A press release from Qualcomm highlighting how desktop GIS and LBS applications will take advantage of its new 3G/GPS chip, the Gobi, got me thinking about the future of what we now call desktop GIS. That's the typically large software install, with local data and lots of analytical power. While it may tap into data or even services from the Web, its core use is local data crunching and map design. In the release Qualcomm noted how Microsoft with its Streets and Trips navigation software and its MapPoint business mapping software would be taking advantage of the location information coming from the chip. DeLorme's Street Atlas USA, Topo USA and XMap were highlighted for the same reason.

Qualcomm is doing the right thing by highlighting how the 3G/GPS chip will be valuable in upcoming laptops and netbooks, as well as mobile devices. Unfortunately, the market for desktop software as we now know it, like the products noted above, is unlikely to grow substantially. Instead, more of the tasks done on the desktop will move into the cloud and be delivered by one or more services. The same functionality provided by a Streets and Trips could be offered by MapQuest as a service. There's no reason a MapPoint territory analysis could not be delivered via an in-house Web service accessible through a browser. And, how much longer will anyone, including engineers, surveyors and utility technicians, be carrying around a state's worth, or more, of topo sheets? The data and tools to mark them up will be downloaded in small chunks, along with editing tools, before heading into the field or when connectivity is available while out of the office.

The other big change these desktop products will have to face is the pressure to put as many field apps onto handheld devices as possible, be they enhanced GPS devices, PDAs or smartphones. iPhones may have begun their lives as consumer tools, but why not put that power to use for business? The launch of the Blackberry store puts that business device squarely in a position to do business, including tasks related to location. If anything, the desktop apps being touted to use the new Gobi chip need to be rejigged as services for use on the desktop or mobile devices. Hopefully, Microsoft and DeLorme are already well into designing those offerings.

Another desktop product, though not in the geospatial arena, sealed its fate in the past weeks. Microsoft decided to end production of its Encarta encyclopedia product. At launch in 1993 the product was a multimedia encyclopedia on CD, then DVD. At the end of its life, the product was available only as a download. What killed Encarta? If I had to guess I'd point a finger at Wikipedia, not because it's a better resource, but because it's a service, rather than a product. It's constantly updated (for better or worse) and directly linked to source all over the information world called the Internet. As Microsoft noted in the FAQ about the decision: "People today seek and consume information in considerably different ways than in years past." And, I'd add, those ways are more about the network than the desktop.

The nature of the desktop is changing across IT, not just in geospatial. There's no doubt that the number and type of devices that know and can share their locations will only increase in the coming months and years. However, that location information will rarely, if ever, be staying on the machine as a parameter fed to local software. (Qualcomm must know that, otherwise it would not have paired a GPS chip with a 3G chip!) That location will be heading to a service on the Web to turn the raw data into information. Qualcomm, in fact, highlights that by citing a third company that will use its new chip. Absolute's software will use the chip's location information for computer tracking and theft recovery. One of its key features? To report the device's location via the Web for