The metamorphosis of the software industry

One of the obvious changes after the crazy dotcom bubble-burst is that IT buyers are much more educated and demanding. They are done signing mega contracts. Instead, they would like to pay as they realize the value of what they are buying and as they see a clear return on investment. This has created a significant move away from monolithic multimillion-dollar software contracts and increased the popularity of pay-as-you-use licensing. Some fast movers like Salesforce.com have capitalized on these new business models with great success cannibalizing the business of giants like Siebel.

The consequence of these aggressive business models and the rising success of open source is that traditional software vendors (such as SAP and Oracle) started applying steep discounts to their software licenses to get new business. During the antitrust trial in 2004, an Oracle executive admitted that they were prepared to cut prices by 70%. The way I read this personally is that those giants realized that the real money is in the recurrent revenue produced by maintenance after the customer is locked in. CIOs often report publicly as much as 70 to 80 percent of their IT budgets are consumed by maintenance. During an earnings call in 2004, Oracle CEO Larry Ellison touted the company's maintenance (which includes fees derived from product updates and support) as an "extremely high-margin business." An article published by CNET shows how maintenance revenue manifestly exceeds (and keeps growing) license revenue for large software vendors.

By now I think I have made the point that the dynamics of the software business have changed and that maintenance, once a boring and unexciting, is where the money is coming from. What does this mean for open source? I think it’s very exciting news!

This simply means that many companies that provide high quality services around open source projects or bundles could potentially become billion dollar companies. Recently, we’ve seen several of those companies getting serious venture money. I was getting bored with Linux (Novell and Red Hat) being isolated open source successes. Seeing VCs inject money into those companies is very encouraging for open-source supporters who put their careers on the line to get their companies to adopt open source platforms. What businesses want is one neck to choke, one vendor to go to for its support needs so that they can focus on what they do best instead of wasting resources maintaining code they never wrote.

Do you think that one of the effects of open source is that software will never be sold the same way again? How many multi billion open source service providers will we see in the next decade? Do you think Orbeon, Optaros, SourceLabs or Spikesource have a shot at being as successful as Red Hat?