F/LOSS User Survey & Firms‘ Open Source Activities

1. F/LOSSUser Survey & Firms‘ Open Source Activities Thorsten Wichmann Berlecon Research GmbH Oranienburger Str. 32, 10117 Berlin tw@berlecon.de

2. Motivation of FLOSS...and outline of presentation • Collect data on importance of and role of OSS in Europe • Assess impact of OSS for policy and decision making Part I of presentation (Final I): • Provides data on OSS use in European enterprises and public institutions on motivations for OSS usage and benefits from using it. Part II of presentation (Final II): • Provides analysis of firms’ OSS development activities and the resulting policy implications.

3. „There are lies, damn lies and statistics.“ • Mark Twain Part I: User Survey

4. Open research issuesRound-up of user survey • What explains differences in OSS usage by country? • Do expected benefits from OSS usage really materialize? Case studies of selected enterprises could help to answer question. • How do establishments evaluate “hybrid” software with OSS components? (e.g. StarOffice, MacOS X, several firewalls and communication server)

5. Who was asked?Methodology I • Firms and public institutions with min. 100 employees in Germany, Sweden, UK • Large number (1452/395 establishments) but few countries since: • Not every company uses OSS • Wanted to distinguish between countries and sectors • Wanted to distinguish between different kinds of OSS • Telephone interview instead of online or print since: • IT decision makers difficult to reach • Minimizes selection problems

6. What did we ask?Methodology II • 1. Current or planned use of OSS • 2. Importance of OSS for IT infrastructure • 3. Questions on OSS assessment in general • General reasons for OSS use • Support of OSS development • 4. Areas of use: (server, desktop, databases, web-sites) • Grouping necessary due to huge number of OSS projects • Many groupings around (different in each publication/site)

7. What did we ask?Methodology III • 5. For each area of use: (server, desktop, databases, web-sites) • How relevant have several motivations been for decision in favour of OSS? • Motivations derived on basis of TCO model, often-mentioned advantages of OSS • Original idea: • Attaching euro figures to to motivations (e.g. savings from stability) • Pre-test showed that nobody could answer such questions. • Decision to pose (more general) questions people could answer.

8. Current and planned OSS usageResults I • OSS used most widespread in Germany • Usage rates higher in public sector than on average (5 out of 6 cells) • Usage rates higher in large than in small establishments (8 out 12 cells)

9. Hitlist of application areas for OSSResults II Germany • Server operating system (30.7% of establishments) • Creating or operating web sites (16.2%) • Databases (15.7%) • Desktop (12.0%) • Same picture in Sweden, although percentages between 3.3% and 9.8% • In UK, OSS usage for databases most important (9.9%), followed by server OS (6.4%) and web sites (6.5%).

10. Most popular OSS softwareResults III Server operating system • 1. Linux (currently 78% of those using OSS SOS), 2. Free/Open BSD (9%) Databases • 1. MySQL (54%), 2. SAP-DB (12%?), 3. PostgreSQL (11%) Web site creation and operation • 1. Apache (72%), 2. Perl (48%), 3. PHP (32%), 4. Squid (26%), 5. OS CMS (10%), 6. Python (8%) Desktops • 1. Linux (49%), 2. StarOffice/OpenOffice (33%), 3. KDE (31%),4. Gnome (23%), 5. Mozilla (23%).

11. Benefits from OSS usage – server OSResults IV Importance of criterion for decision in favour of OSS and against proprietory software

12. Benefits from OSS usageResults V • Results for usage motivations are surprisingly similar across application areas: • Higher stability and better access protection most important • Low or zero license fees come second in importance… then more product features… • Installation and administration cost savings come third • Open and/or modifiable source code only fourth

13. Attitude to OSS in generalResults VI General motivations for OSS usage • For 56% independence from pricing and licensing policies of big software companies is motivation for OSS use. • For 21% better availability of IT specialists is motivation for OSS use. • For 19% OSS use is company policy. Support of OSS development • 36% allow developers to work on OSS projects on company time. • For 35% support of OSS community is motivation for OSS use. • 14% use OSS service companies to support OS community

14. It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest. • Adam Smith, Wealth of Nations Part II: Firms‘ Open Source ActivitiesMotivations and Policy Implications

15. Explanation of OSS development activitiesMany for individuals, few for firms • Motivations for OSS development activities by individuals discussed intensively, e.g. • Signaling effects => show ability to program, be part of team • Learning and training => accumulate human capital • Cooking pot model • Have fun, annoy Microsoft • But motivations by firms are less-well understood • OSS activity often simply explained by complimentary-good argument (give away software, sell server)==> misses spillovers

16. Example IBM: • Around 70 Linux-related projects within Linux Technology Center. • Claims to have invested $1bn in Linux activities. • A few dozen OS projects related to Java, XML, Web Services. Firms‘ Open Source activitiesA quick view at the large software companies • According to their web sites, a third of the world’s 25 largest software companies do engage in major OSS development activities. • 3 out of 25 have smaller projects. • Many of those without projects support Linux with their software.

17. Motivations behind OSS activities • Motivations that explicitly take into account the spillovers from developing OSS: • Standardisation: overcoming the ghost of Linux wars • Open Source software as low-cost component • Strategic considerations • (Enabling compatibility) • Firms can be OS advocates for one software and opponents for others (e.g. Sun)

18. OSS activity as standardisation effortMotivation and policy implications • Standardisation reduces investment risk and can make future path of product more focused. • Participation in activity ensures head-start and helps to influence standard development • Competition policy: OSS development less problematic than other standardisation activities, as OSS is generally open and information exchange takes place in public. • Property rights: “Hijacking” of OSS development possible in theory, but incentive may be lower. ==> Has to be investigated in more detail.

19. OSS activity as basic research Motivation and policy implications • Firms engage in basic research when some returns can be captured. Also results from basic research can best be understood when companies are involved in research process. • Social returns from basic research estimated to be considerably higher than private returns. ==> Argument for policy support • Property rights: Trade-off between tight property-rights to protect research investments and loose property-rights to allow for spillovers.=> Which regime to choose for government-funded projects?=> What are the effects of patents on this type of basic research?

20. Open research issuesRound-up of firms‘ OSS development activities • Do firms’ OSS development engagements pay off? • Can private and social returns of OSS development be quantified? • What are implications of firms’ OS engagements for public support of OSS projects? Which projects should supported? • Is “crowding-out” an issue for these activities? • How do changes in legal regimes influence firm’s OSS development activities?

21. F/LOSSUser Survey & Firms‘ Open Source Activities Thorsten Wichmann Berlecon Research GmbH Oranienburger Str. 32, 10117 Berlin tw@berlecon.de