With The Register reporting in its usual scurrilous style Microsoft’s ongoing difficulties in killing Windows XP, Bill Gates stepping down as the head of Microsoft and Taleb making an instructive blunder on the Mac-versus-Windows religious wars I thought I would indulge myself in a rare techie post.
It is also part of the review Nassim Nicholas Taleb’s Black Swan.
One of the many interesting (and instructive) observations that Taleb made in The Black Swans was the following.
A person can get slightly ahead for entirely random reasons; because we like to imitate one another, we will flock to him. The world of contagion is so underestimated.
As I am writing these lines I am using a Macintosh, by Apple, after years of using Microsoft-based products. The Apple technology is vastly better, yet the inferior software won the day. Why? Luck.
This I think is emblematic of the modern diseased, totalitarian mindset, and a central idea in Taleb’s writing. It is no more reasonable to dogmatically assert a lack of narrative than to dogmatically insist on a narrative. Indeed it is more unreasonable to do so yet it is a central feature of enlightenment thinking. The amusing thing about this example is that Steve Jobs learnt his lessons from his brilliant mentor John ‘Pepsi’ Scully and has surely surpassed the master. Jobs on his return to Apple in 1997 realised that Apple were in the furniture and accessories business (h/t: John Rayfield), and Apple products are packaged brilliantly, but to see Taleb herding into Jobs’s honey trap under the illusion that it demonstrates his independence is hilarious.
Let me explain the narrative that Taleb is missing, why Microsoft’s (and Apple’s) success has nothing to do with luck. I am a computer scientist by training and I have been programming and using computers since the early 1981, starting with the Apple II. I spent most of my computer science career cursing and sneering at the ‘Evil Empire’ and up until 2004 resisted its malign influence by stubbornly insisted on using Linux on my desktop for engineering work.
And I would still much prefer to use a Unix-like environment for all programming and many tasks. Indeed as much as XP has dominated the desktop Unix (now mostly Linux) has dominated the server infrastructure. While there are some tools that can make life more bearable for these tasks on XP (Cygwin) it is still utterly inferior in this department. Apple OS X is based on Unix (actually the Mach kernel, which I worked on in the early 1990s) so as well as having a sexy user interface that non-technical users can appreciate it has a beautiful kernel under the hood that technical people will appreciate—the design goes all the way down bringing the best of the server world and the desktop into the one system. One of my good friends also worked on Mac OS X in the late 1990s.
So what gives? The first point is whether one prefers Mac OS X or Windows XP is quite subjective. From the first moment I started to use XP I found it entirely straightforward to use—everything was where I expected it to be—and ever since I have found that it does exactly what I need it to do. On the other hand I find the OS X interface horrible to use, stumbling over some of the most simple tasks (giving up on one occasion trying to pursuade Safari to download an MP3). Apple’s stubborn insistence on using a one-button mouse—always a stupid idea—but looking more and more stupid with the omnipresence of XP. The appeal of Macs is their difference, like the HP RPN calculators, latterly a badge like designer clothes for people to identify themselves, cluster around and mark themselves out from the crowd. I don’t buy the idea that that the Mac user interface is objectively better than XP—for me it is insufferable. Part of that insufferability is derived from some quixotic design decisions on some basic points that having no rationale other than NIH and being fashionably different—not helping the user.
The second point is that the success of Microsoft has been no accident, and only someone with a pretty stunning ignorance of the computer industry could assert otherwise without a good argument. And the reasons do not lie in savvy business practices and a stupid public as some would have you believe. Microsoft have been careful to provide an easy upgrade path for their users to follow, initially erecting the dreadful Windows 1/2/3.x on MSDross—and they were technically inferior to the alternatives at the time—but they were good enough and met people’s needs. There was no doubt an element of luck to Microsoft early success (and luck in not encountering any subsequent unforeseen events that stopped them in their tracks, i.e., black swans), but to set all of their success down to chance and luck is ridiculous.
The point is that from their initial success Microsoft hardly put a foot wrong up through the development of XP. And this was no accident, suggested by this internal email from Bill Gates email (h/t Sullivan):
So after more than an hour of craziness and making my programs list garbage and being scared and seeing that Microsoft.com is a terrible website I haven’t run Moviemaker and I haven’t got the plus package. The lack of attention to usability represented by these experiences blows my mind.
I thought we had reached a low with Windows Network places or the messages I get when I try to use 802.11. (don’t you just love that root certificate message?)
This is Gates in 2003, not 1980, getting the software out of the box and installing it on a PC, illustrating a top-to-bottom preoccupation of what people really, really care about: not having to watch their precious life disappearing down the drain fighting to get software installed and working. Gates has at every point understood what his customers care about (as well as Jobs understands his own customers’ priorities).
The point about making Windows the success that is decades of development (and I think you can see this in almost all widely-used general-purpose operating systems). Microsoft made their user interface less laughable and almost respectable in Windows ’95 and ’98, but prior to that, in 1992, they commissioned the development of a decent and solid O/S kernel in Windows NT, merging these developments to produce Windows 2000, migrating in the process all their users onto the properly engineered Windows NT kernel (pretty smoothly, somewhat more so than the parallel migration to Mac OS X by Apple). Windows 2000 was refined in late 2001 to give us the smart and stable workhorse of Windows XP. (Over 20 years Microsoft provided a careful upgrade path to get their users from a 1981 home-computer hack to one of the most sophisticated engineering artefacts (ever) that is powering the planet’s desktop computers today. That is an astounding technical achievement which I can’t even begin to do justice to here. That Microsoft have in parallel got everyone to use Microsoft Office is further evidence of their competence, and as we can see from the Web browser wars this need not have been so.
The Netscrape web browser got off to a good start in 1994 but it was quite appallingly engineered, really deserving to lose out to Microsoft’s Internet Explorer, only for Microsoft to became complacent about Explorer and giving FireFox the opportunity rise out of the Netscape ashes and come roaring back to re-establish itself as the browser of choice (considering Microsoft’s natural advantage, an inertia that sees 4 users of IE for every user of FF [see IE and FF] and will protect IE until it gets turned around or stabilised).
The truth of the matter is that in Mac OS X Apple provided too little too late, their own earlier operating system offerings being pretty uneven. In 1984 the Mac interface was ground breaking but it lost its edge and never regained it until OS X. Mac software and the OS kernels alike were generally flaky and not really much better than MSDross. By the time they got their act together in 2001 Microsoft had migrated the planet fairly seamlessly onto a very decent 32-bit multi-user operating system. And Microsoft systems have always been much cheaper than Apple and run tons more software.
The real point about Windows is that it will run on anything while Apple provides an operating system that will only run only on Apple-supplied hardware. (I know it can be hacked to work on anything; that is not the point; Apple don’t have to make it work on everything.) That Windows (from the user’s point of view) just works on everything is a tremendous achievement, and very valuable. That everyone uses the same suite of application programs (which is itself a mighty feat of engineering) is tremendously valuable as it means us losers can exchange documents with anyone (even Mac losers) without having to worry about all the horrors of changing formats. If Microsoft had not ported Office to the Mac then Apple would almost certainly not have survived as a desktop computer vendor.
None of this happened by luck. Not at all. That Microsoft find themselves captive of their brilliant monster that keeps their users happy all these years says it all. And I am really pissed at Microsoft for killing XP—and yes, the competition authorities should look into this.
[By the way, having worked in a corner of the industry watching Microsoft and others pouring vast quantities of money into breaking into the cell-phone market, especially the holy grail of data-stream oriented cell-phones, it is amusing to see Jobs succeed with his iPhone accessory where so many others have failed. This again is no accident. Marketing and narratives are important. They make up our reality. Apple is worthy of respect just as Microsoft; both originating in the microcomputer revolution their genius has always been complementary.]