This article is in response to one posted by an online acquaintance, found here:

In which Mr. Tim begins,

“This musing comes as a result of a topic brought up over on Usenet.

The crux of this article is around a fictitious headline of:


Take a few seconds to consider how you would feel, then maybe be kind enough to hear my view.”

His conclusion:  “Bad idea? Yes completely, here’s why.”  And he makes some good points.  Unfortunately, some of the premises of his argument would not seem to follow.

And looking at everything that follows, I suspect I may have been too wordy with this post.  But maybe it had to be said.

So, let me lay out my view:  I think Office on Linux would make Linux more viable for enterprise desktops, which is an area where it has the potential to shine.

In fact, I think most enterprise desktops would do fine with the LibreOffice suite.  However, as a practical matter, I suspect many businesses have processes dependent on Office, such as complicated Word documents, or complicated Excel spreadsheets.  Loading these files into LibreOffice, and then cleaning them up, might be prohibitively expensive — especially when were talking about saving OS licensing costs.  But for some office managers, this isn’t an issue, they just “want Office”.  Okay, give it to them on Linux.  Why not?

It should be mentioned regarding maintenance costs:  in the modern world, more and more Windows system administrators are gaining Linux experience, and have probably booted one or two desktop distros themselves, just to see what all the shouting is about.  Besides, their web site is probably running Linux anyway.  In short: Linux is not the black-box mystery to Windows administrators that it was (say) five years ago.

So there are the arguments made by Mr. Tim regarding how Microsoft’s software in a Linux ecosystem would cause it to suffer.  One premise he makes is that the software would exist as part of a distro, and that there would be some kind of marketing (read: monetary) encouragement for a distro to “push” MSFT Office over other office suites in its default software manager.  But this is a bit of a slippery-slope argument — because why would MSFT want to entrust the distribution of its software to the default software manager?  I believe it would be more likely to be an online purchase, downloaded and installed by the user.  Other commercial packages for Linux are the same way — why would it be different for Office?

(Note that this is different from the idea that I originally proposed — that Microsoft begin selling, and supporting, its own Linux distribution with the capability to run Windows software, which would include Office.  Crazy?  See below.)

Another factor — and certainly one to make one cautious — is that Microsoft has been a bad actor in the past, and not just with regard to Linux.  Yet many folks may not realize that Microsoft is making contributions to the Linux kernel.  Though those contributions (if I am not mistaken) are to help Linux run better under HyperV, they are still contributions to the Linux community that allow more people to run Linux (even if that is in a scary HyperV environment).  Yes, Microsoft is guilty of a great many sins — but at some point, there needs to be recognition that they may be trying to pivot to being good actors in the global software ecosystem — though, certainly, there is reason to regard such activities with skepticism and suspicion.

So then we arrive at the crux of the matter:  at what point do Microsoft’s contributions to free software show that they are acting in good faith, and have atoned for the sins of their past?  In other, more “religious” terms:  What is the “penance” Microsoft must undergo to be absolved from their sins?

I don’t think I am building castles in the sky with this question, and here’s why:  Mr. Tim makes mention of the Usenet.  Once upon a time, servers that operated to the Usenet’s detriment could be “punished” with something called the “Internet Death Penalty” — a harsh term, but adequately describes when the Usenet Cabal (tinc) would collectively stop listening to a particular server.

So consider, then, as similar idea:  the “Market Death Penalty”.  For some Linux and FOSS enthusiasts, there is no “penance” that Microsoft can undergo that will absolve them of their sins — to them, Microsoft deserves a speedy corporate death, no matter what they try to do to atone for their past behavior.  This is the camp I’ve sat in for over a decade, and I’m still not sure that it isn’t a good idea.

In the field of ethics, this isn’t an idea without precedent:  we know from game theory that the strategy for maximum benefit in iterated Prisoner’s Dilemmas is cooperation, unless the other player is determined to be a “scorpion” (one who never cooperates).  (It’s a counterintuitive strategy, one that many can’t accept, preferring the “tit-for-tat” strategy.)  Folks much smarter than I am have developed ethics based on such mathematical certainties.  And even though they may not use these terms, those that advocate a “Market Death Penalty” for Microsoft have identified them to be a scorpion.

Of course, the world is a much more complicated place than “iterated Prisoner’s Dilemmas”, and we would be guilty of the fallacy of “sweeping generalization” to regard it as such.  As one noted online tome[1] about logic points out, “This fallacy is often committed by people who try to decide moral and legal questions by mechanically applying general rules.”  But, the concept of “scorpion” is useful, so let’s not through that away yet.  Which brings me to my question:

Is Microsoft a “scorpion”?

I think the discussion all boils down to that.  And with MSFT’s admission that they now only have 14% of the total markets that they operate in, I think we have them right where we want them.

Don’t believe me?  Have a gander at this:

(Indeed, this was the link I originally posted to the Usenet, with the Subject heading, “What Nadella’s Manifesto means for Linux?”)

And what is the crux of that manifesto?  Here it is:

“More recently, we have described ourselves as a “devices and services” company. While the devices and services description was helpful in starting our transformation, we now need to hone in on our unique strategy.

“At our core, Microsoft is the productivity and platform company for the mobile-first and cloud-first world. We will reinvent productivity to empower every person and every organization on the planet to do more and achieve more.”

As Ars Technica rightly pointed out, these conflicting reports are very conflicting.  After all:  aren’t “devices and services” another name for “mobile and cloud” — in other words, mobile devices and cloud services?

I’ll skip the issue with mobile devices and talk about the Microsoft “cloud services”, which is essentially Azure.  Simply put, they can’t hold a candle to what one can find with Linux.  Linux is king of the “cloud world”, and a lot of that has to do with its licencing.  AND, if you want to build your own Linux-based private “cloud”, all the software to do so is available for free, and the hardware is commodity hardware.

But, whither Azure?  You can’t build your own private Azure “cloud” (cluster).  (You used to be able to buy one from Microsoft, but I’m not sure that’s even available anymore.)  So it’s a very limited cloud universe with Azure, one that is actually a bit orthogonal to the “real” world of cloud computing.

So that’s what I mean when I say “we have them where we want them.”  I suspect — with some skepticism — that they will adapt to the realities of today’s IT world, and try to become a “good actor”.  That adaption means working within the current software ecosystem, and atoning for a lot of sins of their past.  If they ignore those realities, and continue to act in bad faith, we will start to see signs of that “Market Death Penalty”.  If they act like a scorpion, they will eventually die in the market.

It’s almost like the days of yore with IBM — you would never get fired for recommending IBM, until their day was done.

Also — and I know I’ve been long-winded about this — there’s one other matter that needs to be considered.  I don’t know if there’s any “law” already named for this, so I’ll just call it “Doty’s First Law of Organizations” until corrected:

“Every organization can be broken down into competing factions, all the way down to individual persons.”

This isn’t useful for mitigating Microsoft’s sins — because when the company does something, it’s the whole company that is responsible.  However, when trying to predict future behavior, we often pigeonhole the entire organization as having one mindset.  It’s easier to think about organizations in this way, rather than what they actually are:  a collection of factions, all tugging in directions that may or may not match.

And within Microsoft, there is a free software faction.  MSFT would do well to give great weight to anything they have to say.  Because without their input, MSFT may continue to consider their software as part of an “economy”, rather than an “ecology” — the former being money-oriented, the latter being organic.  While the former isn’t necessarily “ruthless” and “being a scorpion” in the business world, it does seem the larger a corporation is, the less conscience it develops.  In today’s world, rare is the megacorp that isn’t widely hated — and Microsoft is certainly no exception to this rule, by far.

Finally, Mr. Tim did make an important point which I do partially agree with.  The issue of host Microsoft Office software may become moot, if they transition to Office 365 overall.  However, I don’t think he appreciates the virtual panopticon that comprises legacy media, nor that people’s reaction to privacy issues around “the cloud” have actually been quite outspoken (even if reporting on such outrage is sparse).  So I daresay there is a chance that these huge cloud systems that we currently see on the landscape may be replaced, in time, with smaller, more private, clouds.  Will there be a place for a hosted MSFT Office in such a world?  I suppose time will tell.

It would also be interesting to see more cloud providers getting better marks on list such as this: