Interview with Marten Mickos
Marten Mickos is now the head of the Database Group at Sun Microsystems
I have asked Marten a few questions related to the new strategy of MySQL, now part of Sun Microsystems.
See his reply below.
RVZ
Q1. It appears as if the positioning of MySQL has been refocused more predominately on the Web applications / SaaS / ASP market in the last year or so. Would you agree with this, and if so, what does that mean regarding the potential of MySQL to penetrate further into the enterprise?
Marten Mickos:
Great question. We believe that enterprises will move to web-based architectures, and with that wave, MySQL is penetrating the enterprise market.
Goldman Sachs stated in 2006 that “the shift to more web-based applications in the enterprise is unstoppable”. The percentage is still relatively low (10-20% I think) but it is growing.
Q2. Lack of enterprise-grade support and vendor services are frequently cited in surveys as the #1 barrier to the adoption of
FOSS RDBMSs by the enterprise. In this sense, can you give some specific examples of how the Sun acquisition is playing with enterprise clients?
Marten Mickos:
Very true. Before we got acquired we didn’t really believe this (we had so many customers anyhow), but we see a clear change now. Thanks to Sun, we are in active dialogue with CIOs and others of very large corporations.
Q3. One could make the argument that the Big 3 (Oracle, IBM and Microsoft) did not have an appropriately tailored offering (particularly on price) for the build-out of the Web, and that this largely left the field clear for MySQL (as part of the LAMP stack). What do you expect their strategy to be over the long-term?
Marten Mickos:
Yep, I think that’s a valid assertion.
But they are not stupid, and they all have great strategies for the future. Microsoft SQL Server is marching into the web world with the entire Microsoft stack. They are perhaps not overly successful, but within the domain of web apps that run on Windows, SQL Server has a reasonable share. Oracle seems to be attempting to cover the SaaS companies, and they have a reasonably good start there. IBM is focusing on their on-demand story with DB2 on a variety of IBM platforms.
These are just my observations and I may be wrong, of course. Overall I think that the big 3 will continue to have good business for themselves, but I also think that in the most rapidly growing market segments they may have no special advantage.
Q4. What consequences do you think Sun’s acquisition will have for MySQL as an open source product? Can you maintain the user involvement and open source brand? How will you manage the innovation process in the future?
Marten Mickos:
A key reason for us accepting the acquisition offer was that we saw and liked the new open source strategy of Sun. They are fully committed to open source and to the architecture of participation. If anything, this should have a positive effect for us.
Q5: Where Sun wants to brings persistence with respect to objects?
Marten Mickos:
I believe that this is a question for the application designer. As a vendor of software and hardware infrastructure, we at Sun need to accommodate all needs. You get persistence through MySQL or JavaDB (in native Java) or you can use memory-based tools such as Memcached. And with various object-relational mapping technologies you can go from non-persistence to persistence according to your own desires.
Q6: LINQ is leading in database API innovation, providing native language data access. Why is there no LINQ for Java?
Marten Mickos:
I don’t know.
Q7. Sun and Java go together synonymously, ala the change in Sun’s ticker symbol to Java. However, it’s been written that Java users represent a smaller subset of the MySQL community which is largely composed of PHP, Python, Perl, C, C++ developers. How do you plan to increase your appeal to the Java user community?
Marten Mickos:
The P languages are likely to continue to be the most important for MySQL, and Ruby on Rails is growing in popularity. But we always had an initiative to grow the installed base in the Java world. The most important thing we can do is have a great JDBC driver, which I think we have. It is highly performant and it supports the most important functions and constructs. Now as we are part of Sun we will be able to further remove barriers to adoption. We will probably create more how-to documents, tutorials, and sample applications, plus benchmarks etc. – all of which are intended to make it more appealing to use MySQL from a Java app.
Q8. As a company, you need to derive revenue to survive. You’ve done that successfully using an open source model that focuses on services and value added upgrade licenses. Those tools, while establishing a vast user community including 100’s of millions of installations, have driven relatively modest revenues. Going forward, which revenue generating tool do you see providing the most return to Sun’s investment in MySQL, services or licenses and why?
Marten Mickos:
We have always had a business model of providing commercially licensed stuff to customers. This will continue. From a technical standpoint we know that open source is a more efficient way to produce software, but from a business standpoint we have chosen to produce certain smaill add-ons for paying customers only. In this way we can combine the best of open source with a great revenue model. This is probably why we are the fastest growing database business in the world.
///mgm
