I just read Stefan’s post on Microsoft buying ESRI in some attempt to “recover” from the blow of Google Earth (linked through James Fee’s Spatially Adjusted blog).
While I don’t have a ton of time to speculate on if Jack and Laura would/will sell to Bill & Co., I will throw this in there – I don’t think Microsoft needs it. Google has “Earth”. Cool. As far as anyone can tell, they have yet to figure out how to make money with it. The only cash Keyhole made was in their sale, and click-through advertising will have limits. Whoo hoo.
I figure it’s a loss leader type of thing that is mainly meant to bolster the Google(tm) brand. The infrastructure to keep “Earth” running is pricy, and the data is orders of magnitude more costly – even buying older archive data. Additionally, the cool factor will wear off – the more people use it, the more they realize the limits of satellite data – it gets stale very quickly. And even if Google had all the money it wanted, there simply is not enough satellite capacity to have the entire globe mapped a relatively high resolution (say 1 meter over cities, 15 meters over rural areas) every year. That said the Google/Microsoft/Yahoo charge into having global coverage will certainly help out the satellite & image processing industries.
On top of this, there are so few things which lend themselves to visualization on a global scale. Yes it would be cool to see some global scale scientific data up on Google Earth, but that’s what NASA’s WorldWind is for (check this out if you have not – very very cool).
So – what then do you use it for? Search for ALL pharmacies? Or maybe you just want to know where Abu Dhabi is (check it – there’s some good Digital Globe data there!) Anyhow….
So Google Earth is very cool. But to Microsoft, the most successful software company in the world, one simple 3D app, with little more than “wow factor” driving it’s use, is not a make or break issue. I would argue that Google Maps is a little different issue – yes it’s about the maps, but more than that, it’s about the Web 2.0/AJAX interaction model. On this front you see lots of Microsoft action – including the Atlas client side framework (more here). This needed to be addressed – partly because they’ve been doing AJAX for quite some time, and party because they needed to allow their developers to create highly interactive sites easily. As for the maps, Virtual Earth is coming along pretty nicely, and from what I hear, the API is also pretty good. And yes, they may suffer a little slower adoption in the “hacker” community for the simple reason that it’s “Microsoft”.
As for ESRI and Microsoft – from a software perspective, it would be a great match – ESRI’s ArcObjects is one of the biggest COM packages out there, and the integration with the rest of the MS product line is relatively good. Another upside could be that SQL Server may get a spatial data type (would someone please do this already!!). But really, I think that ESRI likes their independance too much to sell out. And in the end, what could Jack and Laura do with a few billion that they can’t with a couple hundred million?? And if you’ve seen one of Jack’s keynotes at the annual user conference, you know he is doing exactly what he wants to be doing with his life – so I’d be very suprised to see ESRI up for sale
