Free Soup, Open Source and Android

Last weekend I went to an Asian-cuisine restaurant with the whole family. It is one of our favorite casual places in Redmond. ll four of us love Miso soup, and this restaurant happened to have good Miso - not fantastic, but pretty good.

When we arrived, we saw a banner informing us the restarurant was "Under New Management". As we ordered, we learned the new management had decided to make Miso soup free. Wow! everyone thought - free Miso! We all got our plates and started drinking our soup (do you drink miso soup?) while our main meals arrived.

We soon learned making it free was not the only change from the new management.The soup was pretty bad. The new free soup did not include tofu, and it did not include any of the vegetable leafs (excuse my ignorance, is it algae?) or the spices that make Miso a simple yet quite yummy soup. But it was free.

What does this hae to do wth Open Source?

 My guess is that the soup was free because the new management figured that there was more trouble in the process of taking orders and bringin soup to tables, plus adding them to bills than the benefit from the $1. or $1.50 they charged for it. In other words, the restaurant was unable to derive enough economic benefit from the Miso soup to make it an area of investment.

Something very similar happened with IBM Cloudcsape, Sun's J2ME and countless other technologies: when the company can't find a way to make moey from the products, they "altruistically, in a token of support of customers" get rid of the technology contribute the unprofitable great software to the open source community. Many companies think maybe the community can further develop and support the software product while they charge integration and support services to customers who are serious about using it.

The problem is, like Miso soup, without an economic model, usually these products don't get much investment. You can bet they are not getting the A-class talent from the company who made the donation, those brains are working on profitable products. This is one of the problems I see with Android. People ask me how is Android going to disrupt the Windows Mobile model if it is going to be free to device manufacturers.

For starters, the ~$15. per unit a device manufacturer can save is not significant when compared to the millions of dollars it osts to bring a device to market, and the ammount is a rlatively small part of a smartphone's BOM. In the context of a phone that produces $100+ per month in revenue for an operator, which over two years is $2,400 the $15 are pretty insignificant.

My question is, how much do you think Samsung or HTC are going to invest in an unproven platform coming from a company who might or might not have a long-term commitment on a product that no one knows if it will be successful or not. If they are giving it away for free the expectations of continued investments are not very good. Of course, Google, like IBM and other companies before, is not making Android free because they want to win a Nobel Prize or because they want to give back to customers. Like any other company, they are in for the money and they surely have a plan to make money somehow. It is unknown and unproven how and how much can mobile searrch be monetized at the OS level.

back to open source, surely there are a few project that have enough smart and deicated developers behind them to produce pretty good code and to provide some expectation of continued investment. But for each one of these, there are hundreds of half-baked open source project that only contribute fragmentation and confusion to the community. I can see how open source can add value to the industry and to customers in some specific cases, but it is not easy. There are not many profitable open source companies either.

As a reminder, all posts on this bnlog represent my personal opinion which may not be the same of  my employer. Your comments are welcome as always.