CrowdProcess

This is a great execution of a bad idea. It's the worst kind of bad idea because it seems like a great idea until you try it. It's novel, interesting, timely, and attractive from a technical perspective, but unless your market is tech porn, you're selling to nobody.

This use case is served by countless hosting services, saturated at all levels from cheap VPS's to Amazon clusters to private farms. The market is really efficient here and doesn't afford much overhead. Distributing computation over volunteer users is enormous overhead both for you and your users -- in terms of molding problems to fit the computation model, maintaining the infrastructure, keeping it reliable and performant -- the list is endless. This overhead means you can't compete with the existing players. Forget about being significantly better, which you would need to be to get significant numbers of people to switch.

Quite simply, the math for this doesn't work. Not yet, maybe not ever.

Maybe there's some amazing secret sauce here, but I don't see it. So I offer a warning that a lot of people have tried this idea and failed because they were oblivious to reality.

First of, market. Not there isn't a market for computing power, but a market for this kind of thing. This is a good parlour trick, to show of browser power, but as a computational framework it doesn't work. There are far better and cost sensitive solutions, even using the cheapest AWS service. I know, because I did the math.

Second there is the ethical problem. You are are off shooting the cost of cloud computing to the website client that will pay for the extra electric power generated by your tasks. How do you justify that? More so, how can you justify a ecological minded person paying for mapreduce geological data from oil company?

Third, and the one that kills you, security. You cannot assure the confidentiality of the data being processed. You can not use homomorphic encryption because you do not know the operations you need to do on your data, and since you are working over it, at least in memory it must exist in plain text. Therefor you cannot assure your clients data confidentiality. Worse, this screws up the market fit. The only people that have big data needing to be processed that isn't confidential is research institutions. The thing is, research institutions buy computer power by the stack. I know, I work at one. There is always some grant that allows them to buy a bunch of computers to do parallel processing.

I've heard about you guys for over an year. And it took you more than a year to complete the full product to launch it. What have you been up to? I've seen clones of your product on techcrunch 500. I made a clone of your product using bitcoins, in a weekend. (I did the calculations and wasn't cost effective and I decided not to launch it) Where I work, Técnico Lisboa, we are actually thinking about deploying my clone in my work, over computers we own and provide that computing power to the faculty. Actually I've seen this in 2009, when it was shown of by Ruben Fonseca on Codebits (http://blog.0x82.com/2010/11/22/map-crowd-reduce). If anyone can do this, how can you assure the competitiveness of your product.

The press also took note of Digsby (one of their distribution partners) and the controversy was huge.

They are now out of business.

http://www.crunchbase.com/company/plura-processing

http://lifehacker.com/5336382/digsby-joins-the-dark-side-use...

I'm also reminded of hivemind from defcon this year. https://github.com/seantmalone/HiveMind

In my time as a web developer, probably the most important lesson I've learned is Never Trust the Client.