The Free Internet, the Promise of Grid Computing

In the beginning, there was the 'Net. Nothing but Net. Gopher was there, and ftp, and other now-ancient protocols, and cruisers of that electronic highway of yore would probably have embraced Mad Max as their avatar.

When I ran an Internet business in the early through mid-90's, even the people with whom I contracted for programming had issues with using the Internet for things as base as commerce. The hue and cry died down about the time people started buying up domains and squatting ahead of companies that were just starting to catch onto this new medium with the potential to 'level the playing field' between the mega-stores and small boutique businesses. One of my investors offered to set up a fund with several of his cronies and have me buy up domains. I refused on moral grounds. Stupid me!

Now I have two firewalls, anti-virus and -spam software, and spend quite a bit of my business life defending against or outthinking people and software 'bots that are attempting to make my life, and that of my charges, impossible. There is a multi-billion dollar industry around Intrusion Detection, Intrusion Prevention, and network authentication, all around keeping the bad guys off our networks. Whether offerings by companies like Mirage Networks will be able to hold the fort without impacting legitimate activities remains to be seen on commercial-scale networks. All this without even touching the real crimes of identity and data theft happening all the time and reported to no one unless discovered. The Russian Mafia is already wise to the kind of 'protection' money already being collected by the Dons of Symantec, Trend, CA and MacAfee. Of course, their marketing methods are a bit more... direct.

Grid computing is at a similar cusp. If you've ever signed up to volunteer your computer to look for aliens, cure cancer, or map the human genome, you've been part of the birth of grid computing. Grid computing has been around for over half a decade in one form or another. In a nutshell, your computer goes to work for someone else when you're not using it.

Like the Internet itself, this simple idea has the ability to become a burden at least as onerous to businesses as security is for the average user, with a questionable return on investment depending on operating burden. Sun and IBM are duking out the back offices of grid computing, while smaller -- and nimbler -- companies such as United Devices are bringing their grid solutions to market.

Up to now we're still mostiy in the "do noble things for noble causes' mode. But to boost this idea to the level of market and profit requires satisfactorily answering the following questions:

1. How can information be kept confidential even if it's on someone else's grid?
2. How can the network be secured from inapropriate activity on the part of the grid it's hosting?
   
3. How can a company make money by hosting someone else's work?
4. How can a host prove to the work provider how much work they did in a measurable way that lets them get paid?
5. How do we match hosts with job opportunities?
6. And how do we do this securely over the Internet?
   

Bluntly, until issues #1 and #2 are solved, I don't think there's any future to grid computing on a serious commercial level. And then the other issues need to be solved to make this a worthwhile market segment. And judging from the reaction from drug companies, they're not thinking security, they're still trying to get their arms around their current IT infrastructure.

There's some good patents out there, if you look around (try here, or here). What we need now is a solid business model, an ROI that makes sense, and a commitment from both OS manufacturers such as Sun and IBM on how they can integrate fiduciarily auditable job performance into their grid offerings. What we have so far are packaging and marketing solutions who benefit the hardware makers, not grid users.