Thanks to a terrific community that nominated Talend Open Studio, I am proud to inform you that Talend has been nominated a finalist in the SourceForge.net Community Choice Awards, in the category “Most Likely to be the Next $1 billion Acquisition”.

This is a great recognition of the success of Talend and how we have clearly become the leader in the open source data integration market, and a model for many other players in this field.

Now we need the community more than ever. PLEASE VOTE!!! Being a finalist is a great recognition but we don’t want to stop here, we want to be the winner.

The process is simple:

Visit this URL: http://sourceforge.net/awards/cca/?group_id=181308. You will need to authenticate into your SourceForge.net account (if you are not a member already, you will be prompted to create an account - it’s a very straightforward process and of course it’s free). Then pick Talend Open Studio as your vote for the category “Most Likely to be the Next $1 billion Acquisition”. You may pick other winners for other categories if you wish, but this is optional - you can just pick “I do not wish to vote” (yeah, this is a bit of a pain… but it’s worth it!)

Don’t wait - voting is only open for a few weeks.

Thanks again to all the community for your support.

Yves

PS: kudos to SourceForge.net, a fellow founding member of the OSA, for putting together these great awards!

This press release from IDC caught my eye. They published a report yesterday demonstrating the rising importance of open source among business end users, and listed the main challenges they face. In particular:

Respondents believe the biggest challenge vendors face in delivering open source-related services is integrating open source and proprietary software components.

Late last year, the OSA published its customer forum series summary report, which highlighted the same issue. As commercial open source becomes more mainstream, issues like interoperability come to the forefront. The OSA has been on the leading edge of identifying and resolving these issues, while other popular thought leaders and bloggers were asserting they didn't exist. So, it's good to see mainstream analysts and media starting to pick up on the same issues, reinforce the message, and remove the confusion around this.

The good news is these are easily solvable issues, it's just a matter of helping companies work together to solve them. OSA members have been doing this all along, delivering more turnkey, interoperable solutions. JasperSoft/Hyperic, Ingres/JasperSoft, and, of course, our own Common Customer View are great examples that can be easily replicated by other vendors in the industry. Now is not too late to join us and participate in our next round of interoperability initiatives!

As a further proof of the maturity of the open source market, the OSA (Open Solutions Alliance) has recently celebrated its first anniversary. As a reminder, the OSA is a consortium of open source software vendors dedicated to making enterprise-class open software solutions work together. It helps customers put open solutions to work by enabling application integration, certifying quality solutions, and promoting cooperation among open solutions developers. Officially created one year, the OSA has just held a new board election and has also announced earlier this year its European chapter. This European chapter has itself chosen a temporary board of directors, whose first mission will be to select where the organization will be incorporated. I am proud to say that Talend’s General Manager for EMEA, François, will be sitting on this board.

Six directions have been decided for the European OSA for this year:

• Promoting and defending interoperability


• Developing an ecosystem through the recruitment of vendors, SIs and ISVs


• Marketing and business development for and around the EOSA members


• Developing relationships with the open source community


• International exchanges and relationships with the other OSA chapters


• Promoting and defending open source in Europe

As explained by Dominic Sartorio, the OSA President, the first year of such a consortium usually carries high risks that can put in question its existence. The OSA has brilliantly worked around all obstacles and starts its second year, leveraging a growing members number (22 to date). Dominic’s article that presents this first year from the insider’s perspective calls for even more collaboration between organizations.

I have no doubt that even more of us will celebrate the OSA’s second anniversary next year.

Bertrand

OK, time to wave the white flag... I'm a former developer, and while it's been over six years since I've written a line a code, I couldn't resist spending an unhealthy amount of time today wading through the 30,000 new pages of APIs and interoperability documentation on the MSDN site.

In particular, I'm looking for solutions to our "Top 5" hitlist of interoperability challenges, as measured by customer feedback during last fall's customer forum series. My fading developer skills must be failing me, or call me an old fogey, because I'm simply not finding what I'm looking for. Rest assured I'm not trying to express skepticism in a back-handed way - Judging from the announcements and resulting coverage, I really believe this is there - or soon will be there. Perhaps it's staring me in the face, but am just not recognizing it. So, a little help, please? If you find it, please reply to this blog or email me directly!

#1. SQLServer internals

Use case: Porting open source applications that used MySQL, PostgreSQL or even Oracle over to SQLServer. Somehow it just doesn't perform as you expect, or may not even work at all. There is tacit knowledge that you lack, regarding parsing/unparsing SQL statements, managing result sets, or whatever, and just reading the driver interface specs isn't helping. You need better understanding of how SQLServer interprets SQL, optimizes for performance and memory management, deals with the network traffic, and so forth. Heck, maybe you have a specialized high-performance app and want to write your own database driver to better manage resource utilization.


#2. Windows desktop protocols

Use case: Internet Explorer runs well on Windows. Very well. Performance, direct use of device drivers and OS libraries, and so forth. You want your own application to enjoy the same level of access, and you don't get it through Visual Studio.


#3. C# Runtime Internals

Use case: There are a lot of custom .Net applications out there, written in C#. Actually, many customers we spoke with have two standard technology stacks, one Java-based and the other Microsoft-based. Inevitably, applications need to be integrated across these two stacks, requiring development of adapters and plug-ins in one or the other environment. Or, one needs to port an application from one to the other - Standard Java to C# (and vice versa) converters exist. However performance is often an issue. Sun has (finally) gone open with its JVM, and even before that, one could license the technology, so that one can understand threading, classloading, garbage collection, and so forth. Would love to find the equivalent for C#.


#4. Sharepoint data structures and storage

Use case: Of course, it's been possible to share data with Sharepoint in a loosely-coupled way for some time, e.g. WebDAV and web services. However we are seeing higher and higher load applications that store and update vast quantities of data, both structured and unstructured, often with high concurrency. We are also seeing many different kinds of applications using Sharepoint as a back-end - Wikis, blogs, customer records, email, document storage, all banging against the same Sharepoint instance. Similar to the SQLServer issue above, at some point you need to tune for performance or create policies regarding what kinds of data can be shared when, and this requires lots of tacit knowledge regarding Sharepoint's internal behavior.


#5. Plug-ins for Granular Data Access in Office

Use case: It's surprising how much business-critical data is stored in Excel, Access, and other Office products these days. As an example, I once encountered a billion-dollar-per-year manufacturing company that did its MRP and closed its books entirely in Excel - No kidding! This is an extreme example, and in many cases, a forklift operation to a more scalable business application isn't necessary. However integrating into that data often is. Import/export, or "Save As..." isn't practical, and you really want to avoid DLL hell. What's needed is simple real-time programmatic access to individual data points, wherever and however they may be stored. Again, the same issues of performance tuning some into play as well.


Anyhow, these are one former developer's "Top 5", and yours may differ. But a little help is more than welcome!

Once again Microsoft is in the news - This time with a significant announcement regarding interoperability. The media response has been tremendous, and reactions from the open source community have been as varied as ever, ranging from giddy enthusiasm that this is a "major turnaround" to skepticism that this means anything or nothing more than tactical self-interest.

This blogger is taking the view of cautious optimism. While it doesn't represent a fundamental change in business for Microsoft, we see it as a pragmatic reaction to customer demand, and nonetheless resulting in a significant net positive for the industry.

The Interoperability Mandate

Of course, regular visitors to this site know that we believe interoperability is a key issue for customers. In our customer forum series report published last December, we wrote that customers pointed to Microsoft interoperability as a particular issue that the industry has collectively done the least to address. What's interesting is that the relative volume that this is being voiced has been steadily increasing in the last year, even from one forum event to the next. We believe this is tied to the recent rapid adoption of open source solutions in the enterprise, as opposed to simply Linux, Apache and other infrastructure. It's in the solutions space that integration issues become more complex and multi-faceted, and where lack of available APIs or expertise becomes particularly painful. Our guess is Microsoft's executives were hearing the same thing, to the point where this became necessary for fear of alienating its customers.

The Real News

The real news here is the opportunity that API availability creates for third-parties, including integrators, developers and other vendors. Now, there are over 30,000 new pages of content on its MSDN site that they can leverage to meet customers' interoperability requirements. We expect there will be a rush of integrators pursuing the opportunities this will create.

The Non-News

There were other aspects to the announcement that we perceived as non-news and, if we had the opportunity, would have advised against even discussing, because of the predictable derision these have endured. If anything, these just fanned the flames of long-running controversies regarding Microsoft's handling of IP and our patent system in general, and distracted from the real news regarding API availability and the resulting third-party opportunity. These include:

Assurance not to sue developers working on projects that may infringe Microsoft IP, as long as done for non-commercial purposes. Microsoft has said this in the past, thus isn't new news.

Stating they would publish the list of 235 patents allegedly being violated. Our opinions regarding our patent system notwithstanding, it is our current law, and most community developers want to abide by it, even if just to avoid trouble. If given the opportunity, they will make good faith efforts to avoid duplicating others' IP. Publishing this will be helpful, but until that actually happens, this is non-news.

The Big Picture

Nonetheless, overall, we see this as an important step in a series of steps that Microsoft is taking to be a more open company. And become more open they must, as must any software company in the 21st century. Open, transparent and collaborative ways of doing business are becoming the norm - and this isn't just about opening up one's source code, but throughout all aspects of running a business and engaging with customers and partners. Customers are increasingly demanding this, the next generation of professionals raised on the Internet is demanding this, and those companies that don't evolve towards this reality will gradually become marginalized.

Of course, many in the community would wish that Microsoft (and other proprietary companies') management would just roll over and go open, in one climactic, cathartic moment. This is not realistic. Any of us who have worked for large organizations know this: True change can be agonizingly slow, with fits and starts, and is not for the impatient or faint of heart. The old way of doing business has made many successful careers and many individual fortunes, leading to entrenched interests and a deep desire for continuity. It has made organizational models and work habits that countless professionals take for granted in their day to day work. It has made sacred cows that many forward-thinking leaders and change agents may find difficult to overcome. In such environments, one has little choice but to take small steps, tied to tangible, external demands (such as customers') that are consequently easier for the organization to accept.

I can only imagine how this dynamic unfolded in Microsoft leading up to yesterday's announcement, and I don't envy the challenge. Congratulations to Microsoft's "change agents", including Bill Hilf, Sam Ramji and many others, for making these small and achievable steps towards a more open Microsoft ecosystem.

A day has passed since the end of this year's Olliance Think Tank in Napa, California, and I've had a chance to reflect (and recover).










This is a unique event on the open source calendar, well-attended by numerous entrepreneurs, executives, industry experts and thought leaders in commercial open source. Key trends and issues facing the industry were discussed, plus ample opportunity for offline networking. Not to mention the all-important choice of venue (Napa Valley is the capital of the USA's wine industry) and numerous opportunities to partake in the local product, which helped dull the edges of the many Type A personalities in attendance!

There were several takeaways from this event, with the most significant being that it's no longer a matter of whether open source will succeed, but how. The was a pervasive sense of optimism and excitement at this event, permeating every presentation and offline conversation.
Recent M&A activity was the elephant in the room, pointing the way towards the most viable and successful open source business models. Last month the OSA predicted that, by year-end, the question of what are the "best practice" open source business models would be settled - We may be approaching that already.

Nonetheless there is room for improvement, especially according to the CIO panels. Representatives from ABN Amro, Golden Gate University, Rolls Royce, Silicon Valley Bank, Seagate, and the always irrepressible Tim Golden from Bank of America, described their use of open source. All shared their positive experiences and the reasons why open source was strategic in their organizations. Lower cost, less risk of lockin, ability to innovate with people outside their (and their vendors') organizations, and the spirit of open-ness and collaboration were all reasons for using open source. But all cited a pattern of open source products lacking the same polish, "fit and finish" of typical proprietary products, "an over-emphasis on elegant technical solutions at the expense of better deployment experiences, better documentation, better support and services, better interoperability, and more professional communications with the end user", in the words of one of the panelists.

Another theme was the proliferation of products and projects, and the lack of cooperation between them. Of the hundreds of thousands of projects on Sourceforge, how many are viable and can meet enterprise requirements? Of those that can, how well do they work together and with proprietary technologies, so that larger enterprises can readily make use of them? To achieve this, their sheer number means there is clearly a need for more multi-lateral behavior on the part of these projects, much more so than the already-consolidated proprietary vendor community needs in order to succeed. But instead, open source seems to be balkanizing into specific ecosystems, e.g. Eclipse, Apache, and so forth, with developers choosing to be part of a given "club" over others. Similarly, many leading companies choose to "go it alone" and neglect working with other companies to improve the overall "fit and finish" of their products. If we're not careful, this trend can further limit our success.

But the good news is these challenges would be front-and-center at an event like the Think Tank. CIO's were raising these as issues, brainstorming sessions focused on these, and offline conversations frequently turned to these. Knowing we have a problem is half the battle.

The OSA was well-attended, including AdaptivePlanning's Bill Soward, Black Duck's Doug Levin, EnterpriseDB's Andy Astor, Hyperic's Javier Soltero, Ingres Roger Burkhardt, Iona's Peter Zotto and Larry Alston, OpenITWorks' Michael Grove, Sourceforge's Ross Turk, SpikeSource's Kim Polese, and Talend's Bertrand Diard. Thanks to all of you for your support, and for helping make the OSA part of the ongoing dialog on our industry's evolution.

At this point, the OSA could take credit for helping bring these issues to the forefront, but that would defeat the spirit of community and multi-lateralism that we have championed since our very beginning. Instead, let's continue to work together to drive commercial OSS forward, delivering solutions that customers need, in the form that they need.

And finally, hats off to the Olliance Group for organizing this event. The quality of people at this event was second to none, with numerous networking opportunities and very timely and relevant presentations. With every conversation there was something new to be learned and new perspectives to be considered. Andrew Aitken at one point mentioned that there will be a time where this event won't be needed anymore, but until then, we hope this continues to be a key event that helps move the OSS industry forward.

If you can understand French, you will certainly enjoy these three videos from Solutions Linux:

• Cédric Ingrand of LCI interviewed Bertrand for the Plein Ecran segment he filmed at Solutions Linux, that aired on LCI (La Chaine de l’Information), the French 24 hour news channel (you can also download the video file - MPEG, 84 MB).


• Bertrand was also invited by Philippe Guerrier of VNUNet to discuss proposition 58 of the Attali Commission report.


• And in another video podcast with VNUNet, Bertrand presented the OSA Europe initiative.

In addition, the multiple announcements we did at Solutions Linux generated lots of interesting coverage. I won’t list them all here since many are in French, but all of this testifies to the phenomenal growth and relevancy of open source in general, and especially business open source.

Yves

These past few weeks featured a lot of announcements by Talend, by the OSA, and by the open source community in general. The timing of Solutions Linux was very good with that regard, and many members of the press are highly interested in hearing from open source thought leaders.

On Wednesday alone, Bertrand was interviewed and filmed by three different media outlets: LCI (the French 24-hour information channel) - see photo, VNU.net, and NetEco. All segments are expected to air within one week, we’ll post links to them when they are available online. Topics addressed in these interviews ranged from open source adoption to interoperability, from the launch of the OSA in Europe to the Attali Commission report, and a few others.

Dominic, the OSA president, and myself were also interviewed by several high profile media outlets in the US, the UK, France, etc. on the OSA Europe launch.

Yves

Today during my keynote at Solutions Linux in Paris this morning, I announced the creation of the European Chapter of the Open Solutions Alliance, the open source consortium Talend co-founded one year ago. For the OSA, this means “acting local”, while still “thinking global”.Indeed, more open source projects have originated in Europe than anywhere else in the world, which gives the region a unique foundation for building business value for users. Even for global vendors like Talend, it is important for open-source companies to work together to maximize the opportunities for open solutions in the European market.

This new chapter structure will better enable the OSA to address the needs of vendors such as Talend, who are deploying business-ready open solutions all over the world, but with a regional focus that is tailored for regional differences in culture, business practices, regulatory environments and open source adoption patterns.

Over the course of the next three months, we will finalize the creation of this chapter. Stay tuned for more information.

Bertrand

Greetings from SolutionsLinux in Paris, France. It is now 2:30am in my local timezone - Such are the joys of "Going Global"!

The OSA has had a great first year, attracting significant membership and driving several important interoperability projects. With many of our early members being USA-based, many of our activities have been in the USA. As we continue to grow, we have recognized the need for driving more activity around the world, and we have given careful thought to how to do this in a scalable way. How can we think globally, while ensuring greater activity on a local basis? How can we better serve member companies, no matter where in the world they happen to be based?

Time differences just scratch the surface of the challenges of global coordination. Open solutions are truly a global phenomenon, with adoption increasing in most regions of the world. However those adoption patterns differ. The European public sector, for example, has been driving adoption to a greater degree than in the USA, which has been mostly private sector led. Differences in culture and business practices also affect where and how commercial open source products are adopted. This all adds up to a need for local spins on our message and methods, and a need for local people to drive activities and their relative priorities.

Early feedback from many European companies has been positive, with several prospective members indicating they would join a local chapter that would serve as a regional focal point for collective action.

We have heard the same from similar companies in Japan and other parts of the world. Indeed, by this time next year, we expect a large and vibrant European chapter, as well as new chapters in Japan and possibly other parts of Asia, and Latin America.

2008 will be a turning point year for open source solutions, with clear separation between winners and losers, as we have recently written. Many companies will figure out that "go-it-alone open source" won't be their path to success. Our goal is for the OSA to be there for them, no matter where in the world they happen to be. Vibrant chapters in key markets will be the way we make that happen.

By the way, if you're at SolutionsLinux, please stop by our session track on Tuesday afternoon!

We announced today a technical partnership with Microsoft. The goal of this partnership is very simple - and very transparent: ensure interoperability of Microsoft systems with the rest of the information system. Nobody can ignore the fact that IT systems are made of a combination of open source and closed source technologies, of databases and packaged applications, of legacy and SaaS systems, of… you name it, it’s probably in there!

Interoperability is the core of our business, and it’s a subject that is close to my heart (this is why Talend is a founding member of the OSA, for example). So when Sam Ramji from Microsoft invited us to their Open Source Lab in Redmond, and offered to help us work better with their systems, we decided to use these resources for the greatest benefit of our users. And we also liked the fact that only very few vendors are admitted in the lab: MySQL, Firefox, XenSource and now Talend are among the chosen ones. A clear tribute to Talend’s success.

Specifically, our work with Microsoft was three-pronged: improving our performance on Windows (including Windows 2008), supporting SQL Server 2008, and providing connectors for the Microsoft Dynamics applications.

Of course, this does not lower our commitment to the support of open source technologies: Linux operating systems, open source databases and packaged applications, etc. Open source is also freedom of choice, and interoperability between all systems is the path to this freedom of choice.

Bertrand

A funny thing happened on the way to the office

I hopped on a plane in New York this morning and planned on writing a blog about new year's resolutions. Resolution one is to "Stop Procrastinating" which might explain why that particular blog is three weeks behind schedule. When I landed in SFO, my blackberry almost melted under the strain of all the email traffic around the MySQL and BEA acquisitions.


Full Story

Sun's purchase of MySQLfor a cool $1 billion is all the buzz today and rightfully so. I think this news goes to show the power, visibility and growth in adoption of open source. And in this particular case MySQL.

As enterprises continually seek competitive advantages they find their way to open source solutions. Sun has certainly recognized this with Jonathan Schwartz at the helm, but today's news takes another step towards ensuring they are going to be a major player in offering such solutions. It will be interesting to see how this purchase affects Sun as the inherent value proposition of MySQL is that it is an open, potentially low cost alternative to Oracle. The other potential hitch is that companies want best-of-breed solutions for their needs and Sun just might be becoming a monolithic vendor offering one set of solutions.

I am thrilled to see the increased adoption of open source and the excitement of my friends over at MySQL, but also see this news as an opportunity for system integrators (SI's). With IT spending and acquisitions on the rise, it furthers the need for technology agnostic SI's (Unisys, Cap Gemini, Accenture and others) in this space to play significant roles in customer IT implementations. Companies are looking for the best independent solution for their needs, not for one monolithic vendor to lock them into an IT environment of their choosing. The OSA’s common-customer view reference architecture, as one example, demonstrates interoperability across multiple vendor products to achieve best-of-breed functionality.

Another piece of big news today (other than the continued economic impact in the Financial Services sector as a result of the sub-prime lending fallout) is Oracle’s purported acquisition of BEA. Obviously this would further position Oracle against IBM in the middleware space and further Oracle's goal to dominate that market. But this also creates another “integrated stack” that possibly precludes clients from choosing best of breed product, hence furthering the need for large Systems Integrators to help fulfill that gap.

Yes, that's "billion", with a 'B'! I'm sure the valuation negotiations considered the media impact and attention of a nice round number. They succeeded. Already the punditocracy is a-twitter, with lots of ink being spilt and keyboards being worn out about how this should prove (once and for all) that open source business models are viable or even superior to other ways of running a software company.

This industry observer begs to differ. MySQL didn't succeed just because it was an open source business. They succeeded because they mastered how to deliver customer value, and brilliantly executed a strategy that is tried-and-true in the software business: brand, platform, and meeting customer needs. Open source was a means to an end, not an end unto itself.

Let's look more closely at their keys to success:

Platform. MySQL correctly read the rapid emergence of Web 2.0 based applications and websites, and customers' de facto use of a new technology stack (LAMP). They successfully positioned themselves as the centerpiece of this new technology. How open source helped: By making it easier for developers to get their hands on the product and start developing, and fostering developer loyalty. But this couldn't succeed without...

Brand. MySQL invested heavily in marketing its brand, trumpeting customer success stories as well as the strength of its product features. By aggressively positioning as the de facto database for Web 2.0 based apps, this drove those developers to adopt MySQL's product instead of others, when they otherwise may have been ambivalent regarding choice of database. But this branding effort couldn't succeed without...

Product meeting customer requirements. MySQL didn't try to be an all-things-to-all-people database. They focused on the Web 2.0 market, and built a product that met this market's data storage, security, management, performance and other requirements. How open source helped: Open source enabled more product innovation outside the company, which helped improve the product. But let's not discount the discipline of an excellent product management and engineering leadership team that laser-focused on its target market. Of course, this was also helped by focusing on a technology platform that could be leveraged and built upon.

In short, they executed the "building a business based on customer value" playbook extremely well, with a multi-pronged mutually-reinforcing strategy of brand, platform and product excellence.

The lesson is that companies that listen to its customers, and execute a strategy based on customer value, will succeed. Those that resort to ideology ("we're better just because we're open source") will not. Open source should support the strategy, not the other way around. I know Marten (he sits on my employer's advisory board) and he's as pragmatic and customer focused as they get. This pragmatism and customer focus will continue to separate the winners from the losers.

Now, after dominating the LAMP based market, I would suggest the next stage of the MySQL business unit's growth would require thinking about interoperability with other technology stacks, and this is where the OSA can help...