Port 25: The Open Source Community at Microsoft : Community, Interop, Sam Ramji

Releasing the Linux Integration Component Drivers...

Sam Ramji — Fri, 24 Jul 2009 01:38:00 GMT

Microsoft on Monday contributed the Linux Integration Component drivers to the Linux community for the reasons stated in our release. Microsoft chose the GPLv2 license for the mutual benefit of our customers, partners, the community, and Microsoft.

Microsoft's decision was not based on any perceived obligations tied to the GPLv2 license. For business reasons and for customers, we determined it was beneficial to release the drivers to the kernel community under the GPLv2 license through a process that involved working closely with Greg Kroah-Hartman, who helped us understand the community norms and licensing options surrounding the drivers.

The primary reason we made this determination in this case is because GPLv2 is the preferred license required by the Linux community for their broad acceptance and engagement. For us to participate in the Linux Driver Project, GPLv2 was the best option that allowed us to enjoy the tremendous offer of community support. The community's response even within a few hours of posting the code was welcoming and we appreciate it greatly.

We arrived at the decision to release the drivers to the community under the GPLv2 through this process. Both Greg K-H and Jim Zemlin of the Linux Foundation have reiterated that this is the same process that other companies follow when deciding how to release new device drivers to the Linux community.

We are looking forward to the positive collaboration and acceptance that has marked the vast majority of our interactions with customers and community members regarding this important project.

Cheers,

Sam

Updated 7/25/2009 @ 11:54 AM Pacific: Dave Roberts of Vyatta posted a blog entry rebutting recent cloims that we were accused of a licensing violation with some detail on the technical issues.

Microsoft Releases Device Driver Code to the Linux Community

Peter Galli — Mon, 20 Jul 2009 16:48:00 GMT

In what many may see as a surprising move, Microsoft today released 20,000 lines of device driver code to the Linux community under the popular General Public Licence v2.

The code includes three Linux device drivers, and has been submitted to the Linux kernel community for inclusion in the Linux tree.

The drivers will be available to both the Linux community and customers, and will enhance the performance of the Linux operating system when virtualized on Windows Server 2008 Hyper-V or Windows Server 2008 R2 Hyper-V.

In an article posted to Microsoft's PressPass site, Tom Hanrahan, director of Microsoft's Open Source Technology Center, notes that this is a significant milestone because it's the first time the company has released code directly to the Linux community. "Additionally significant is that we are releasing the code under the GPLv2 license, which is the Linux community's preferred license," he said.

In the same article, Sam Ramji, senior director of Platform Strategy at Microsoft, points out that Microsoft communities and open source communities are growing together, which is ultimately of benefit to customers. An example of this is the Linux community, which has built a platform used by many customers. "So our strategy is to enhance interoperability between the Windows platform and many open source technologies, which includes Linux, to provide the choices our customers are asking for," he said.

Ramji also alluded to the fact that people are often surprised when they hear how much open source community and development work is happening across Microsoft, which is largely due to the fact that these collaborations focus more on getting the work done and engaging with the various communities on a one-to-one basis and less about promoting them.

One example of how Microsoft participates with, and contributes to, open source is its relationship with the PHP Community. The company's involvement includes contributing to the PHP Engine, optimizing PHP 5.3 to perform strongly on Windows, and working to improve the performance of numerous PHP applications on Windows. Then there is the ongoing participation in various Apache Software Foundation projects, such as Hadoop, Stonehenge and QPID.

"In short, we're focused on building sustainable business strategies for open source at Microsoft ... we see open source playing into three key areas, one of which is the use of 'inbound' open source and the open source development model to make our software development processes more efficient."

"Good examples of this include what we did recently with jQuery in Visual Studio 2008, the implementation of OpenPegasus connectors and adaptors into System Center Operations Manager, and work that the Microsoft High Performance Computing team did with the Argonne National Lab (ANL) to source its MPICH2 implementation, which is a portable implementation of the Message Passing Interface (MPI) used in cluster computing and super computers," Ramji said.

We'll be posting a number of other articles on the release of the device driver code to the Linux community over the week, several of which will be penned by Hank Janssen from Microsoft's Open Source Technology Center, so look out for those.

The Linux Foundation's Collaboration Summit

Peter Galli — Thu, 23 Apr 2009 14:00:00 GMT

A lot has been written by the press and blogosphere since the Linux Foundation's annual Collaboration Summit was held earlier this month, particularly about the panel that included Microsoft's Sam Ramji, Sun Microsystems' Ian Murdock, and Linux Foundation executive director Jim Zemlin.

The panel was entitled "Why Can't We All Just Get Along," which struck me as not only divisive, but also a little outdated given the level of collaboration that already takes place between proprietary and open source software vendors alike.

For example, Microsoft and Sun already have a long-standing working collaborative relationship; Microsoft also has a technical collaboration agreement with Novell, an agreement with Red Hat to test and validate our respective server operating systems running on one another's hypervisors, and a number of arrangements in place with other open source companies.

The panel discussed this in greater depth, looking at how collaboration, cooperation and competition exist: not just between proprietary and open software vendors, but also between Linux and open source ones.

This prompted panel moderator Zemlin to suggest that the three make an even greater effort come together and collaborate where it makes sense.

Interestingly, the Summit also spurred renewed discussion about whether there need to be more critics in the Linux community, with one blogger taking Zemlin to task for what he described as the "tall claims" he made at the Summit.

Ramji, the Senior Director of Platform Strategy at Microsoft, also used the panel to remind the Linux and open source communities of his offer for them to reach out to him and others within Microsoft and share their frustrations, problems and issues, so that they could be better educators and advocates on this front across the company.

Ramji also, again, stressed that Microsoft's customers want interoperability with open source software, including for PHP on Windows, but that making this happen sometimes took time.

Sun's Murdock seconded this, talking about internal inertia and how Sun also had had to deal with hearing from customers and developers that they wanted interoperability with technologies other than their own.

At Microsoft, there are cross-group, company-wide open source discussions and initiatives underway, with each group given the autonomy to decide for itself how this plays out with regard to their product set and business model.

While Port 25's mission is to be the voice of the open source community at Microsoft, it is far from the only voice on this topic. There have been blogs across the company on open and interoperability initiatives, from groups including security, Live and the Mac Business Unit, to name just a few.

It is also important to remember that Ramji and other executives like Bob Muglia, the president of Microsoft's Server & Tools business, have often said that open source is a journey that Microsoft is on and that much more needs to still be done. Many groups across the company are already responding to that call.

Open Source Interoperability Projects

Jean Paoli — Fri, 07 Nov 2008 07:21:00 GMT

Interoperability has always been a focus area at Microsoft. Being a platform company, Microsoft has engaged in interoperability at many levels - product features, participation in standardization bodies, publishing many technologies under open licenses and working closely with customers, governments and partners to understand the heterogeneous IT landscape and discuss practical interoperability solutions.

Earlier this year, these activities were formalized under the Interoperability Principles for all of our high-volume products.

I am the General Manager of Interoperability Strategy at Microsoft, and I have worked across the company on many interop initiatives. I am happy to see many interop projects now coming out of Microsoft and, personally, having many of them based on XML makes me doubly happy.

My team has built several bridging technologies and solutions for many of our products to enable interoperability. These are being run as open source projects and released under a broad BSD license so that our customers and partners can use them in many open and broad scenarios.

Interoperability has been getting enhanced attention at a lot of conferences lately and Microsoft has also upped its participation at many open source conferences such as OSCON, the Eclipse Conference and ApacheCon.

At Microsoft's Professional Developer's Conference last month, the interoperability story was part of almost every announcement and keynote address. As Sam Ramji writes in his latest blog, Microsoft is also participating at ApacheCon and highlighting the interoperability work we are doing. These are indeed exciting times!

On the interoperability front, my team has been working with the WSO2 since the TechEd 2007 Conference to demonstrate interoperability using our StockTrader reference application.

This week, the WSO2 proposed a new Apache incubation project, known as Stonehenge, to further this work. The aim of this project is to set up sample applications to demonstrate interoperability with multiple underlying platform technologies by using currently defined W3C and OASIS standard protocols. We look forward to working with WS02 on the scope of this project, and having discussions with the community.

I also want to highlight some open source interoperability projects that my team has been working on with third parties, companies and members of the community at large, which may be very relevant to the readers of this blog.

Eclipse Tools for Silverlight

Eclipse4sl allows Java developers to develop code for the Silverlight platform within the Eclipse development environment, and contains both an advanced project system for creating Silverlight applications and media experiences as well as a compiler for packaging Silverlight applications for deployment.

Interoperability with the Azure Services platform

Announced at PDC recently, the Azure Services Platform is an internet-scale cloud computing and services platform hosted in Microsoft data centers. It provides an operating system and a set of developer services which can be used individually or together. Microsoft .NET Services is a key component of the Azure Services Platform that offers a set of Microsoft-hosted, highly scalable, developer-oriented services that provide the key building blocks, like, Access Control, Service Bus, and Workflow service.

The Azure Services Platform, built from the ground up to be consistent with Microsoft's commitment to openness and interoperability and in that spirit, we have built two cross-platform SDKs for .NET services - for Java and Ruby.

Information Cards Interoperability

Windows CardSpace is Microsoft implementation of Information Cards on the Windows platform. Information cards are a core part of Identity Metasystem and help both site owners and visitors to manage, control, and exchange digital identities more safely and consistently.

We have also built four open source projects that help Web developers support information cards on diverse platforms:

OpenXML-ODF translators

The goal of this project is to provide translators to allow for interoperability between applications based on ODF (OpenDocument) standard and Office Open XML standard. The translator is based on XSLT transformations between two XML formats, along with some pre- and post-processing, and is available on Sourceforge under a BSD-like license.

OpenXML-UOF translators

The goal of this project is to provide translators to allow for interoperability between applications based on UOF (Uniform Office Format) standard and Office Open XML standard.

UOF is an emerging standard, which is being developed by the Chinese Office Software Work Group (COSWG), led by the China Electronics Standard Institute (CESI), the Ministry of Information Industry (MII), major suppliers of Chinese office software suites, and other academic institutions.The translator is based on XSLT transformations between two XML formats, along with some pre- and post-processing. It is available at SourceForge under a BSD-like license

I would like to hear your comments and feedback on these projects and also welcome open engagement on what Microsoft should be doing for interoperability. Tell us what other interoperability scenarios we should be looking to address.

I also want to thank the multiple third party companies and the community members we cooperate with, as well as the members of my team: Vijay Rajagopalan, Sumit Chawla, Kamaljit Bath, Claudio Caldato, Jean-Christophe Cimetiere and many others for working on these projects and building technical solutions for interoperability with key Microsoft products and technologies.

struct.new("future", :open, :microsoft)

Sam Ramji — Thu, 06 Nov 2008 08:49:00 GMT

I delivered the keynote at ApacheCon in New Orleans today, where I talked about some of the new milestones we have chalked up on the journey inside Microsoft towards greater participation and growth with open source communities, and our strategy of "architecting for participation."

This strategy focuses on four significant themes: community; contribution; partnerships; and choice. Microsoft believes that the next ten years of software will be a time of growth and change where both open source and Microsoft communities will grow together.

We also believe that in an increasingly interconnected world, where more people have a greater opportunity to use more technology to do more things than ever before. We support those choices and are expanding interoperability between open source technologies and Microsoft technologies.

So, on the interoperability front, we have been working with the WS02 since our TechEd 2007 Conference, to demonstrate interoperability using our StockTrader reference application. Today, the WS02 announced they would build an open source version of the sample application under "Project Stonehenge," which hs been proposed as a new Apache incubation project.

WS02 will use the project to set up sample applications that demonstrate seamless interoperability across multiple underlying platform technologies, using currently defined W3C and OASIS standard protocols.

My team has been working closely with that of Jean Paoli, the General Manager of Interoperability Strategy at Microsoft, whose team is driving much of this interoperability work. You can read more about all this in Jean's blog post.

Microsoft has also decided to move the development of protocol parsers for Microsoft Network Monitor - a free protocol analyzer and network sniffer - to an open source model, on CodePlex, which will host the development of parsers for public protocols and for protocols described in our Open Protocol Specifications for Windows.

An updated parser package has been released and a source tree created on Codeplex. We want Netmon to be the best-of-breed tool for network monitoring at Microsoft, not just for Windows.

Microsoft also recently joined the AMQP Working Group as a participant, with the goal of contributing towards the development of the specification and to enable greater customer choice in the marketplace.

At the request of community members, we have now committed to participate in the Apache Qpid project, a widely adopted open source implementation of the AMQP specification that addresses the customer need for choice and improved messaging interoperability.

Our customers are telling us that they would like to see the Apache Qpid project extended to interoperate with Windows, so the next few months of participation will be focused on understanding the community's effort to build Windows based AMQP software.

Participation will give us the opportunity to learn from other project participants, so that we can be in a position to consider making a valuable contribution. But it is important to note that the Apache Qpid project is just one of many AMQP specification implementations, and we are open to supporting additional projects.

You can read an interesting technical research paper from Ohio State University analyzing the performance of the Qpid implementation of AMQP here.

Microsoft also announced, at PDC 2008, our commitment to include "Oslo" - an upcoming set of technologies for modeling - in the Open Specification Promise. This will ensure that the "Oslo" declarative modeling language, codenamed "M", is interoperable with prominent industry standards such as WS* specifications, XML formats, industry protocols, and security standards.

Two of the core focuses for Oslo are integration and interoperability. As such, it will integrate with next-gen Microsoft technologies, including System Center, Visual Studio and BizTalk Sever. We also plan to work with partners and the industry, so as to make Oslo interoperable with important standards and industry protocols.

One of the key ways we think customers will achieve customization for their platforms is through the use of textual and visual DSLs, which can be written uniquely by the developer for vertical industries and specific domains, or they can use pre-existing DSLs in these same scenarios.

The hope is that we will establish a broad and open ecosystem around "M" that will enable customers to bring the power of model-driven applications and systems to their heterogeneous environments.

Finally, on the Live Search front, the Powerset team recently resumed its participation with HBase, which is elated to infrastructural storage technology enabling large scale data processing.

The HBase project receives significant lift from the active community that supports the project, and Powerset's continued participation on HBase could allow us to accelerate the integration of Powerset's technology into Live Search, resulting in improvements to the end-user experience.

So, stay posted. There's a lot more to come!

Opening Day: Azure Platform Debuts

Sam Ramji — Mon, 27 Oct 2008 14:00:00 GMT

Today at PDC in Los Angeles, Ray Ozzie unveiled the Azure Services Platform, which will enable developers to build the next generation of applications - spanning all the way from the cloud to the enterprise data center. My team's focus has been on making sure that this platform treats open source development technologies as first-class citizens.

A key components of the Azure Services Platform is Windows Azure, an infrastructure that provides core capabilities such as virtualized computation, scalable storage, and automated service management. Developers will be able to build or extend parts or complete service-based applications using Live Services, .Net Services and SQL Services.

They will also be able to choose from a range of open source development tools and technologies, and be able to access Azure services using a variety of common internet standards, including HTTP, REST, WS* and Atom.

The Azure platform's goal is to support all developers and their choice of IDE, language and technology. We are also providing programmable components that can be consumed by other applications, and Microsoft is funding and sponsoring open source software development kits to enable Java and Ruby developers to take advantage of Azure.

This is significant as this is the first time we are delivering cross-platform software development kits at the same time as Microsoft Developer Network software development kits.

We are also funding these open source projects, under the BSD licensing model, in collaboration with Thoughtworks Inc. and Schakra Inc., and they will be run on open source portals RubyForge and SourceForge.

Much of this interoperability work was undertaken by Jean Paoli, the General Manager for Interoperability Strategy, and his team, including Vijay Rajagopalan, the Principal Architect for Interoperability Strategy, so a big thanks is due to them on this front.

In addition, as part of Microsoft's commitment to openness and working with open source communities, I asked the Open Source Technology Center (led by Tom Hanrahan) to come up with some specific examples that show how open source communities can access Windows Azure.

This work has allowed us to deliver several ‘proofs of concept' which show open source developers that they can create applications that run as services and have access to services in the cloud. These ‘proofs of concept' demonstrate that:

A developer using the Eclipse IDE can write a C# application that runs on Windows Azure
Gallery, the leading PHP photo application, can access Windows Azure cloud storage
A blog engine hosted on Windows Azure can authenticate users with OpenID.

Specific to Gallery, we've done two simple things: we created wrappers to convert the Windows Azure API to PHP objects, and we created a Windows Azure subclass inherited from the Windows NT Platform class. The net of all this is that, with a small amount of code, we were able to connect one of the top PHP application to Windows Azure, specifically, photo images stored as BLOBs in the cloud.

Finally, Microsoft is also going to publish the "M" language specification, including MSchema, MGrammar and MGraph, under the Open Specification Promise. This will facilitate the interoperability of the "Oslo" declarative modeling language, codenamed "M," with prominent industry standards such as WS* specifications, XML formats, industry protocols and security standards.

Stay tuned, because there's more to come.

Cheers,

Sam

Open Messaging

Sam Ramji — Fri, 24 Oct 2008 11:00:00 GMT

We've been working with a range of open source projects in the last few years, and each one has taught us something - both what to do more of, and what to change. One of the things we've learned in listening to very specific customer needs, is that there is an emerging pattern of shared software development that will drive changes in how companies buy vs. build software.

Messaging (and I mean enterprise messaging, rather than email) is an area that is of keen interest to customers like JP Morgan Chase and Credit Suisse. As they run their businesses on real-time messaging, they need to be deep experts, and drive changes in their messaging platforms to fit their business. Along with companies like Cisco, Novell, iMatix, RabbitMQ, WSO2, and Red Hat, these industry leaders have built a standard for ubiquitous messaging: AMQP.

The Advanced Message Queueing Protocol is an open specification supported by open source communities and currently implemented by Apache QPID, RabbitMQ, and OpenAMQ. The contributors established the AMQP Working Group as a body to manage the process of developing the specification.

It's my pleasure to announce that Microsoft has been invited to join the AMQP working group by the six founding members. We have committed to participate in the development of the specification and are keenly interested in the developing need for interoperability in enterprise messaging.

While message-based transports with security and transactional integrity are a vital infrastructure component throughout financial institutions, the AMQP specification and related implementations may also provide greater interoperability for a number of other vertical scenarios, including insurance and healthcare. AMQP specifies a wire-level protocol (think of a transport like TCP or HTTP) and FIX, FpML, SOAP, and other messages can be sent of AMQP in LAN and WAN environments.

I think it's particularly interesting to see this trend of industry-specific shared software and protocols. In the case of AMQP, the known implementations are open source (using MPL, BSD, GPLv3, and Apache licenses). In a sense the customer/end-user organizations involved in AMQP - competitors in their core business - are choosing to act as a technology keiretsu within a highly competitive industry.

Our work in AMQP will be consistent with the commitment to openness outlined in July. The AMQP Working Group requires a limited royalty-free patent licensing commitment from its members and, as a participant, we have agreed to grant royalty-free patent licenses on specified terms to implementers of the specification.

The AMQP Working Group is also not a formal standards-setting organization like ISO or a standards consortium such as the IETF, OASIS or the W3C, but rather a group of companies and organizations that have come together to develop a specification to improve interoperability for messaging solutions. Microsoft will help, as appropriate, the Working Group to take the AMQP standard specification to another standards-setting organization, should it decide to do so at a later stage.

So, in short, we hope to contribute to the development of the AMQP specification in ways that will promote interoperability for existing and new implementations.

Cheers,

Sam

A Brief History of Open at Microsoft

Sam Ramji — Fri, 22 Feb 2008 17:06:00 GMT

Four years ago, we started the Linux Lab at Microsoft.
Two years ago, we established the Open Source Software Lab at Microsoft.
One year ago, we initiated the Linux Interoperability Lab at Microsoft.

Yesterday, we announced the broadest change to the way the Microsoft builds software and works with open source communities and developers.

By now you’ve probably read the announcement – “Microsoft Makes Strategic Changes in Technology and Business Practices to Expand Interoperability” and are wondering what it all means, and where it came from. In a nutshell, the documentation for the APIs, document formats, and protocols used in Windows Vista, the .NET Framework, Windows Server 2008, SQL Server 2008, Office 2007, Exchange Server 2007, and Office SharePoint Server 2007 will be made publically available. All developers will be able to access the documentation with no need to sign a license or pay any fee.

We are also announcing the launch of the Open Source Interoperability Initiative – a framework that will let us consistently support community development teams who build implementations of these specs with labs, technical support, plugfests, and joint testing and development.

To me, it’s a logical progression from the work and learning we’ve done with the Mozilla Foundation, the Eclipse Foundation, the Linux Foundation, the Apache Foundation, the Samba Project, MySQL, and PHP. We’ve learned how to make agreements with community projects – including those which lack a legal entity for formal agreements; how to deliver technical support; who to listen to; and how to prioritize our work. We have seen how positively developers and users respond to these kinds of collaborative efforts. This is reflected by the progression of our approach: the creation of the OSP (Open Specification Promise), the IECC (Interoperability Executive Customer Council), the IVA (Interoperability Vendor Alliance), the submission and approval of the Ms-RL and Ms-PL by the OSI, and the PFIF/Samba agreement and ongoing collaboration.

It’s also a major evolutionary step, and significant commitment for our engineering teams. Ray Ozzie says it best: “Customers need all their vendors, including and especially Microsoft, to deliver software and services that are flexible enough such that any developer can use their open interfaces and data to effectively integrate applications or to compose entirely new solutions. By increasing the openness of our products, we will provide developers additional opportunity to innovate and deliver value for customers.”

In order to meet these new and higher standards that we’re setting for ourselves, engineers will need to build public documentation of the new formats, protocols, and APIs they develop as they advance our products. For those of us who write (or have written) code, we realize that this is a significant additional phase to the development cycle: design the feature, specify the feature, implement and test it, then proof and test the documentation of the specification, build user documents and sign off on the whole thing.

Because we are a platform company first and foremost, it will be entirely worth the investment both due to the increased transparency to developers, and due to the expanded range of innovation that can be built on Microsoft technologies.

I think this is a great day not just for Microsoft, but for the software industry. And I thank the people who have helped us learn what it’s taken to get here – most notably Jeremy Allison, Matt Asay, Mike Schroepfer, Andi Gutmans, Jim Zemlin, Mike Milinkovich, Zack Urlocker, Marten Mickos, Andrew Tridgell, Miguel de Icaza and Stephen Walli. We will continue to look to their perspectives and advice as we continue down the open road.

Here are a few of the responses we’ve seen – and I’ll quote from the industry publications and blogs:

LWN.net: “The announcement is sweeping enough to make one check the calendar, but we are still a month and a week early for pranks. Microsoft is making available specifications for APIs and communication protocols for Exchange, Office, SQL Server, SharePoint, and others without requiring a license or royalty payments. They will indicate what patents they believe cover any of the protocols and "will license all of these patents on reasonable and non-discriminatory terms, at low royalty rates." There may be lurking dangers, but it appears to be a sincere effort at providing interoperability.”

Matt Asay (Alfresco/The Open Road): “As a Microsoft admirer, critic, and competitor, I can't help but applaud the depth and breadth of this move ... All in all, a huge day for Microsoft. Will there be gaps in Microsoft's efforts? Undoubtedly. For one thing, it hasn't really made much progress on its covenant not to sue commercial open-source providers, despite what Ina writes. But I'm impressed that it's even bothering to try.”

Andi Gutmans (PHP/Zend): “I believe Microsoft has finally understood that their closed nature has significantly hindered the growth of their eco-system. In many ways the threat of Linux has by many been interpreted as a threat of open-source (wrongly so in my opinion) …. Microsoft is now enabling the open-source community to grow its contributor base around such technologies and significantly improve the delivered quality. As most open-source developers and users live in heterogeneous environments this will benefit many.”

Jeremy Allison (via The Register): "It's definitely a positive step. Doesn't mean any change for us [Samba] as we already had all these docs, and the promise not to sue is only for 'non-commercial' open source, which is a bit meaningless. But that's the same thing we had really (they're listing the patents etc.). At least everyone now gets access to the same info, which I'm very happy about. As for the rest, the devil is in the details. If they can follow through with this, the world will be a better place.”

Zack Urlocker (MySQL/Open Sources): “… even if it was legislated, it's still good for the industry. And it’s good for Microsoft customers. And ultimately, it's probably good for Microsoft to be more open. If Microsoft wants to attract the next generation of developers and users, they should take the hint: Open works.”

We recognize that the communities’ judgments of the significance of this announcement will be entirely based on the actions that follow. The optimistic undertone that I’ve seen so far suggests that we can make real progress.

We will report back frequently on the progress and details of this work – especially on the Open Source Interoperability Initiative – here on Port 25. This announcement is the starting point of the next phase of Microsoft’s work with open source, and as Port 25 readers know, we are here for the full marathon.

The interoperability principles are posted here: http://www.microsoft.com/interop/principles/default.mspx

If you're surprised, you're not paying attention

Sam Ramji — Thu, 20 Dec 2007 05:03:00 GMT

First, let me say thanks to Jeremy Allison and Andrew Tridgell for their decades of hard work and their optimism.

Back in March, Jeremy invited me to talk about Samba and Microsoft, and how we could work together. It turned out that our first opportunity to meet was actually at the annual Samba developers’ conference, SambaXP in Gottingen, Germany in late April. I spent three days there listening to the Samba Team's reports on work they were doing, their observations relating to Microsoft protocols, and at breakfast with Tridge, Jeremy, and other team members we established a potential roadmap for collaboration. Frankly, I think my commitments were viewed with disbelief by some but with cautious optimism by Tridge and Jeremy – as well as by Dan Shearer and by John Terpstra, a man of vision and entrepreneurial spirit.

I worked with legal and engineering teams at Microsoft once I returned from Germany, and over a few weeks in May I got consensus that we could help the Samba Team by delivering on the roadmap. This included donating software licenses (MSDN Premium subscriptions) to the core team, building a test bed and beginning to share testing tools, preserving the UNIX extensions in CIFS to ensure that the work Jeremy and Steve French were doing would continue to be compatible with Microsoft implementations, accepting Samba Team’s observed bugs in Microsoft’s CIFS implementation and vice versa, providing some technical support on CIFS questions, and sending Microsoft engineers to the CIFS Conference @ Google in September 2007.

About the same time, Tom Hanrahan of IBM’s Linux Technology Center and the OSDL joined my team at Microsoft. His experience in working with Linux – and with Tridge – made it clear that we could sustain the work required to support the roadmap. Apart from his three decades of software engineering and management, one of Tom’s greatest assets is his combination of patience and perseverance; we knew it would take time and progress would be slow, but worthwhile. We’re still early in the process of doing joint testing and engineering with the Samba Team, and have many milestones to achieve (for example, shared test suites & frameworks). Thanks to Tom’s work with key engineers and managers in the company, we have already made progress and are committed to the long term.

Based on the dialog we’d established with Tridge and Jeremy, when the European Commission published the terms that would satisfy them in regards to Microsoft protocols, I saw an opportunity to continue aligning our work with the Samba Team. The terms were good, but the Samba team wanted Microsoft to make some changes to fully conform with the existing practices of the Samba developer community. Jeremy and Tridge saw the opportunity as well, and thus began a 6+ week process of improving and correcting the agreement to arrive at terms that both dramatically expanded their access to protocol information and enabled the Team to continue developing Samba as they have in the past. Attorneys and technologists (always an odd combination) on both sides worked hard to refine the language and do so in a clear and cooperative way. The discussions were masterfully led by Microsoft’s GM of Protocol Programs, Craig Shank (ex-Lineo!) and Samba’s Andrew Tridgell.

Today the Samba Team announced that they’re satisfied with the agreement, and are taking a Work Group Server Protocol Program (WSPP) trade secret and copyright license. This will give them access to Microsoft specifications for the protocols in WSPP (such as file, print, and user and group administrative services) and allow the Samba Team to create, use, and distribute implementations. I expect that this will significantly improve the process of Samba development, and produce better quality interoperation between Windows and Linux/UNIX environments.

What this process has shown me is that if we focus on technology, and patient, diligent execution, we can make real progress together.

This is a historic moment, and one that I’m proud of. But it is only a moment, and now it’s time to get back to working on interoperability, one day at a time.

Cheers,

Sam

[PostIcon:28]

Silverlight on Linux

Sam Ramji — Wed, 05 Sep 2007 10:45:00 GMT

For those of you who have met Miguel de Icaza, you know right away what I mean when I say that he is one of the most energetic people I’ve ever met. Clearly, the whole Moonlight team (whom I haven’t met) would also qualify for this appellation – in roughly 21 days between May and June, they collectively built an alpha implementation of Silverlight on Linux, based on many pieces of the Mono codebase.

After a great deal of work between the Moonlight and .NET teams, we’re ready to formally announce that we (Microsoft and Novell) will be bringing Silverlight to Linux, fully supported and including application and media codec compatibility.

The expansion of the existing work between Microsoft and Novell to include support for Silverlight on all Linux platforms is a major step in the journey of interoperability that we are on. We’ve heard clearly from the community that a full cross-platform web development solution is not only Windows and Macintosh, but must include Linux. I think this is a big deal. While we’ve licensed media codecs before, this represents a fully heterogeneous implementation of a strategic client technology.

Here’s to a better web and support for all users. Hopefully this will help breed further productive conversations about what developers and users need, and in someone else’s famous words, we can all “just get along.”

Cheers,

Sam

Microsoft-Novell Interoperability Lab – Sneak Peek

Sam Ramji — Wed, 14 Feb 2007 14:06:00 GMT

Based on the email I received I would say that many Port 25 readers noticed my post last week on job openings in my new lab. Thank you for your positive responses (and especially the resume submissions)! Brad Cutler, my counterpart at Novell, has been overwhelmed with responses as well, so thank you on his behalf.

I’ve called this a sneak peek because there is much work ahead of us, but it’s time to talk in a little more detail about what the lab will be doing. I and my colleagues at Novell and within Microsoft have been putting in long hours for the last several weeks – nights and weekends as well – detailing the plans for our work together. As you may have seen covered in the news this week (“Microsoft and Novell Announce Collaboration for Customers”), we’ve got a solid long-term plan that covers our cooperation in the following areas:

Virtualization: We will rigorously test the functionality and reliability of SUSE Linux Enterprise Server on the next generation of Windows Server virtualization as well as Longhorn Server on Xen. This focuses on paravirtualization and enlightenments. Here’s a great discussion of Xen paravirtualization and a set of presentations on Windows Server enlightenments. Part of the challenge in delivering enterprise-grade heterogeneous virtualization is in ensuring correct behavior and performance across a broad range of hardware – AMD and Intel, single/dual/quad socket, and single through multi-core CPUs.
Directory and Identity: Directory interoperability is the basis of identity interoperability - directories contain the structure and content that provides the raw material for identity. Through our ongoing testing in the lab, Microsoft and Novell will improve directory and identity interop between Active Directory and eDirectory, using open specifications such as WS-Federation and WS-Security.
Management: We’ll test WS-Management for interop between Microsoft System Center and Novell’s WS-Management implementation, which Novell is developing in the open source community under the openwsman project.

Why are these the most important areas for us to work on?

As part of the Interoperability Customer Executive Council, I heard from the heads of IT from Goldman Sachs, UNICEF, American Express, NATO, and 25 more global organizations that server consolidation is essential in allowing them to reduce costs. In order to fully achieve server consolidation, they need to be able to move their existing workloads – both Windows and Linux – to a common set of server hardware. Without interoperable hypervisors, IT shops would be forced to support two separate sets of hardware, software, and personnel in order to consolidate their servers: one set for Windows and another for Linux. We don’t think that’s good enough.

Hypervisor interoperability is critical, but for this scenario it isn’t enough to deliver the full benefits of virtualization for an enterprise. Once the workloads are running on the same server and the same hypervisor, access control and authorization needs to work across the entire environment consistently – otherwise you’re just shifting the interop problem up the stack, only to suffer later. This is where WS-Federation is essential – implementing an open specification to federate identity between existing directory servers enables you to have consistent security policies across your heterogeneous workloads. This is a continuation of work we’ve done with IBM, Apache, Ping Identity and SXIP Identity.

Operations relies on strong management tools to provide availability and reliability across a broad server environment. Ops teams typically have training on specific toolsets to monitor, administrate, and manage their infrastructure. Realistically, moving Windows and Linux workloads onto the same set of servers requires that existing management tools be extended to the new environment. We believe (as do HP, IBM, BMC, CA, and many others) that WS-Management is the solution. Implementing this open specification will enable servers, applications, and services to communicate with management consoles from multiple vendors.

I’ve had a few people approach me about this project who pontificated “If you [Microsoft] would just implement the specifications as they’re written, you wouldn’t have to do all this work!” In fact, this is an incorrect understanding of software engineering and interoperability. Making protocols truly interoperate in every realistic circumstance is one of the great challenges in engineering. In real life, you have to implement the specification correctly – and then the work begins. Were there platform-specific assumptions in the code (as basic as big-endian vs. little-endian format)? Were there parts of the spec that were subject to interpretation? Due to the extensive development and testing embedded in technologies like TCP/IP and HTTP, it’s easy to forget that it took years of work by many parties to deliver what we now take for granted.

This work across virtualization, identity, and management is a pretty awesome undertaking, and I expect that as we continue to progress here we’ll discover new things we need to do in order to deliver interoperable computing. I’m looking forward to reporting on it here, and have submitted a presentation abstract for OSCON ’07 to walk through the Joint Interoperability Lab’s operations in detail. Hopefully I can shed a little light on what makes interoperability so challenging, even in an age of open specifications.

Cheers,

Sam

SML Update

MichaelF — Mon, 15 Jan 2007 20:01:00 GMT

Back in August we posted a podcast with Praerit Garg regarding the announcement of the Service Modeling Language draft specification put together by Microsoft and a number of other leading technology companies. We wanted to follow-up and provide an update on the specification and the workgroup.

On September 12th the public feedback workshop was held and a good deal of feedback was provided both by community members in attendance and by those submitting feedback via email. One of the key topics was the name of this language as many felt the SML title didn’t full capture the intent or capabilities of the specification. Pratul Dublish, Senior Program Manager at Microsoft, has a blog entry regarding this discussion here (along with a number of entries regarding SML and the Working Group).

The Working Group has published the second draft of the SML specification and the first draft of the SML Interchange Format specification. It has also announced an Interoperability Workshop for interoperability testing between different implementations of the specifications. The workshop is open to companies and individuals willing to bring an implementation of the latest published specifications to the workshop. The workshop will be held during January 16-17, 2007 in Austin, Texas. The invitation for this event can be found here.

Per the terms provided in the specification, there is nothing that prevents an Open Source project from implementing the SML Spec. Eclipse has started a project called COSMOS which implements the SML runtime, modeling tools for SML, and the infrastructure to enable the use of SML for model based management. Take a look at: http://www.eclipse.org/proposals/cosmos for details.

The URL for the SML Working Group is: http://www.serviceml.org, take a look at this site if you are interested in learning more about the specifications, the Working Group’s activities, and the Interoperability Workshop.

We will continue to keep the Port 25 community up to date on the progress of the SML Working Group and the evolution of the specification. If you need help connecting with the group, please let us know.

Please take a minute to provide feedback to the Working Group if this is a topic of interest either personally or professionally.

-Sam

Sendmail on Sender-ID

Sam Ramji — Mon, 20 Nov 2006 17:03:00 GMT

Eric Allman (founder of Sendmail, inventor of Allman notation, original hacker) graciously took some time with me to discuss his take on Microsoft’s opening of the Sender-ID specification. Thank you, Eric!

While privacy is important for email, so is reducing spam. A few years ago, Microsoft established an approach defined as Sender-ID which is in principal similar to a modern phishing filter – ensure that the email’s sending address correlates with its IP address of origin. If not, there is probably something wrong.

While the specification was a good one, the approach to licensing the specification was not. Since billions of emails are sent each day via open source technologies, a comprehensive solution to spam has to include open source MTAs and developers’ rights to use the solution. Microsoft has corrected this error by opening the Sender-ID specification under the OSP (Open Specification Promise) – a free and open guarantee that the specification can be used by anyone.

Eric was very complimentary of the move to the OSP, and believes that Sender-ID and DKIM (Domain Keys Identified Mail – see here for an overview) can and should coexist.

Listen to the podcast to get Eric’s insights first-hand.

Cheers,

Sam

Interoperability: Open Source ODF/Open XML Translator and Microsoft Office

jcannon — Thu, 06 Jul 2006 19:42:00 GMT

Linux Format reported on Port 25 recently with the tagline “Reports of snowballs seen in hell as Microsoft offers to work with Linux developers,” which I thought was funny. It’s apparently getting even colder down there as we’ve now announced an open source project that adds support for ODF to Microsoft Word 2007 ("Microsoft Expands Document Interoperability").

A few months ago I started working with Jean Paoli, whose leadership on Interoperability at Microsoft is steadily moving product teams toward the goal of consistently delivering high-quality interop. Brian Jones notes this in his blog but doesn’t call out Jean by name. You can be sure that you’ll see more of Jean’s handiwork in the coming months and years.

During the time I’ve worked with him I’ve been greatly encouraged by his commitment to openness in documentation and in implementation. The Open XML Translator project is a great example of this – it’s an open source project hosted on Sourceforge.

I couldn’t help but hop over to Slashdot and check out the reactions to the news – and as usual there was a mixture of the rational and irrational, hope and fear, insight and suspicion of conspiracy. It’s worth making one point over and over.

The Open XML Translator is an Open Source project.
The Open XML Translator is an Open Source project.
The Open XML Translator is an Open Source project.

By definition it can’t conceal its implementation, is open to experimentation, modification, and commercialization (it uses a BSD license), and is owned by the community.

If you think it needs improvement, then improve it. If you think it doesn’t matter, ignore it. But above all, really think about it and what it means that we’ve taken this step before reacting reflexively.

This is actually something new and different.

Hats off to JBoss

admin — Fri, 14 Apr 2006 17:03:00 GMT

Congratulations to Marc Fleury and team for their success this week in selling their company to a like-minded partner that pioneered Open Source business models.

Running a startup is hard. Keeping it going and focused is harder. Selling it while maintaining your principles is nearly impossible – but JBoss has scored a hat trick. From 97-2001, I was in a string of startups in Silicon Valley, and I know first hand just how hard it is to make them work.

When a company reaches the level of publicity and success that JBoss has achieved, there are pressures to compromise in many dimensions – exemplified by the rumored half-billion dollars that “a large database company” offered them only last week. Marc could have taken that and walked out of the industry altogether – enjoying a comfortable life on his own private island. Instead, he has chosen to marry his company to a long-term resident of the Commercial Open Source Software industry that will sustain the JBoss principle of interoperability and heterogeneous support, and is well-aligned to evolve the business model in a pure Open Source tradition.

My lab is working with JBoss on the technical collaboration project that Microsoft announced earlier this year – good, interesting work – and will be exploring new areas to work on interoperability, such as JBossWeb, an Apache-based web server with native support for .NET, Java, and PHP.

We’re continuing to read and sort the suggestions we’re getting – including emails – and looking for clusters of requests that we can turn in to projects. What would you like to see us do from a Java interoperability standpoint? What would be most valuable for you?

Cheers,
Sam