% fortune -ae paul murphy

The top five VC proposals I'd like to write

Suppose you had the opportunity to pitch a business opportunity to a high end, Silicon Valley style, venture capital firm. What would you talk about?

Here's my top five list:

  1. a new financial model for massively parallel games.

    The existing financial models for multi-player games, essentially subscription and licensing, were used by time share services suppliers in the 1970s. I know of a better model and I'll bet other people do to, and when someone gets the money in place to bring something new forward, they'll revolutionise gaming and own the market.

  2. a better google - easy and effective search, no misleading ad hits, and no enormous backend data centers.

    Google's front end attracts the eyeballs that power the whole business but its money is going into driving data and applications from your desktop to their data centers. That's the business plan: in their view, as in mine, the PC is a historic hangover - a fundamentally absurd, insecure, and expensive way of doing some pretty simple stuff.

    From a business perspective, however, what they have is a vision that works, but without enough visionaries to support it - and that makes them a sitting duck for the next company to come along with a better search engine, a better business model for advertising, and a better network computing strategy.

    None of that's technically very difficult and sooner or later somebody's going to do it. If it's sooner, Google won't have completed its move to network services, eyeball revenues will dry up, and the money will flow to the next guy, whoever that turns out to be.

  3. QDOS3: an open source, patent conflict free, BSD adaptation to meet Microsoft's PPC network OS needs.

    Microsoft is growing desperate about this: they have several competing internal development projects underway and I believe that they all illustrate the usual iron law of software development: the bigger the team, the worse the product. Now, they may well have another half dozen or more skunk works products laying around, but no one's paying attention because the focus is on the guys on who got the bucks and, well, 70 million lines of Windows code can't all be wrong, but enough of it is to prove the point.

    What Microsoft needs, and what the people at the top have a history of buying, is a solution - and various BSD players have the goods. Put together the right team - a small number of technical people, a lot of lawyers and managers - and pretty soon you too could be meeting with Messrs Gates et al to sell them QDOS3: a billion dollar open source product to power their next ten years.

  4. a simple, easily extended, identity management technology that covers everything from network access control and single sign on to entertainment industry digital rights management.

    The technology for this exists and is reasonably well understood - but the big players are locked into mutual rigor by their own legal commitments and historic antagonisms. The liberty alliance, for example, exists only as a response to Microsoft, American federal DRM legislation treads on real and assumed rights for some to protect real and litigated rights for others. China, meanwhile, plays finders keepers with everyone else's rights and Europe continues to hire new EU directors whose periodic pronouncements on both sides of the DRM issue might as well be publically traded.

    It's swamp no-one's draining despite the plug's being in plain sight -and it's also the number one problem faced in data centers and entertainment studio offices.

  5. Counter-flood: a technology to end spam, while cutting down on phishing and inhibiting the spread of malware.

    Everybody hates spam, malware is pointlessly destructive, and various forms of fraud - from VoIP misdirection to web site spoofing - make the headlines almost every day. So why aren't we doing anything useful about it?

    It's pretty simple: for every internet account there is someone who writes a check: so if an account is used to send spam or proliferate malware: just automatically return it ten times over to the account holder until the misbehaviour stops. How? have internet routers mutually authenticate and have the first one through which things pass onto the internet, write account holder information in every packet stream. Modify mail and network clients to counter-flood on spam delivery.

    It's a bit more complicated, but that's fundamentally it - a billion dollar market win for the first router manufacturer with the guts to do it.

Of course technology business proposals don't have to actually advance technology. For example HP is over ripe for market exploitation. "Just" buy control and split the company three ways:

From a market cap perspective both major spin-offs, HP (printers) and Compaq (x86) would be worth more than HP is today - and that catch-all services group has lots of sole source contracts, meaning that it can be made to look profitable enough in the short term to support a sale or IPO just by laying off more staff and raising prices.

So, anyone have the phone number at Kleiner Perkins Caufield & Byers?


Paul Murphy wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specializing in Unix and Unix-related management issues.