As Microsoft prepares to launch Windows 7, the software giant is wisely describing Windows 7’s connectivity to Windows Server, Exchange Server and other popular back-end solutions. But is it time for Microsoft to think outside the Redmond box, and talk about Windows 7’s connectivity to Linux servers? Plus, what about Windows 7’s ability to run open source applications?
Let The VAR Guy set the scene: He’s at Microsoft’s Worldwide Partner Conference 2009 (WPC09) in New Orleans. There’s a reasonable amount of buzz for Windows 7. Faster load times. Better hardware support. Windows 7 certainly seems to be everything that Windows Vista should have been. Yes, Windows-centric businesses should eventually migrate to Windows 7.
But here’s a rather timely question: Where are the killer applications for Windows 7? Fact is, more and more of the world’s fastest-growing software companies focus on either (A) open source or (B) software as a service (SaaS). And many of those SaaS instances themselves run on Linux-centric servers. Meanwhile, Microsoft is promoting Corel as a major ISV supporting Windows 7. Hmm…
Open to Open Source?
So, the big question for Microsoft: Is it time to start marketing Windows 7’s easy connectivity to Linux servers? Plus, is it time to market Windows 7’s ability to run a range of open source applications?
Rewind to the 1990s, and Microsoft wisely evangelized Windows 95’s connectivity to Novell NetWare servers. Fast forward to the present, and Microsoft should do the same with Linux servers and open source.
Sound far-fetched? Microsoft already helps to certify open source applications on Windows Servers. The software giant should do the same on the desktop.
Where’s Novell?
To be sure, the Microsoft Worldwide Partner Conference 2009 is filled with thousands of loyal VARs and solutions providers. But it’s strange to see such a massive software gathering without any open source ISVs on hand.
Come to think of it: Novell — which works closely with Microsoft on Windows/Linux server integration and virtualization strategies — is not listed among the event sponsors. Our resident blogger needs to check in with Novell to see if the company is at the conference.
Follow The VAR Guy via RSS; Facebook; Identi.ca; Twitter; and via his Newsletter; Webcasts and Resource Center.
Read More About This Topic
Share This Post
Tags: Microsoft Linux Servers | Microsoft Worldwide Partner Conference 2009 | Windows 7 | Windows 7 Linux Servers | Windows 7 Novell | Windows 7 Open Source Applications | Windows 7 SUSE Linux | WPC09
Interact: Add a Comment | Trackback Link | Permalink
Subscribe: RSS Feed
“So, the big question for Microsoft: Is it time to start marketing Windows 7’s easy connectivity to Linux servers?”
Doubt it since this would be ceding ground on the gains made by tying their desktop to the server with the workgroup protocols fiasco. Microsoft got spanked pretty hard by the EU Competition Commission and eventually threw in the towel.
The best option is to support the upcoming version of Samba since the EU agreement forces Microsoft to play nice. It’s better to have interoperability on those terms then on Microsoft’s terms alone.
Warp99: Thanks for offering your perspectives. The VAR Guy still thinks it’s time for Microsoft to talk a bit about open source app support on Windows. But our resident blogger appreciates the reader feedback.
Microsoft needs all the help it can get. The downturn has hit proprietary software companies particularly hard, while open-source ones seem barely touched. It makes sense to put ideology to one side and go where the profit is.
Whether Microsoft’s top management understand what makes sense or not is another matter.
Lawrence:
It’s really not fair to talk about “open source companies” so glibly. There are very very few publicly traded open source companies for which we have publicly available financials for. Most “open source companies” are still in start up mode or otherwise still reliant on venture capital funding. Which “open source companies” have released enough details about their financials to know whether a quarter was a net money maker? I see a lot of press releases about sales growth..but very little disclosure about corresponding expenses. You need both revenue and expenses to see where the profit is or is not.
Out of the publicly traded companies that could be stood up as an “open source company” Red Hat is singular (or nearly so) as being able to show a profit. It certainly not clear to me that Red Hat is representative of open source businesses more generally…there just isn’t enough financial information available to be certain. It could be that Red Hat is bucking the trend even among open source companies.
-jef
Jef: Well stated. The VAR Guy was thinking about those very thoughts earlier today. On the one hand, the traditional closed source model seems to be under attack from open source and SaaS. On the other hand, Red Hat is the only large, independent, public, extremely successful open source company in the world.
Then, you’ve got fast-growing public SaaS companies (Salesforce.com, RightNow, NetSuite). Some are profitable. Some aren’t. Most depend on their own proprietary software platforms.
Talk about confusing but exciting times in the software space.
The Var Guy:
Most SaaS businesses depend on proprietary software platforms? I don’t think you can say that either. They could have completely BSD or GPL code running on their iron and be a completely open source shop according to the OSI definition and you’d never know it until they told you. And they aren’t telling. SaaS as a model makes it harder to draw a bright line between proprietary and open. Its more than a new business model, its a cultural shift.
It’s actually more than that…SaaS is a frontier…it is a gold rush…its a land grab. It’s as scary as it is exciting. Because those sort of rushes to exploit the new market/resource is full of irrational exuberance. The only thing you can know for sure about SaaS is this… the majority of companies who are currently spinning up a SaaS business…will fail. That’s a guarantee. Until there is a clearly successful business model that people can copy its going to be intense phase of trial AND error.
-jef
Jef: Unless you can take the application and move it to another service provider, The VAR Guy considers it proprietary. Can you take Salesforce.com and run it elsewhere? Can you take NetSuite and run it elsewhere? Can you take RightNow and run it elsewhere?… … …
The VAR Guy:
Ah, sorry. Proprietary from a data lock-in sense not from an open/closed source definition sense. tsk tsk, you subtly change the subject halfway through the conversation. Lawrence was using proprietary as opposite of open-source. Service data lock-in is orthogonal to open-source. Any time you contract with any service provider you run the risk of losing control of your data running on their servers.
I think in practise is that what you find that most web services are going to have proprietary aspects using your definition. Even web services with AGPL codebases…its doubtful that competitors using the same codebase will provide exactly the same services even when you can export data from one service to another. Your definition for proprietary assumes that multiple providers will offer the same service featuresets. I don’t necessarily see that happening..until a massive die-off and consolidation in SaaS providers happens. Even then competitive differentiation of featuresets will still be happening.
You’re also assuming that services will implement and allow export and import of user data. Even AGPL licensed codebases don’t require service providers to expose that sort of data fluidity. Being able to export data is only going to happen if customers start demanding it. I don’t see any sign of that happen enmasse. I think that’s going to require legislation as the SaaS matures.
-jef
Jef: Sorry The VAR Guy subtly switched subjects. Your points are well taken. Yes, customers are not demanding the export data feature yet. Our resident blogger needs to concede that point to you.
Novell is nowhere to be seen because it does not benefit Microsoft. The relationship is one sided – what Microsoft can get into Linux.
As he first commenter stated, MS does not want to be seen to be ceding ground to FOSS.
“Microsoft wisely evangelized Windows 95’s connectivity to Novell NetWare servers”
Not quite, the sequence went:
Microsoft releases their Advanced Server.
Bill Gates announces that ‘the next version of MS-DOS/Windows 3′ may not support Netware.
Novell decides that it needs a client OS that will _always_ support Netware so buys DRI for DR-DOS.
Novell enhances DR-DOS 7 to multi-task and plans to include n x DR-DOS 7 licences with each Novell server n x user licence.
Microsoft and Novell have talks and agree: DR-DOS gets dropped, MS to support Netware and announce this.
The next step was that MS would sell 100 client Advanced Server that would act as a proxy to a 25 user Novell Netware Server such that 100 user MS-AS + 25 user Netware was less that the cost of 100 user NetWare but the proxying gave the effect of the latter.
Jef, Var Guy: Most “open source companies” are not software houses, they are corporate employees working on in-house infrastructure projects that also happen to be open source projects. Think “part time corporate subsidiary” of parent companies that do not sell software. As a concrete example the people who are writing Apache are ISP employees.
Trying to gauge the economic impact of open source by looking at “publicly held open source companies” is never going to make sense.
karl:
You are broadening the discussion. So far this isn’t about gauging broad economic impact. This discussion is specifically about an apples to apples comparison of the financial performance of software companies.
Apples to apples is all I’m saying. If Lawrence wants to stand up companies who are open source companies that make for a reasonable comparison to MS..he’s got very few examples to choose from.
I am NOT saying that you can measure the broad impact of open source development by just looking at public companies. What I am saying is you can’t take all the glowing sales and deployment numbers we are hearing from some private companies like Alfreso (just for example, not picking on them) and compare it to the financial reporting of a publicly traded company like Microsoft or Red Hat. It’s not a fair comparison. As much as I want to believe that many “open source” software companies are profitable I’m not going to let my desire for that reality to supplant the need for objective financial information.
That being said, to do what you are proposing and to try to take a clear look at impact of open versus closed software development in a broad sense…across companies which are not traditional software development houses but who develop software secondarily to their business mission…you’ll need far more granularity in financials than even public traded companies are required to expose. You are absolutely right, there’s a lot of people across a wide range of companies who are contributing to open source project codebases because it enhances their ability to be more effective in their day jobs. No question about it. But you aren’t going to easily capture that impact in financial statements.
At best you can do a set of case studies. And I’ve personally never trust case studies to really give me a sense of the mean or median trend. I tend to view case studies as a way to capture best and worst cases and not an objective sampling of the average.
-jef
Jef:
Yup. I certainly continued the conversational drift and wound up far from the original post.
I am annoyed with the metrics spewn about that measure FOSS in the same way that one would measure a traditional corporate product. The unstated assumption is always that these reports should be given equal weight with reports on corporate products.
The best that can be done is to apply proper statistical methods to a random survey. If you’re talking finance it would be a survey of people who could quantify the numbers. This is quite expensive, I guess, and the people who pay for such things are rarely willing to share the results — although occasionally (http://www.osnews.com/story/21035/Ballmer_Linux_Bigger_Competitor_than_Apple) the results are made public. Never though, with a clear explanation of what’s actually measured.
[...] was also not much about SUSE. In the context of GNU/Linux servers, the Var Guy mentioned Novell as follows: Rewind to the 1990s, and Microsoft wisely evangelized [...]