Microsoft and Attachmate were not Novell's destiny

Big bets, brass balls, and other lost assets

Open...and Shut Novell, a collection of mostly legacy software businesses, has announced that it will be devoured by another collection of legacy software businesses, Attachmate.

It's not Novell's finest hour, and is made worse by the sale of 882 patents to Microsoft-friendly CPTN Holdings, even despite Novell retaining its Unix copyrights. Novell, the one-time networking giant, sold for scrap, and possibly selling out the open-source community in the process.

Sadly, it's not clear that this route was preordained for Novell.

In early 2004 I attended a meeting in which Carl Ledbetter, then Novell's senior vice president of engineering, detailed Novell's quandary for a small group within the company. Since mid-2002 I had been working in Novell's Developer Services group, which reported up to Ledbetter. He explained that Novell's NetWare revenues were declining at a steady double-digit percentage each year, and that Novell needed to find new business to replace this dying revenue stream while doing everything possible to slow its decline, so that NetWare could fund this new business.

Enter Open Enterprise Server and other machinations designed to prop up the NetWare dying horse that the market seemed intent on beating - again and again and again. None of them worked. At least, not well enough.

In retrospect, it's fairly clear why: Novell played more defense than offense. It spent more resources tightening the tourniquet on the gaping wounds that were NetWare, GroupWise, ZENworks, and other legacy businesses, rather than taking bold steps into the future. After Ximian and SUSE, Novell largely abandoned any attempt to bet big and take risks.

It didn't help that Novell lost key executives like former vice-chairman Chris Stone (to pursue "other" opportunities) and Ledbetter (joined Pelion Venture Partners), replacing them with professional managers like Ron Hovsepian. I admire Hovsepian for his operational expertise, but he was never the sort of dynamic visionary that Stone was - the sort of leader Novell needed in order for the company to become relevant. I really like Hovsepian as a person and think he's an efficient manager. But he was unlikely to be the one to brainstorm Novell's way out of its slump.

Under Hovsepian's watch, Novell cozied up to Microsoft, gaining short-term revenue but losing long-term leadership points as Novell distanced itself from the open-source community that has supported Red Hat's rise to a $750m Linux and middleware juggernaut. More recently, Hovsepian managed to spend eight months shopping Novell after an Elliott bid, only to eke out six per cent more from Attachmate. Was it worth it?

No, but apparently there simply wasn't much interest in buying Novell, as The 451 Group's Brenon Daly points out:

[B]eyond all of the complications around matchmaking is the fundamental fact that Novell just isn't that attractive, regardless of whatever business we look at inside the company. Each component of its revenue (license, maintenance/subscription, services) has dropped so far this year, which is part of the reason why Novell has come up short of Wall Street expectations every quarter this year. Overall, sales have dropped six per cent in 2010, and current projections call for Novell's revenue to decline next year, too. So as we look at it, the board probably did a fair job to get Novell valued at $1.2bn (net of cash), which works out to basically 1.5 times sales.

Again, it didn't have to be this way. Novell under different leadership might have swung for the fences. It might have still struck out or it might have hit a home run. Maybe it could have gone all-in on middleware, as Red Hat has done. Or maybe it could have led out in virtualization, beating Red Hat to the punch.

It did neither. Instead, it shepherded its dwindling assets and sold them to the highest bidder, which turned out to not be that high at all.

Meanwhile, the big bets that Novell took early on to redeem itself, particularly the SUSE bet, are about to wither on the Attachmate vine. Attachmate has no experience managing an operating system. Much less an open-source one.

Attachmate has told SUSE's development community not to worry, that life will be much the same as before. This isn't very reassuring. Attachmate has zero credibility in the open-source world. Name a single open-source project that Attachmate contributes to in a meaningful way. A search for "Attachmate open source" yields nothing. Well, it does reveal this little piece of anti-open source FUD in Attachmate's FAQ for its FIPS 140-1 product:

FIPS 140 validation provides third-party confirmation that a security software implementation meets the highest possible standards. This validation sets Reflection for Secure IT apart from competitive software, such as open source software, which is not validated.

Welcome home, OpenSUSE developers!

This is the final end of Novell, a company that consistently squandered its opportunities to regain relevance, slouches its way into Attachmate, while potentially poisoning the Linux industry through the sale of its patents to Microsoft/CPTN Holdings. While this IP doesn't include the Unix copyrights, it does include 882 of Novell's roughly 2,000 patents which almost certainly involve IP surrounding WordPerfect and other technologies that give Microsoft more cause for sabre-rattling against open-source projects like OpenOffice.

It may be, as Brian Proffitt persuasively argues, that there's little cause for alarm: that no one has ever shown Linux to infringe Unix intellectual property and, even if it does, Microsoft would have IBM, Oracle, and every other Linux-friendly vendor rising to Linux's defense if Microsoft dares.

Perhaps.

But the fact remains: Novell has once again gone for a short-term remedy and loaded itself and the industry with long-term problems in the aftermath. When I left Novell in 2005 - I'd served as director of Novell's Linux Business Office and Open Source Review Board - I did so because I didn't feel that Novell had the heart to compete, that it was soft. As 2010 closes, it's abundantly clear that I was right.

I wish I weren't. I wish things would have turned out differently for Novell. I still have friends there, friends who believed that the company could turn itself around. I can't see that happening under Attachmate's leadership. Novell might have been a contender. I'm truly sorry that it is not.

To Attachmate: SUSE is still the crown jewel within Novell's remaining product portfolio, along with developers like Miguel de Icaza who create impressive technology like Mono and SUSE Studio. I don't see how you, given your history, can manage these people and products better than Novell has done, but try. You don't want to spoil these assets. They were and are Novell's best chance at becoming a contender again.

Matt Asay is chief operating officer of Ubuntu commercial operation Canonical. With more than a decade spent in open source, Asay served as Alfreso's general manager for the Americas and vice president of business development, and he helped put Novell on its open-source track. Asay is an emeritus board member of the Open Source Initiative (OSI). His column, Open...and Shut, appears every Friday on The Register.