Information Manager Journal

Virtual battle

At last, the fundamental technology battle of our time is at hand. Will you want a virtual version of your Web stuff on your desktop or your handheld device -- the way Google will serve it up -- or will you want a virtual version of your desktop or handheld device on the Web -- the way Microsoft will serve it up?

I'm betting on Microsoft's approach, but only because Google's privacy assurances aren't very convincing, and the Web is subject to reliability problems, as anyone who's ever tried to watch a Netflix movie online can confirm. But millions of others will opt for Google, because of the continuing headaches of deploying and patching desktops like Windows.

Finally, which side will Apple take?

Pay to play

This trend is bound to give Microsoft a huge seat at the open source table, like it or not.

MySQL anti-patent page update

Updating my earlier post, the MySQL "public patent policy" page now says:

"The MySQL patent program is in the process of being integrated into the Sun patent program.  In general, Sun uses its patent portfolio to defend communities and indemnify customers.  Check back here for further updates.

"In the meantime, check out the blogs of Sun's CEO, Jonathan Schwartz, and CTO, Greg Papadapoulos, for Sun's view on software patents."

In the spirit of open source

In the spirit of open source, I'm going to give back to a community and industry that helped me labor to write my (now abandoned) book. Starting today, I'll spend a few minutes a day posting some notes from my book, really just quotes and links, with an occasional comment here and there. I hope you enjoy them.

Declaring blog & book bankruptcy

This is hard to write.

Every day, people declare "blog bankruptcy," stating that they are no longer able to keep up with the blogs they're reading. I'm declaring that today. But there's a bigger declaration in this case.

For six years, I've been laboring to write a book. The research became a monster that took over and choked off my common sense. Rather than trying to limit the scope of the book to a discussion of software development trends back then, I went for a grand vision, trying to sum up three megatrends: what was happening to open up computing to provide more choice, what was happening to try to thwart that, and how security concerns threatened the current and emerging models of computing.

After five years, my book collaborator Howard Pearlmutter, after having tried valiantly to inspire me to finish writing based on some visionary notes of his from 2002, said five years was enough time to finish. He was, of course, right, but I had become obsessed with nitpicking every little assumption Howard had made in his notes about how he saw things playing out. Howard's a great guy and I'm afraid in the process of an argument last year with him, I did an injustice to our friendship by becoming too concerned that this world-renowned programmer and architect would send me a bill for his time. I'm so sorry, Howard. I appreciate all you did for me.

But I compounded my error by deciding to soldier on without him. I recast my book with a new title and three themes, instead of one. It would be a history of this decade's computing: Everybody Has a Share: Myth, Madness and Momentum in the Digital Decade. The last third was titled momentum. Here I made my last error. I assumed that by 2010, there was still a chance for the existing PC and client/server-based computing models -- best represented today not by Microsoft Windows but by Apple's OS X -- would emerge bloodied but viable, more open and more secure, as they faced the challenge from software-as-a-service. My rationale was that the service levels and quality we expect from PCs and servers would never be exceeded by cloud computing -- or at least the differences wouldn't change for a very long time to come -- and that privacy concerns would slow down wholesale migration to keeping all our data and applications on someone else's computers -- some small group of service providers each with massive facilities.

This week I realized I was wrong on all three counts that would make my book an essential read. The current computing model is giving way, slowly but inevitably, to a cloud computing model. Does it exceed the service levels of running apps on your PC? No, not yet, but work is now well underway to fix that. Is your privacy assured? No, but that hasn't stopped millions from adopting Google's applications or a host of other Web 2.0 offerings, and now you can move entire databases and applications to services such as Amazon's EC2 and soon similar services from Microsoft, and ultimately Google. Moreover, with the planet's global warming crisis, it makes ever-more sense to millions to save the planet by shutting down their own energy-wasteful PC servers and data centers and hand over everything to service providers who can squeeze every watt out of computing.

While open source remains a triumph of developer will over commercial greed, it is now being ingested by every aspect of what remains of the software industry. My eternal admiration goes to Richard Stallman, who I thought was kind of kooky when I first heard about GNU at the Hacker's Conference in 1984. If not for him, politics and the law today wouldn't recognize the ability of creators to disseminate their works in the new ways that make Linux and Firefox and so much more possible. Kudos also to all the open source committers and supporters whose actions spoke louder than words and finally created an effective counter to Microsoft's world-dominating ambitions.

Open standards are another thing. They too are being ingested, but in the process, all the old tricks and schemes of software capitalism are being brought to bear. Is CalDAV an open standard? No, because it's being tweaked behind closed doors. But I celebrate the efforts of CalDAV's working group because they will have beneficial interoperability when all is said and done. Most customers still only care about interoperability, not standards, and software companies have been playing the interoperability game as long as there's been software, giving information only when forced to by customers or governments or rivals. And by the time the most important standards reveal themselves to developers, that computing model has morphed enough that it just doesn't matter as much anymore.

Yes, there's a book worth of standards madness to be written, but it's not an essential book looking forward (someone please prove me wrong!) and anyway, the problem of how open proprietary data is isn't going away any time in the indefinite future. Don't believe me? Try to move your Facebook contact list over to some other social network, spam your list to get them to join the new network, and release their personal details to a bunch of strangers on the new social network. Sure, standards make it possible, but they don't make it right.

As for the momentum, that book is now out. It's Nicholas Carr's The Big Switch, and he was willing to say what I wasn't, that cloud computing is all that matters moving forward. Yes, he's a bit early on the adoption, but the trend is inevitable. The most telling quote was this one I read in Information Week:

Q: If companies are starting to use the Internet for data processing, is security a huge problem?

Carr: I don't think it's a huge problem. The onus is on the suppliers to prove their reliability and security and earn the trust of the buyers, but my own feeling is that ultimately the utility model will offer greater security than we have today, because today our IT system is incredibly fragmented. Some companies and some individuals are very attuned to security and are very good at it, and others aren't.

Carr is absolutely right. I wasn't a big fan of Carr's first book, because I do think strategic thinking about IT does matter, and makes a huge difference to success or failure of business today. But this security thing will sink any computing trend that smashes into it. That includes email, which all these years later still hasn't solved the spam problem.

None of this means that I think that classic personal computing or client/server computing will just disappear next week. But the writing is on the wall and the trend is accelerating. I personally will continue to guard my own information's privacy more than the average person, but that's me and my generation. The next generation is still concerned about privacy, but is more willing to use cloud computing services despite their concerns. And as far as reliability, or availability, who's lost more data in the past twenty five years -- service providers, or personal computer owners whose flaky hardware and bad backup habits conspire every day to erase millions of files by accident?

To all those who have waited patiently for this book, and cheered me on, my heartfelt thanks. To my family, my message is, tomorrow morning Scott's going to focus on using his abilities to tell stories for money again. Or whatever I end up doing, since that whole landscape looks nothing like it did six years ago. Calendar Swamp will continue on a limited basis, because it represents the community of calendar-sharers like nothing else on the planet. The focus is squarely on interoperability, not open source or open standards, and that's where it will remain.

And to you reading this, keep up the good fight for open, secure and private computing, but remember the words of George Eliot, which still adorn my old domain's home page:

Blessed is the man, who having nothing to say, abstains from giving wordy evidence of the fact.

When I do have something to say that relates to this blog's past themes, I will say it here, at least for now. I'm definitely a wiser man for all I've tried to achieve, but now I need to get back to work.

Array of Windows PCs, and standards, to the rescue

I've maintained a small array of Windows XP PCs for several years. I take a lot of ribbing for not dumping them all and replacing them with a single Mac. But I get more done between the four of them, and occasionally, it comes in handy. This morning, I was trying to re-rip a Red Sovine CD whose final track had some skips when iTunes played it. I had to try three of the four PCs before I could find a CD drive that was able to rip the track with no errors. The PC to which that drive belonged didn't have iTunes installed, so with some trepidation, I fired up RealPlayer, a program I still generally avoid, to rip the track. It did so beautifully, and I was able to drag the track file across my LAN back into the proper iTunes folder. Kudos to RealPlayer for supporting the AAC audio standard. (Yes, I know I should use MP3 format instead, but AAC sounds better.)

Secrecy and open standards: do they mix?

This quote from IBM's Bob Sutor gives me the perfect opportunity to add to this longstanding debate:

"Whatever the outcome of the vote, that secrecy is one of the things that should change in the way IT standards are developed, Sutor said."

There is another recent example of real progress being made on a standard that probably isn't open if secrecy is a no-no. That's CalDAV, which I've written about recently over at Calendar Swamp.

CalConnect meetings, where the main interoperability demos occur, are invitation-only affairs (all members of CalConnect are automatically invited, and membership in CalConnect is open to anyone who can pay the required membership dues). CalConnect does release quite a bit of information about CalDAV and its progress at its Web site, but throughout the process, detailed info about the demos in each meeting are under nondisclosure, the CalDAV spec and related specs evolve while implementations and interoperability testing proceeds in parallel, and standards bodies such as the IETF consider the fruits of CalConnect's labors.

Is the proof in the process, or in the pudding? CalDAV isn't fully rolled out yet, and we can't assess its interoperability across an industry at this time, so I can't answer that question yet. And connecting calendars is a smaller problem than making all XML-based documents portable across all products and services that read and write such documents. But CalConnect seems to put the lie to Sutor's assertion that IT standards must be developed in the open. There's a lot of room for embarrassment of companies whose interim implementations of a standard happen not to work with other implementations. Open that up to wide scrutiny and the companies may just avoid working on the standard, slowing down progress substantially.

I don't have an answer, but blanket statements such as Sutor's strike me as noble aspirations, not ironclad requirements, for open standards to work.

Mozilla's Asa Dotzler takes aim at Apple

Former Wired Magazine cover-guy Asa Dotzler accuses Apple of not fully open-sourcing its WebKit Web browser and gets piled on in his blog's comments, including this comment by Mozilla's own Brendan Eich:

"Mozilla advocates risk looking holier than thou, or just whiny, when there's real work to do."

At some point, with all the money sloshing around, do leaders of Mozilla-type organizations start exhibiting more of the behavior or deportment of commercial software companies? MySQL may not be the only "community" drifting in that direction.

(None of which means Apple shouldn't be as open as possible.)

There's logging out, and then there's logging OUT

Here's an interesting idea...two-tier logouts:

I found this example at Linkedin.com. I wonder if this will become a widespread notion. In these days, could it hurt?

Zoho Writer: Rosetta Stone between ODF and OOXML?

Could a Web 2.0 service become a Rosetta Stone-like gateway between the two incompatible file formats of the next decade--ODF and OOXML? Maybe so. Zoho Writer just announced the ability to export documents to OOXML, and in response to my comment on TechCrunch, Sridhar Vembu of Zoho says it will also be able to import documents from OOXML soon: "Give us a few weeks!" Since the Zoho Writer Web site says it already reads and writes OpenOffice documents, this could be big news, provided it works well. Fortunately, I've got a good test document to try out.