Continuent is probably best known for its database clustering technology for MySQL, as well as PostgreSQL, but the company has for some time had its sights set on expanding beyond open source databases and enabling horizontal database scalability.

It has just taken a major step towards delivering on both counts with the launch of Tungsten, its new stack of open source middleware technologies designed to enable low-cost databases to scale horizontally for database failover and continuity.

Tungsten includes includes Sequoia, the existing synchronous multi-master replication project, as well as Tungsten Replicator for asynchronous master/slave replication, Tungsten Manager for cluster-aware service management over LAN or WAN, Tungsten Connector (previously known as Myosotis) for native wire protocol proxying, Bristlecone for scale-out performance testing and Hedera for group communications.

Tungsten brings Continuent technology to Oracle environments for the first time, although as CTO Robert Hodges notes on his blog, a more significant immediate target will be addressing some of the previously mentioned MySQL replication issues presented by Monty Widenius last April.

The Tungsten stack is designed to let users mix and match the components to fit their specific replication requirements, and Continuent will be doing that itself as it delivers commercial products based on the open source code and commercial extensions next year.

I?ve spent the past three weeks profiling open source policies and adoption projects at the 16 nations competing in EURO 2008. Congratulations are due to Spain, which deservedly won the football championship on Sunday with a 1-0 win over Germany.

Just for fun I thought I?d also declare a 2008 Tour of Europe Open Source Champion. In deciding the winner I decided to follow the same organizational structure as the football, so read on to find out which eight nations made it out of the group stages and how I whittled it down to an eventual champion.

If you disagree with any of my decisions feel free to add a comment explaining why, but remember: the referee?s decision is final. Although the football has finished, I?ll also be continuing the tour to take in some of the nations that did not qualify. So after a short break while I get my breath back, look out for further profiles in the weeks to come, including the UK, Ireland, Finland, Denmark, Norway, Belgium and Estonia.

Group stages
Group A
It is a close call between three of the four nations in Group A, with Turkey bringing up the rear. However, with sporadic deployment projects already underway, good grassroots interest and government support for the Pardus project, Turkey is catching up and will pose a greater challenge in a few years. Switzerland, Czech Republic and Portugal are all more advanced with policies and/or projects in place. However, despite early successes the Czech Republic?s flow of open source projects does seem to have run dry and the recent deal with Microsoft is not going to help.
Portugal goes through as group winner with Switzerland runner up.

Group B
There was only ever going to be one winner from Group B but separating the other nations was not so easy. Poland shows a lot of promise but relatively few projects and a watered-down policy favouring open standards rather than open source. Austria and Croatia are hard to compare. Austria has established projects, but Croatia has the forward-looking policies. Ultimately though Vienna is more progressive in terms of open source adoption than Austria as a whole, and even that project has stumbled.
Germany goes through as group winner with Croatia runner up.

Group C
As was the case with the football, Group C saw three European heavyweights drawn together. Romania was always likely to suffer by comparison despite a lot of grassroots interest and emerging projects. Meanwhile Italy has been more progressive over the years but as Roberto Galoppini recently reported ?is still missing a clear strategy about how to foster the Italian open source ecosystem through training, education, research and outreach?.
France goes through as group winner with The Netherlands runner up.

Group D
It also a close affair in Group D, at least between the top two. Greece was the rank outsider thanks to its private partnership agreement with Microsoft, and while Russia shows a lot of promise it is definitely one to watch, rather than a current challenger.
Spain goes through as group winner with Sweden runner up.

Quarter finals
Portugal versus Croatia
A close one to call given Croatia?s commitment to open source as a means of reducing its dependency on suppliers, promote interoperability and match the EC?s position on open source. However, Portugal has the advantage of maturity enabling more projects to have got up and running. Croatia bows out with heads held high.
Portugal wins.

Germany versus Switzerland
A David versus Goliath match-up to be honest. Switzerland has one of the most forward-thinking open source software strategies and a number of significant national and regional projects, but nothing near the adoption levels seen in Germany.
Germany wins.

France versus Sweden
Sweden can boast Programverket, a project to help the public sector adopt or convert to open source software, as well as a number of deployment projects, and of course MySQL, but ultimately pales in comparison to France.
France wins.

Spain versus The Netherlands
A very close one to call, but ultimately while The Netherlands has seen a number of national and regional projects based on a policy of adoption of open standards in general and positive discrimination in favour of open source software, it cannot match up to the commitment to open source as an economic enabler seen in Spain. S
pain wins.

Semi finals
Portugal versus Germany
It could be said that the draw favoured Portugal somewhat in that it avoided some of the more significant adopters, but it more than meets its match in Germany. It would be interesting to see how different things might have been had a Portuguese bill that would have mandated the use of open source software in public administrations been approved in 2003, but that?s all ifs and buts. As it is the country?s significant adoption projects are focused on the education sector.
Germany wins.

France versus Spain
A very, very difficult one to call. However the political structure in Spain eventually swings the decision in France?s favour. Despite significant adoption across its autonomous regions what Spain lacks, compared to France, is a cohesive national strategy driving adoption. For example, there are at least eight different Linux distribution projects funded by local governments. Recent policy changes suggest that the national government is catching up and if Spain can link together some of its regional projects and replicate Andalusia?s open source repository to reduce duplication of effort, it could be a force to be reckoned with.
France wins.

The Final
Germany versus France
It?s the final that most open source industry watchers would probably have predicted, I guess. Germany and France have been the two heavyweights of open source adoption and promotion in Europe to date. In order to separate the two it?s necessary to take a closer look at the information we?ve been able to gather according to the categories of policy, national and regional projects, and vendor ecosystem.

Key policies:
Germany has been publicly committed to open source promotion since 2000 and in 2001 the Bundestag passed a resolution in to promote open source software in the federal administration, based on the principle that open source is a special opportunity for the European software industry. Several studies and guideline documents have followed, along with a key arrangement with IBM for discounts on Linux systems.

France considered but declined legal proposals to enforce/encourage the use of open source software in public administrations in 1999 and 2000. In 2001 ATICA took matters into its own hands with a decision to promote open source, which was followed by guidelines, recommendations, and practical guides. Recently an economic commission set up by France?s president Nicolas Sarkozy recommended that France should increase its use of open source software and consider tax benefits to stimulate open source development.

Score: Too close to call ? 1-1

National deployment projects:
German federal projects include Federal the Federal Finance Office, the Bundestag, the German Aerospace Centre, the Foreign Office, Deutsche Bahn, the Employers? Liability Insurance Association, the Monopolies Commission, German air traffic control and the Federal Institute for Geosciences and Natural Resources.

French national projects include the Ministry of Equipment and Transport, the Ministry of Defense, the Family Allowance Agency, the Ministry of Foreign Affairs, the Ministry of Agriculture and Fisheries, the Tax Ministry, the Directorate General for the Modernisation of the State, the Ministry of Education, the Culture and Communication Ministry, the gendarmerie, and the National Assembly.

Score: Close again but 2-1 to France based on depth as well as breadth of adoption.

Regional deployment projects:
Munich is probably the most famous Linux deployment project in the world but Germany also boasts projects in Hall, Mannheim, North Rhine Westphalia, Lower Saxony, Heidenheim, Berlin, Treuchtlingen, Osterburg, Stuttgart, Frisia, Friesland, Freiburg, Nordrhein-Westfalen, and the German Alliance of Cities and Communes.

France also has a number of regional projects including Arles, Grand Nancy, Lille, Val d?Oise, Marseille, Brest, Grenoble, Lyon, and Rennes.

Score: Germany draws level ? 2-2

Key vendors:
Germany?s tradition as an open source adopter is no doubt due in part to its championing of SUSE Linux AG, now of course owned by Novell. Other German open source vendors, such as Open-Xchange and Collax, have also migrated to the US. While Credativ and Synerpy remain, Mindquarry came and went.

France has done a better job of keeping hold of its vendors, which include Mandriva, Ulteo, Nuxeo, Talend, and Linagora, as well as the OW2 Consortium.

Score: In the context of this study, the vendors remaining in Europe has to be a plus point. 3-2 to France.

Result:
France wins and is the 2008 Tour of Europe Open Source Champion.
Germany runner up.

To coincide with EURO 2008, I’m embarking on a virtual European tour, taking a quick look at open source policies and deployment projects in the 16 nations that are competing in the tournament.

Sweden crashed out of EURO 2008 last night a Russia qualified for the knockout stages with a well-deserved 2-0 win. As home to MySQL Sweden might be expected to be one of the more progressive adopters of open source but while there is significant interest, details of deployment projects are relatively hard to find.

Key policies:
The Swedish Agency for Public Management?s 2003 feasibility study identified a number of existing projects (see below) and recommended that open source software should be judged on a par with proprietary software in a procurement process.

In 2005 an association of county councils, municipalities, and private healthcare providers known as Carelink published a report (PDF) on Sweden?s legal framework for open source software which stated that ?wider distribution of open source software that has been developed internally by health and social services has the potential to enhance the effectiveness of IT in health services and would achieve faster diffusion of common solutions in health care.?

Also in 2005 the Swedish Association of Local Authorities and Regions created Programverket, a project to help the public sector adopt or convert to open source software. Programverket is based on the principle that ?software development financed by tax money should in principle provide open source software, unless there are specific reasons to the contrary.? Further details on the projects can be found here.

Key projects:
Examples included in the Swedish Agency for Public Management?s 2003 feasibility study included the Meteorological and Hydrological Institute, the Premium Pension Authority, the National Labour Market Administration, and Malmo General Hospital.

Since then a number of further projects have been added to the list. Not surprisingly MySQL has had some success in its own back yard, including a project with the Swedish National Police, which saved the equivalent of 400 fully-equipped police cars on moving to an infrastructure based on Linux, JBoss, and MySQL.

Other state projects include Apoteket Sweden?s largest state-owned pharmacy chain, which migrated to Red Hat Enterprise Linux at 900 pharmacies as part of a move to Intel hardware, and the Swedish Armed Forces? deal with Red Hat to enable migrations from Windows NT to Red Hat Enterprise Linux.

Key vendors:
MySQL is the leading Swedish open source vendor, or at least it was until the Sun acquisition, but Sun?s new database business unit is keen to stay true to its roots. The country also boasts a host of open source services firms and consultancies, as can be seen from the member list of Open Source Sweden.

And another thing:

As always we welcome your input. If you have examples of open source adoption in Sweden that we’ve overlooked, please leave a comment below. For more stops on the European tour, see this post.

Switzerland is co-hosting EURO 2008 along with Austria and will be kicking off the tournament with a game against the Czech Republic on Saturday. The country is of course famous for its neutrality but has shown itself to be less than neutral when it comes to open source (see what I did there) with the federal government having adopted an open source software strategy as long ago as February 2004.

The policy had three core priorities: equal consideration for open source and proprietary software packages; the sharing of software among federal agencies; and the implementation of pre-requisites for open source adoption (such as standardization, technical support, training, assessment and legal considerations). Given the latter it was probably an example of one of the most forward-thinking open source software strategies.

More recently it has been suggested that the policy should be extended even further such that the use of proprietary software would need to be justified if open source alternatives are available.

Key projects:
The Swiss government has also shown that it is prepared to put the strategy into practice, with the Swiss Federal Office of Construction and Logistics issuing a tender for a decision-support software system in February 2005 that met the open source strategy.

Later that year Novell revealed an agreement with the government of Switzerland to migrate 3,000 of its servers to SUSE Linux. Local administrations have also followed suit with Zurich deciding in 2005 that open source and proprietary software should be treated as equivalents but that open source would be the preferred option if the project was reasonable.

More recently it was revealed that the Swiss canton (region) of Solothurn would complete the migration of 2,000 desktops to Debian by the end of the year, while the desktops used by 70,000 students and 7,000 teachers in Geneva are in the process of moving to Ubuntu and OpenOffice.org.

The Optaros effect:
Switzerland has certainly proved to be a happy hunting ground for open source consulting firm Optaros, thanks to its offices in both Geneva and Zurich. The company?s case studies include the Federal Supreme Court?s use of Alfresco to control its repository of legal documents; the State of Vaud?s e-government system (also based on Alfresco) and workplace condition tracking application (SugarCRM, MySQL, Apache, JasperSoft); Swisscom Hospitality Services? broadband Internet service; and Swisscom Mobile?s Labs customer engagement site.

And another thing:
The flag of Switzerland is one of only two square sovereign-state flags, the other being the flag of the Vatican City. (Facts like this are what Wikipedia was invented for.)

As always we welcome your input. If you have examples of open source adoption in Switzerland that we’ve overlooked, please leave a comment below. For more stops on the European tour, see this post.

There was some interesting discussion following my post last week asking whether there is a growing rift between commercial open source software vendors and some aspects of the open source user community.

Amongst the comments, Chris Marino of SnapLogic suggested that some of the tension might be eased by open source software vendors being more upfront about their intentions via the publication of social contracts. Examples include the Debian Social Contract and also Funambol’s Open Source Project Social Contract.

As Chris noted, the problems come when companies start changing how they interact with open source communities. Nevertheless, this sort of contract at least lays out the ground rules so that users and developers know what sort of vendor they are involved with and have something definitive to refer to when disagreements arise.

If vendors are not proactive about writing and publishing declarative statements about their open source involvement, they may find that community users start judging them on the community’s terms. From Milking the GNU comes the suggestion that a new independent organization could be formed to judge vendors that claim to be open source on a number of criteria, such as patent policy, business model and development model.

“Equitable Open Source” as it is called, is only a suggestion at this stage, but is an example of the caution being expressed by some users towards commercial open source vendors. As the idea is described, it would at least create a level playing field upon which vendors can be judged.

This would help to avoid debates like this one, in which Baron Schwartz argues that MySQL is free software but not open source. His point is that the development model is not open.

While respondents have pointed out that, according to the Open Source Definition, MySQL is in fact open source, the fact is that the OSD only defines the license used to distribute the software, not the method used to develop it.

Simon Phipps has recently suggested adding open source patent and trademark definitions to the Open Source (Copyright) Definition. Does the industry need an open source definition for business and development models?

“Sun to Begin Close Sourcing MySQL” screamed the headline on Slashdot last night. The headline is not entirely accurate (although slightly more accurate than the bizarre statement that “Sun has had a very poor history of actually open sourcing anything”).

So what is going on at MySQL? To get to the bottom of that you have to weave together a number of posts and comments from a number of sources. First the article behind the Slashdot headline:

“Just announced: MySQL to launch new features only in MySQL Enterprise,” states Jeremy Cole, which is a much more accurate description of the state of affairs. “MySQL will start offering some features (specifically ones related to online backups) only in MySQL Enterprise. This represents a substantive change to their development model ? previously they have been developing features in both MySQL Community and MySQL Enterprise.”

Marten Mickos confirmed Jeremy’s post in the comments section, stating: “In 6.0 there will be native backup functionality in the server available for anyone and all (Community, Enterprise) under GPL. Additionally we will develop high-end add-ons (such as encryption, native storage engine-specific drivers) that we will deliver to customers in the MySQL Enterprise product only. We have not yet decided under what licence we will release those add-ons (GPL, some other FOSS licence, and/or commercial).”

So to clarify. Sun (or MySQL) is not going to begin closing the source code of MySQL features, but it is going to introduce new features into the Enterprise Edition that will not be available under an open source license.

To some extent there is nothing new here. The company previously announced that the Standard Edition of the recently introduced MySQL Workbench would include functionality not available in the open source Community Edition, while the MySQL Enterprise Fall 2007 release saw the availability of replication monitoring and advisory functionality only available with the Enterprise subscription.

Before that the company introduced Network Monitoring and Advisory Services with the Enterprise version in October 2006. Additionally, MySQL removed the Enterprise tarballs from its community ftp site in August 2007.

Mickos also responded to the Slashdot post itself; pointing out that “the business decision on this was made by MySQL AB (by me as the then CEO) prior to the acquisition by Sun, so this has nothing to do with Sun” and that “everything we have released under GPL continues to be under GPL”.

In a later comment he added: “If the world were perfect, we would only produce GPL code and we would have a great business that can fund the software development. But we have found that the world is not perfect. We have been experimenting with a variety of business models around FOSS (dual licensing, support only, simple subscriptions, different binaries for community and enterprise, non-open source features) to find the best one. And we will continue to experiment until we are satisfied. We need to find a model that allows us to produce a ton of great code under GPL while having the financial strength to do all this.”

I was reminded of an article Mickos himself wrote in 2006 about the 13 different business models used by open source vendors (the original article appears to have vanished but you can see my response to it and a list of the business models here).

From this list it is clear to see how MySQL is in the process of moving from “3. Software is free but if you embed it in closed source, you better pay a fee (Trolltech, DB4Objects, Funambol, MySQL, etc.)” to “6. Software is free but some enterprise features are not (SugarCRM, Zimbra, JasperSoft)”.

Of course, whether you believe this to be the correct model for MySQL is another matter, and Matt Asay for one would prefer to see MySQL opting for “5. Software is free but on-going maintenance, monitoring and provision of binaries is not (Red Hat)”, which is the direction the company had appeared to be going in.

The fear, as far as the community users is concerned, is that MySQL might end up using “7. Software is free but we built a closed-source product around it (EnterpriseDB, GreenPlum)”. However, given Sun’s business model is “9. Software is free but we sell everything else on the planet, including closed source software (IBM)” there appears to be little chance of that.

As Matt also notes, the problem MySQL has right now is a public relations problem (or a community relations problem). Given that there does not appear to be a vast change in strategy (despite the headlines this is a bend in the road rather than an about turn), this should not pose a long-term problem for the company.

Zack Urlocker has also clarified the situation here.