Tracking change and innovation in the enterprise software development community
Posted by Al Tenhundfeld on Jun 22, 2009
Microsoft announced System.Data.OracleClient will be deprecated after .NET 4.0. Classes in the namespace will be marked obsolete in .NET 4.0 and removed from future releases. OracleClient is the ADO.NET provider for Oracle developed by Microsoft and shipped as part of the .NET Framework Class Library.
This decision has sparked controversy among the community of .NET developers working with Oracle. While many enterprise .NET applications use a 3rd party Oracle provider, System.Data.OracleClient is often used in small applications and typically has better integration with other Microsoft tools.
Microsoft insists this decision was made after much deliberation and research:
After carefully considering all the options and talking to our customers, partners, and MVPs it was decided to deprecate OracleClient as a part of our ADO.NET roadmap.
Part of the reasoning for this decision is the increasingly availability and improvements of 3rd party ADO.NET data providers for Oracle. There have been significant performance improvements and enhanced multi-version compatibility among the popular Oracle providers:
Despite the community backlash, Microsoft has shown no signs of reversing the decision, and none should be expected. Microsoft's official line is that "many of the third party providers are able to consistently provide the same level of quality and support that customers have come to expect from Microsoft." Therefore, it's not worth the investment to bring OracleClient up to parity with the third party providers, and this frees MS resources to focus on ADO.NET.
Some are calling this move an underhanded strike to raise the bar for .NET development against Oracle, but many are viewing this decision with a guarded optimism that Microsoft might be rethinking its not-invented-here bias. For example, it is a common complaint among the ALT.NET crowd that Microsoft reinvents the wheel for no reason, creating their own versions of products when good alternatives already exist. See MSTest vs. NUnit or Entity Framework vs. NHibernate. In the context of recent decisions to officially support jQuery and provide the source code for ASP.NET MVC, this decision could be interpreted as Microsoft further admitting that it doesn't need to control the entire stack; they can rely on their partners and the community to provide some pieces.
Also, the Connected Show Podcast has an episode discussing this move and examining why it's not a disaster.
I support that.
It's a long way from Microsoft realizing that Oracle is not part of their stack to acknowledging that they don't need to control the entire stack at all. ;-)
I agree it's optimistic, probably overly optimistic, but there have been some encouraging signs that Microsoft, especially Scott Guthrie's group, are listening to their developer community's pleas for more openness. I'm not trying to say MS is doing anything just to be nice. I think they're intelligently responding to the fact that there are many options for programmers nowadays, and if they maintain their closed, iron-fisted approach, they will lose some of their most vocal developer evangelists to languages/environments like Ruby, Scala, etc. Is deprecating OracleClient part of this shift towards openness? Maybe that's a stretch; I think the ADO.NET team just decided their resources would be better spent elsewhere. But I don't think deprecating OracleClient is part of some evil conspiracy to marginalize Oracle; this move will probably be beneficial for .NET Oracle development in the long run. And I wouldn't be shocked if, in the next few years, Microsoft released the source code for OracleClient under the MS-PL.
I believe you're right about the encouraging signs and ScottGu and friends. Just saying that OracleClient is probably not one of those signs. That was just a move to establish .NET in Oracle environments. No need for that any more, Oracle provides their own provider, and they're probably pouring more resources into it anyway.
Download
new Adobe®
Flash® Builder™
Book excerpt
How are messages
stored by
Apache ActiveMQ?
JBoss vs. Websphere
Cost, Performance,
Efficiency,
Innovation
IBM wins.
Scaling Scrum
in the enterprise?
ScrumWorks Pro
can help.
Start a trial today
Jason H Christensen presents the evolution of mobile devices, how to integrate mobile systems with the cloud and use RESTful services, what is a basic mobile architecture and how it can be implemented
Kresten talks about current languages and problems they solve. He tries then to portray the language that would take us to the next level helping to tackle issues we face today such as concurrency.
Tim Smith introduces the licenses offered by Oracle/Sun and Microsoft, with a focus on how they may affect third party implementation such as Google’s Android platform.
Russ Miles and Toby Hobson outline many factors to be considered when adopting a cloud solution, creating a wider view of the cloud from the development and business perspective.
Kelley Horton discusses the reasons why her organization transitioned to Lean-Agile, the approach used and the visual tools helping them minimize WIP, concluding that visibility leads to success.
Alex Blewitt, Kevin Seal and Alex Buckley answer Java modularity-related questions: when is modularity needed, how to address it, and what are the improvements in OSGi-based development.
Adam Blum discusses the current trends in mobile development and smartphones, trying to predict what will happen in this area over the next 5 years so a developer would know what to expect.
Roy Osherove discusses the difficulties met when trying to test code embedded in a framework (cog), presenting several solutions to create unit tests for cogs, using Silverlight code as example.
4 comments
Watch Thread Reply