What’s next: A butler for your idle computer?

Idle grid computing — applying unused computer time to a task — is a great solution in need of a problem. Hundreds of millions of computers sit unused two-thirds of the day. Most of these PCs have the CPU, storage, and internet access to run complex tasks, but instead just suck power.

Dozens of free programs let people use this time for worthy causes: Seti@Home searches for ET, Folding@Home analyzes proteins, neuGrid studies brain diseases. Most are non-profits that just offer good karma.

But there are some untapped startup opportunities to use this computing power. Let’s look at one idea: a program that downloads interesting media when your computer is idle.

I’ll call this hypothetical startup PushMedia. What it would do is find and download rich content that’s interesting to you. This may seem like a solved problem with so many sites like Digg and StumbleUpon already aggregating content, but most have two problems: They require that you choose what content you want, and you have to download it in real-time. For larger files like songs, movies, TV shows, books, and applications, “pulling” this media is slow. When internet access isn’t available, it can’t be pulled at all.

PointCast tried pushing content in the dot-com boom and famously flamed out after refusing a $450 million buyout. But this may be an idea worth dusting off; immediate gratification is always a good bet.

PushMedia would be an installed program that downloads media from a wide variety of sources. You could specify sites or categories you like and give optional demographic data for targeting. When your computer is idle, PushMedia would scour the web and “push” the files to a specified folder. You could specify when this downloading happens or let it run in the background.

PushMedia differs from most grid computing projects in key ways: it solves a common problem, it provides direct benefits to users, and most importantly, it can generate revenue. PushMedia could monetize through advertising, premium subscriptions, and market research.

Media companies could pay-per-download for content they want to advertise. While many companies are fighting content pirates, they still want awareness for new, lesser-known, or highly-monetizable titles. When CBS launches a new show, U2 releases a new album, or Stephen King publishes a new book, pushing a teaser episode or chapter to millions of targeted users, knowing what’s been viewed and shared, and tracking conversions to purchases is pretty valuable. File-sharing is expected to grow to a $28 billion market by 2011.

For consumers, the base program would be free, but users could pay for premium features like early access to popular content, advanced search and filters, and faster downloads. File-sharing companies like Rapidshare, Kazaa, and eMule have successfully upsold users on similar features. Media research firms like Nielsen and Arbitron might also pay for daily reports to learn what people are consuming, especially if users give demographics.

Unlike other grid computing projects that need scale to be useful, PushMedia’s basic features are useful with just one computer. Ten years ago, when PointCast made its play, it was harder to learn what was popular, but now the program could easily scour content ratings and view counts to get an early read on what’s hot.

The real challenge is getting a lot of installs. StumbleUpon greatly benefited by being one of the first Firefox add-ons; founder Garrett Camp has said he would not start StumbleUpon today because of user aversion to installs. PushMedia could face the same pitfalls as Joost; because users don’t have to connect with others for basic features, it’s hard to grow like inherently viral products such as Skype.

Possible ways to overcome this include partnerships with media creators and aggregators, paying for initial installs, and social features that encourage sharing and chat. A radical option is to build PushMedia as a website that automatically downloads content when users visit, but they likely wouldn’t do that consistently. Like most web ideas, PushMedia’s key challenge is distribution.

While few companies are taking PushMedia’s tack, it would compete indirectly with a crowd: user-generated aggregators like Digg and StumbleUpon, media outlets like Hulu and Joost, and bookmarking sites like Delicious. PushMedia’s competitive advantages would be to eliminate the user’s need to browse and to provide immediate consumption of large files. Unless these companies build a competing client, they cannot match that.

The technology to build PushMedia is readily available. BOINC (Berkeley Open Infrastructure for Network Computing) is a free, open-source platform for grid computing projects; others are also available. If PushMedia decided to scour copyrighted content, it’d risk the wrath of billion dollar lawsuits, but plenty of interesting content is now available to legally sample or download. Users bear the bandwidth and storage costs, so the marginal costs to run the company are minimal.

Overall, the company’s success likely depends on its marketing and ability to provide good suggestions. What do you think?

Do you think PushMedia is a good startup idea?(myspace polls)

If you liked this post, be sure to check out last week’s startup idea here.

Mark Goldenson advises entrepreneurs and is launching an innovative web venture in health care. To submit an idea for the What’s Next series, email Mark at mjgold3@gmail.com. Selected ideas will receive attribution.