presentation-layer technology (including Websphere Portal) will be branded as Lotus …

  6. […] , redmonk , web services , wsdm I’ve been meaning to respond to Monkchips speculation over IBM and Amazon from last year this follow-up why Amazon don’t need IBM. James and I met-up briefly before […]

  7. […] back to that one day. What we didn’t get to discuss was Amazon, cloud computing and James on/off theory on IBM and […]

Leave a Reply to Adventures in systems land Cancel reply

Your email address will not be published. Required fields are marked *