Total Pageviews

Friday, December 28, 2012

Dreams of \'Open\' Everything

Software is not merely about automating every aspect of our lives anymore. Some of its makers want to change the way we all interact, spreading their supposed egalitarian excellence.

Whether this is liberation into a new and better mode of being (and yes, the people thinking about this take it to that scale) or the folly of an industry in love with its success is one of the more intriguing questions of a world rushing to live online.

GitHub is a San Francisco company that started in 2008 as a way for open-source software writers in disparate locations to rapidly create new and better versions of their work. Work is stored, shared and discussed, based on the idea of a “pull request,” which is a suggestion to the group for some accretive element, like several lines of code, to be “pulled,” or added, to a project.

“The concept is based around change: what is the right thing to do, what is the wrong thing?” said Tom Preston-Werner, GitHub's co-fo under and chief executive. “The efficiency of large groups working together is very low in large enterprises. We want to change that.”

Mr. Preston-Werner's own company is something of a proxy for how he sees the world. GitHub has no managers among its 140 employees, for example. “Everyone has management interests,” he said. “People can work on things that are interesting to them. Companies should exist to optimize happiness, not money. Profits follow.” He does, however, retain his own title and decides things like salaries.

In his blog Mr. Preston-Werner has written about how important it is for  companies to expose as much of their inner workings as possible. Another member of GitHub has posted a talk that stresses how companies flourish when people want to work on certain things, not b ecause they are told to.

This style and sentiment echoes those at another company, Asana, a corporate social network aiming to improve the pace of work. Founded by Dustin Moskovitz, who was a co-founder of Facebook and for many years ran its technology, Asana bases work on a series of to-do lists that people assign one another. Inside Asana there are no formal titles, though like GitHub there are bosses at the top who make final decisions.

For all the happiness and sharing, real money is involved here. In July GitHub received $100 million from the venture capital firm Andreessen Horowitz. This early in most software companies' lives, $20 million would be a fortune.

Companies pay to use GitHub, and it has become an exceptionally popular way for people to do all kinds of software work; in 2012 its number of users jumped t o 2.8 million from 1.2 million. The number of “repositories” - containing code, its documentation, images associated with a project or other work - increased to 4.6 million from 1.7 million last year.

Many of these are open-source projects, and GitHub does not break out their revenue.

GitHub's popularity has also made it an important way for companies to recruit engineers, because some of the best people in the business are showing their work or dissecting the work of others inside some of the public pull requests. Its founders and backers, however, want to use the GitHub model to make mobile and enterprise software applications, and possibly much more.

Mr. Preston-Werner thinks the way open source requires a high degree of trust and collaboration among relative equals (plus a few high-level managers who define the scope of a job and make final decisions) can be extended more broadly, even into government.

“For now this is about code, but we can make the burden of decision-making into an opportunity,” he said. “It would be useful if you could capture the process of decision-making, and see who suggested the decisions that created a law or a bill.”

Can this really be extended across a large, complex organization, however?

As complex as an open-source project may be, it is also based on a single, well-defined outcome, and an engineering task that is generally free of concepts like fairness and justice, about which people can debate endlessly. Even on a less lofty plane, companies like GitHub and Asana will ultimately test themselves against complex corporate processes lasting years, and involving skills in both science and the humanities. Google once prided itself on few managers and fast action, but has found that getting big can also involve lots more meetings.

Still, these fast-rising successes may be on to something more than simply universalizing the means of their own good fortune. An ea rly guru of the Information Age, Peter Drucker, wrote often in the latter part of his career of the need for managers to define tasks, and for workers to seek fulfillment before profits.