It seems there's a fundamental problem for many state governments that want to acquire and use free software: they can't buy it. Literally. That's because the state software acquisition process doesn't know what to do with software that nobody owns or sells. As political consultant Tom Adelstein puts it, "Governments require bidding. Since there's no Apache, Inc., Apache doesn't get on the bid list." Nor do Linux and other forms of free and open-source noncommercial software--no matter how useful they may be and how much money they save.

This is the situation being addressed by Texas Legislature Bill SB 1579, sponsored by Sen. John Corona and introduced on March 17. It follows Oregon House Bill 2892, sponsored by Rep. Phil Barnhart and introduced on March 5. Unlike the Digital Software Security Act initiative in California, neither of these bills mandates use of free or open-source software. Instead they simply open those forms of software to consideration in the bidding process. In other words, they prevent exclusion of non-commercial forms of software.

Cost is an issue as well. The Texas bill addresses "determining best value for purchases," defined by the "lowest overall costs". In addition to price, those include compatibility, upgradability, reliability, training, technical support, standards compliance and other factors. Basing acquisition (and not only purchase) on these considerations should reduce far more than Texas' $1.5 billion annual computing expenses.

Or, as Senator Corona puts it, "Linux could provide significant savings through better exploitation of existing hardware, the ability to consolidate servers and substantially lower licensing costs."

These legislative moves come at a time when corporate aversion to Linux and open source is waning fast. It only makes sense that private and public software acquisition reforms would follow parallel paths.

Linux Journal is taking a special interest in these developments, because we know far more is going on than commercial vendors can ever tell us. IBM, HP, Oracle and other companies are good at feeding us all kinds of Linux success stories. But they simply are not in a position to report on highly useful and popular stuff with costs that round to zero but benefits run in the millions or even billions of dollars. We're sure the same is true in government, where even commercial vendors often don't have much to say.

Right now, we're studying how smart companies make themselves smarter using Linux. It would be interesting to include governments in that same study. If you have any interesting accounts to report, write to me.

Doc Searls is Senior Editor of Linux Journal.