The Open Road

Microsoft is launching an open-source foundation. Google is promising to keep user data portable. Both moves seem to cut against the financial self-interest of the two technology giants. Have the gods gone crazy, or are the business strategies of the industry's biggest players more subtle than "Embrace. Extend. Extinguish"?

With a steady adoption of open-source business and development strategies, Microsoft has gone from open-source hater to open-source embracer in just a couple of years:

• Created its own open-source foundation, the CodePlex Foundation.
• Launched CodePlex, an open-source project-hosting site.
• Started actively contributing to outside open-source projects, including those of the Apache Software Foundation, which it also financially supports.
• Embeds open-source code such as JQuery into its own products.
• Releases complementary open-source projects to augment its customer relationship management software, SharePoint, and other products.
• Received "unanimous approval" by the Open Source Initiative for a few open-source licenses.

This isn't to whitewash all that Microsoft has not done well vis-a-vis open source (e.g., I'm not a fan of its patent-licensing arrangements, including the "interoperability" agreement with Novell), but clearly, Microsoft has been actively adopting open source as part of its business strategy. I'll address the "Why?" question below.

Google, for its part, has long supported open-source software. And it's easy to see why: the company makes its money from data, not software. The more people that have access to a great Web experience through Firefox or Chrome, or have computer access through low-cost Chrome OS-based Netbooks, the better, as they'll almost inevitably find their way to data-rich services from Google.

Google, in other words, has a strong interest in promoting open source and closed data.

All of this makes Google's Data Liberation Front--"an engineering team at Google whose singular goal is to make it easier for users to move their data in and out of Google products--so intriguing. The DLF appears to be giving away Google's single best option for monetizing its user base.

(Credit: Google)

What is Google thinking? One answer may be that Google is trying to head off government scrutiny and intervention. As CNET News' Tom Krazit posits, "anything Google can do to show that it isn't planning to create an impenetrable fortress surrounding user data, it's going to do."

That's one cynical and likely accurate view. But I think that there's more to the story.

Google has created an array of services that increasingly dominate their respective markets. Consumers and businesses are apparently very happy to give more of their time and attention to Google products.

As such, Google's primary concern revolves around keeping those users from leaving. While the DLF makes it easier for customers to leave Google, it also obviates the need to do so. So long as Google customers feel sure that they can leave on their own terms, they likely won't.

Microsoft is starting to learn the same thing. Its customers tend to use Microsoft products because they work, not because some evil genius in Redmond dreamed up diabolical ways to keep them locked in through closed file formats.

Don't believe me? Look at Microsoft's support for CMIS (Content Management Interoperability Services), a new content standard that promises to do for content management systems what SQL did for the database market. CMIS enables information portability between different content repositories. (Disclosure: Alfresco, my employer, was a founding member of CMIS, along with IBM, Microsoft, EMC, and others.)

In other words, CMIS makes it easy to move content out of SharePoint into, say, Documentum. It also enables application vendors to write to the CMIS standard, rather than specifically to SharePoint.

CMIS Interoperability Standard

(Credit: Microsoft, EMC, IBM)

Microsoft has been actively engaged in drafting the CMIS specification and appears to be a strong proponent of it. Why? Why would Microsoft, which has much to gain from SharePoint being the center of a new lock-in strategy, support an open standard that makes it easy to move content out of SharePoint and into competing repositories?

Because Microsoft knows that it can win.

Take Microsoft's pre-CMIS partnership with Documentum. As CMS Watch anecdotally references, SharePoint is much easier to use than Documentum, making any partnership/integration between the two a largely one-way street from Documentum to SharePoint, just one reason that SharePoint has boomed, even as the economy has busted. This is only going to get better for Microsoft with CMIS interoperability.

Interoperability favors the vendor whose products are easier to use. By opening up, Microsoft is opening its doors to more customers and, hence, more money.

Google and Microsoft aren't supporting open source or open standards or open data because they grew up as Boy Scouts or Girls Scouts, and feel that it's the right thing to do.

Rather, they're increasingly engaged in open business strategies because they recognize the financial rewards that can stem from doing so. Openness is not a religion; it's a business strategy--a strategy that Microsoft and Google are learning to play too.

Open-source advocates need to get their stories straight. Are we a big-tent movement, or a parochial club that is hell-bent on limiting membership...and efficacy? Unfortunately, it increasingly seems that the open-source community is determined to be the latter, and has taken positions on various events that are out of keeping with the founding principles of open source.

Take Microsoft. The company has long been a controversial figure in open source, as well as in the broader technology industry, and for good reason. Conviction for abusing monopoly power will do that to you.

But Microsoft has spent the past few years extending an olive branch to the open-source community, only to be criticized, questioned, and rebuffed. Last week the software giant created the CodePlex Foundation, designed "to enable the exchange of code and understanding among software companies and open-source communities." The foundation has assembled a solid core of directors and advisors, including Stephen Walli and Monty Widenius (formerly of MySQL).

"I don't believe it for a minute," thunders Steven J. Vaughan-Nichols, speculating that "Microsoft doesn't mind stealing from open source, but any deals it makes are only good while there's a clear, short-term benefit to Microsoft." This is probably true, but it's equally true of every profit-seeking company on this planet, minus the emotionally-charged (and inaccurate) term "steal."

Do Google, IBM, SAP, Jive, MySQL, Red Hat, etc., use open source to advance their self-interest? You bet they do. They have a fiduciary duty to do exactly that. And do they selectively adopt open source without always contributing back? Of course.

Let's face it: no one--even open-source community contributors--writes open-source software purely out of the goodness of their hearts. This shouldn't be a surprise: open-source luminary Eric Raymond wrote about it over a decade ago.

But it's telling that Microsoft is the company singled out, more often than not. It is very likely that Microsoft could open source every line of its code and still be treated like a pariah by the open-source community.

Or, rather, by a vocal segment of that community. It's the part that doesn't have to meet a payroll. Perhaps the sort of person that Hugh MacLeod was referencing in a tweet he made: "It's easy to spot a purist. They're the ones without any skin in the game."

When your only job is to yell down others, you don't need to pay much attention to what you say.

But Microsoft isn't the only company to get pilloried. Oracle gets its fair share of abuse, too, and often enough on this blog. It seems that abuse is proportional to one's income statement, and the potential to abuse one's market position to the detriment of customers, as BusinessWeek recently wrote of Oracle.

Hence, ever since Oracle announced its intention to acquire Sun and, hence, MySQL, some within the open-source community have been wringing their hands at a frenetic pace. "Oracle will kill MySQL!" they moan.

Gartner rubbishes this concern, insisting that "the fact of the matter is (Oracle) cannot destroy the (MySQL) product." It's licensed under the GNU General Public License, after all, which preserves the freedom to fork the code. In fact, MySQL has been forked several times already.

This isn't to suggest that Oracle couldn't damage MySQL by slowing its development, or shutting down internal development altogether. Of course it could: Sun/MySQL employs the overwhelming majority of developers who write MySQL. To control them is to control the code.

But if there's any truth to open source's claims that it provides freedom (through the right to fork), then owning MySQL, the company, shouldn't be tantamount to owning MySQL, the code, and Monty Widenius, and others could merrily pick up where Oracle left off.

That is, assuming we really believe open source is a liberating force. Do we?

I do. That's why I don't worry about Oracle's impact on MySQL. Heck, I figure Red Hat or someone would simply hire the MySQL engineers and start MySQL II if Oracle attempts to kill the project (which I don't think it has any intention to do).

It's also why I welcome, not reject, Microsoft's attempts to open itself to open source. Those with no skin in the game find it easy to point fingers and malign others' imperfect attempts to engage. They forget that it's hard for closed-source companies to open up, as SAP's Dirk Riehle writes, but with the incentives to open up increasingly visible, companies will find a way.

We should be encouraging them to do so, not second-guessing their every move. And we should recognize that there are times when the open-source alternative is not ready to displace a proprietary incumbent, as Esther Schindler notes, which means that we're going to need to learn to get along for many, many years.

I'm not suggesting that Microsoft or Oracle has been perfect. But don't believe that IBM, Red Hat, Alfresco, MySQL, or (insert vendor of choice) has been perfect, either. Each of us is making this up as we go along. Sometimes we screw up, but that doesn't mean it's intentional.

Jeff Schick, IBM

(Credit: Jeffrey Gluck, IBM)

Most vendors must guess what customers want to buy, and how they'll use it. For IBM, however, with about 400,000 employees, it has the potential to be its own best laboratory, one that becomes even more potent when mixed with active participation in open-source communities.

That potential, as I discovered in an interview on Friday with Jeff Schick, IBM's vice president of social software, isn't a "gimme," but is powerful if you can enable the right sort of corporate culture and processes.

For example, Schick mentioned that IBM has a technology adoption program for employees that spans the gamut of new products, add-ons and patches to existing products, and still-raw technologies direct from IBM's labs. While the invitation list and process is different for each particular item, IBM generally encourages its product groups to "experiment" upon each other. The earlier in the development process, the better.

At the heart of this open approach to technology adoption are open standards and open source. When I pressed Schick on the relative importance of both ("If you could only choose open standards or open source, which would it be?"), he responded:

Our products may include open-source components, and often do, but ultimately open standards are the most important consideration for customers. As customers integrate our products into their various enterprise systems, open standards are critical for ensuring they work.

Point taken, but it's impressive just how much open source influences IBM's product development. Gartner estimates that 80 percent of commercial applications will include open-source components by 2012. At IBM, the number may even be higher.

Despite IBM not releasing its core software products under open-source licenses, Schick noted just how integral open source is to IBM:

From a development perspective, as we build our social software products in Lotus, we're always looking at ways to improve quality and time-to-market. Open source often helps us with both areas.

For example, we were blogging within IBM for a long time before deciding to build the Lotus Connections product, which is fast approaching hundreds of millions of users. After some study, we decided to build the blogging piece of Lotus Connections using the Apache Roller project, an open-source Java blog software. We have become active contributors to the project since then.

But it's not just in Lotus Connections. As you look across nearly every capability across our social-software strategy, open source plays a critical role. Open source is an integral part of how we build products. Our engineers are very much in tune with the wide variety of open-source components that are available to them, and use and contribute to them. Regularly.

IBM seems to have figured out better than most how to marry the global open-source laboratory with a massive internal laboratory. Talking to Schick, there appears to be a very blurry line between "internal" development and "external" development, giving the company a significant advantage over proprietary (Microsoft) and open-source (Liferay, Open-Xchange) competitors alike.

Some competitors may be able to match IBM's scale, but few to none have managed to marry internal scale (employees) with the power of external scale (open-source communities) in the way that IBM has.

Follow me on Twitter @mjasay.

Though our bodies get older, our minds remain relatively young. Sure, we're scarred and matured by experiences, but our bodies age much faster than our minds. It turns out that companies are much the same.

Take Oracle, for example. We sometimes give Oracle grief for being the quintessentially Machiavellian company, with a hard-driving sales culture and bent on nefarious designs to lock in customers, but the company was founded under very different principles.

Oracle made its fortune promoting the SQL standard which, despite its problems, freed the world from mainframe lock-in, as Alfresco CEO John Powell, an early Oracle employee, reminded me recently. (Disclosure: John is my boss.)

Prior to Oracle, if you wanted to write database technologies, your choices were IBM's IMS, Cullinet's IDMS, or other proprietary solutions that were locked to specific mainframe hardware and the application was locked to the data.

Oracle (and IBM) opened up the market with SQL-based relational databases, thereby allowing independence between data and their associated applications. Oracle's message was "freedom of customers from mainframe lock-in." Starting with the VAX, Oracle gave customers freedom to negotiate between different mini-computer hardware suppliers.

Oracle was, in other words, the open-source vendor of its day, delivering customer choice.

Oracle has since become a massive corporation, and attracts all the suspicion that success often breeds. But perhaps its soul (early employees) is still young and concerned with openness, even if its body (the infrastructure) may not be.

It's especially intriguing for me to watch one of Oracle's longest-serving employees, Ken Jacobs (Employee number 18), take on increased responsibilities within Oracle's open source-related businesses. Jacobs has been involved with InnoDB, Oracle's first foray into MySQL, and it's likely that he'll play a big role in managing the company's MySQL business, too.

This may well be the perfect fit for Jacobs: he grew up touting Oracle as a freedom fighter. Now he gets to do it again, at a time when the industry sees Oracle very differently than when Jacobs started at Oracle over two decades ago.

Follow me on Twitter @mjasay.

It turns out that IBM is the company behind the new Open Cloud Manifesto, a document that defines an open, interoperable cloud-computing vision that is so much motherhood and apple pie that few will disagree with its hazy tenets.

However, it's increasingly evident that IBM wants to use the manifesto to rain on Microsoft's own cloud efforts, inviting Microsoft to join the open-cloud party on terms that Microsoft, as well as Amazon.com, Salesforce.com, and Google apparently can't countenance, as ZDNet's Mary Jo Foley reports.

Because of the closed process, as well as some of the tenets of the manifesto, those companies--the world's primary cloud vendors--have thus far declined to sign up. If these bellwethers of cloud computing don't sign, will it matter that a host of small start-ups are set to join Cisco Systems and IBM in signing it? The answer, as Daryl Taft suggests, is "probably not."

It's one thing to be open. It's quite another to actually get used. Open clouds that no one uses are, well, somewhat useless.

It's too bad, because arguably, the Open Cloud Manifesto could have garnered broader support simply by opening up the drafting process, probably resulting in a more inclusive document.

CNET's Ina Fried takes a look at the document and suggests that "it's easy to see how this might prove challenging for those with existing cloud platforms, folks like Amazon and Microsoft," given its emphasis on open source to undergird open clouds.

IBM seems to have crafted the Open Cloud Manifesto in its own image, and then foisted it upon everyone else.

Thus, as ZDNet's Larry Dignan suggests, the manifesto has something of an anti-Microsoft bias, and IBM, which has no great love for Microsoft, almost certainly intended this. In fact, IBM, master of public relations that it is, probably never really intended Microsoft to participate.

All of this makes me wonder if IBM simply meant for the manifesto to be a PR wedge to beat up Microsoft. All's fair in love and standards, and no one uses standards to greater advantage than IBM, but its arguably less-than-open approach to creating an Open Cloud Manifesto could doom the manifesto from the start.

I like the manifesto's tenets, and I agree that without cloud interoperability and open data, we're going to end up re-creating the past two decades' proprietary desktop wars in the cloud. But unless we're willing to not only articulate open principles, but also to create them through an open process, we seem to be doomed to repeat the mistakes of our proprietary past.

Follow me on Twitter at mjasay.

Open standards are good for an industry's development as they tend to create much more customer demand. In the growing cloud-computing market, standards are also critical to ensuring customers don't get locked into any particular cloud provider. Ironically, this is almost certainly why we're unlikely to see open standards widely adopted in cloud computing any time soon.

ZDNet notes that cloud-computing vendors are increasingly talking the open-standards talk. The problem is that each company, in a Prisoner's Dilemma sort of way, has an incentive to maximize lock-in of its customers while simultaneously encouraging open standards for everyone else.

The cloud-computing market, in other words, might be better off for open standards, but what's in it for me?

It's probably not realistic to expect a market leader to embrace open standards. I wouldn't, therefore, look to Salesforce.com for leadership on open standards, unless they're standards that tilt the field in its favor.

It is possible, however, that we could see a newcomer, perhaps Sun, challenge the incumbents with truly open standards. Now would be a good time, given that there is no overwhelmingly dominant cloud vendor.

I'm not suggesting that Sun has a perfect track record when it comes to open standards, but openness favors the insurgent, and Sun is a credible insurgent in the cloud-computing space, despite its heft in the server market. But be it Sun or some other company, I'm hoping a credible contender will drive truly open standards for the cloud-computing industry, or we face a few more decades of more proprietary lock-in.

Follow me on Twitter at mjasay.

On March 5 Apple dropped a small bombshell on the World Wide Web Consortium (W3C) standards body, excluding one of its patents from the W3C Royalty-Free License commitment of the W3C Patent Policy for Widgets 1.0. The patent in question covers automatic updates to a client computer in a networked operating environment.

The announcement has generated no apparent response, yet could portend serious consequences. The Apple exclusion could mean that a W3C standard on widgets (or, really, any standard in the Web Application Group) at W3C that uses or includes something which touches this patent will either need to negotiate directly with Apple for rights, must pull the code out of publication until such time as a work-around to replace said functionality is created, or be de-published entirely.

None of these options is particularly appealing.

Apple gets a lot of credit in the open-source community, but this move, while understandable from the standpoint of responsible guardianship of intellectual property, shouldn't win it many friends.

If anyone can elaborate on the significance of this move by Apple, please do so in the comments below or by sending me an email. It's possible that this is more smoke than fire.

Follow me on Twitter at mjasay.

By open sourcing Java Sun probably thought it had earned 'brownie points' with the open-source community. Instead, as a recent vote over Java Enterprise Edition 6.0 suggests, the community may be intent on teaching Sun that there's more to 'open' than source code, with the Apache Software Foundation (voted against) and SpringSource (abstained) siding against the pending Java specification, as reported by Gavin Clarke in The Register.

Apache is the only member of the 16-member Java Community Process to vote against the 6.0 specification, arguing that Sun "has breached the terms of the JCP's Java Specification Participation Agreement (JSPA), by inserting terms into the licensing of the all-important TCKs that limit uptake of Java," but it did so for good, albeit self-interested, reasons:

[Apache representative Geir] Magnusson claimed today, as the did two years back, that Sun's refusal to change its licensing terms is damaging Apache's Project Harmony, an open-source implementation of Java SE. Harmony had a modular, OSGi implementation in Java SE before Sun's official Java SE. Without a TCK, Apache cannot prove Harmony complies to Sun's Java specification, meaning people will shy away from using it.

"Sun's doing tremendous damage to the project," Magnussen said.

It looks like Sun's rigorous adherence to the law is also hurting the one thing the company holds so dear: compatibility. Google's Android uses Harmony's class libraries.

Sun's tight control of Java begs for opening up. As noted by The Register, Sun's goal to maximize compatibility is almost certainly hurt by its efforts to control the standard. It's a battle that Sun has fought for a decade, with fierce debates springing up at the O'Reilly Open Source Convention and elsewhere, but it's not a problem that is going away anytime soon.

I'm supportive of Sun's desire to both profit from and guide the development of Java. But Sun can't have it both ways, as the 6.0 specification vote demonstrates. Control and community don't go nicely together, unless that control is exercised by the community.

Sun has to choose how it wants Java to succeed. If it continues to upset key Java community members like Apache, it will fail to drive the kind of Java adoption it desires.

Last week, Tim O'Reilly called for Amazon.com to open up its Kindle e-reader, or "Amazon will wind up another online pioneer who ends up a belated guest at the party it planned to host."

On Wednesday, Amazon demonstrated that it understands the value of openness, even if it's not yet prepared to embrace open standards for the Kindle, by providing an iPhone application that enables users to read their Kindle content on Apple's iPhone, as CNET reports.

This is a shrewd move. It's unlikely that many will want to trade the Kindle reading experience for the iPhone's, but it should prove a useful complement that drives more Amazon revenue.

As Mozilla's John Lilly opines, the iPhone Kindle application is "useful, if I'm somewhere and forgot my Kindle...and I'm sure that I'll buy books with it to read a snippet, then really read on my Kindle."

In sum, by providing a Kindle for iPhone application, Amazon has opened up a compelling complement to its Kindle device, one that will likely feed more revenue to Amazon while simultaneously crippling rivals' efforts to build a critical mass of iPhone e-book readers.

Genius.