If I read the runes correctly, the hopes of many of the committed followers and proponents of Open Source have been disappointed by its progress in recent times. Admittedly, the attempt by SCO to stop Linux in its tracks (or get a royalty from its use) looks to have failed and last year saw Microsoft almost accommodating Open Source in a deal with Novell (but not without spreading a little FUD).

However, it also saw Oracle train its big guns on Red Hat, threatening to spoil its whole business model simply by competing directly—and this was not long after Oracle had embarrassed MySQL by buying up the Open Source providers of two of the database’s key complementary components. On the desktop and laptop it is not Linux but Apple that is providing genuine competition for Microsoft and in other areas, such a mobile devices, it looks like Linux is not really a significant player.

What I think is happening is neither the defeat of Open Source, nor its saturation. I think we have become too accustomed to treating Open Source products as if they were commercial products and thus trying to judge their progress in terms of growth and market share. I don’t think you can look at it like that.

A good deal of Open Source adoption doesn’t occur in the same way as the purchase of commercial product. I was at an IT site recently that suddenly decided to use Hibernate. Why? Because some contractor on the project had used it. The product adoption process must have taken a whole ten minutes (about the time it takes to download). But nobody would adopt Open Office in the same way. It would be a major decision and a staff training issue would need to be addressed.

I still keep seeing signs that some Open Source products are likely to increase market share. Microsoft’s introduction of IE7 seems to have had no effect on the population of Firefox users (despite one or two analysts suggesting that users would switch back). The Firefox user base continues to grow (and so does that of Apple’s Safari and Opera, by very small amounts). IE7 is simply replacing IE6. The adoption of Open Office may be slow, but as far as I can tell it is increasing (and so is the user base of Apple’s iWork).

Microsoft clearly doesn’t have a full lock hold on the office market. It will be interesting to see what effect the new version of Office has. Microsoft has introduced a new file format for the first time in 10 years and a new interface. Although the product is clearly improved, I expect it to have a very slow take-up. I also expect that it will cause some defections to Open Office. (When there’s going to be a learning curve anyway, why not go for the zero cost product?) How many defections? Significant numbers but not vast is my expectation.

Another interesting Open Source play comes from EnterpriseDB which has embarrassed Oracle by acquiring some high profile defections from Oracle (Sony and Vonage). If you didn’t know EnterpriseDB is a version of PostgreSQL which has a full PL/SQL capability. (Ninety percent of Oracle applications work with no changes). EnterpriseDB is not zero cost but it cuts the Oracle cost to ribbons. If I were Red Hat looking for a way to defend myself against Oracle I’d buy, merge with or strongly partner with EnterpriseDB. I expect EnterpriseDB to succeed—in time it will challenge SQL Server and DB2, if it continues to have success against Oracle.

Another factor in the Open Source equation is the OLPC (the $100 laptop or OneLaptopPerChild) initiative. It hasn’t taken off dramatically yet, but it has clearly gained enough traction for us to know that it will persist. In time, this initiative will boost the Linux laptop market in a dramatic way. It is quite likely that the OLPC initiative will lead to the manufacture of 10 million laptops per year for use in developing countries. This constitutes about 10 percent of the global laptop market.

So in summary, the Open Source story is, like the curate’s egg, good in parts. Linux is still a growing force and the commercial use of Open Source is becoming increasingly common in many areas. There was even a recommendation to use it from the EC recently, which stated quite clearly that the evidence is that Open Source reduces costs. Well, duh!