11 June 2007

Book Critique: In the Beginning was the Command Line.

BCIS301-07A Book Critique

In the Beginning was the Command Line
Neal Stephenson (1999)

Critique by Mick Gregg
For Trevor Nesbit, Christchurch Polytechnic Institute of Technology

24 May 2007

Introduction
The story goes that Neal Stephenson originally started to write In the Beginning was the Command Line as an article (1) on slashdot (http://slashdot.org), but that it bloomed into a full manifesto of how he saw the then state of computing for the masses. 'Then' was 1999 when the desktop world was divided between Windows9x/NT and the pre-UNIX Mac OS, BeOS was alive and the age of Linux was still in single digits. What Stephenson wanted to tell the world was that since the days when he learned programming on a teletype across a modem to a mainframe, the power and workings of computers have been abstracted away from users for the sake of expediency and commercialisation, in the form of the Graphical User Interface (GUI). We are told that this disconnection allowed the businesses involved in operating systems and computer hardware - namely Microsoft and Apple - to dictate the terms of our relationship with technology, and for a few experts in the field to become dangerously powerful in a world where users are educated to expect only the most simple of choices.

While In the Beginning focusses on the importance of computing, Stephenson is happy to cite examples of the ills he perceives from the world outside as well, hinting that the Land of Operating Systems is really a microcosm of the Developed World. Stephenson sees a world pacified by "moral relativism" and a "global monoculture" that has "done away with the ability to make judgements". He concludes that, "It simply is the case that we are way too busy, nowadays, to comprehend everything in detail. And it's better to comprehend it dimly, through an interface, than not at all." For the computer specialist and the casual user alike, this may be an epiphany revealing why philosophies within the IT world, not just trends and prices, need to be firmly in the minds of decision makers.

Summary
The relatively recent history of electronic computing means that many formative protagonists are alive and well, many commentators have been involved since what they perceive to be day zero, and opinions haven't yet been muted and dulled by the parity-of-blame technique that makes academic historical study appear to be unbiased. Stephenson is one of these commentators and In the Beginning was the Command Line is all about having an opinion and taking a side, opening with a brief history of computing from his driver's seat. He begins as a schoolboy on a teletype and explains how he was using an interface that had existed for a century before to send commands to a distant mainframe, the move from paper output to a glass monitor, and the shift from batch processing to a real-time command line. For him, the GUI came about in 1984 when he began what he calls his "personal love affair with the Macintosh", and he acknowledges that providing a simplified interface to the computer through the GUI opened a newly created market to a new kind of computer user. He defines the end of the GUI debate as the point when Microsoft eventually adopted one as well, albeit as a layer on top of MS-DOS, and goes on to explain the "interface culture" that requires so many things to have a simplified GUI, abstracting technicalities below, so that users never have to properly learn and instead leave the hard stuff to an adept minority.

Stephenson draws parallels from the world outside computing to demonstrate that this hands-off culture is endemic in America, with the abandonment of command line interaction presented as both a symptom and cause of an apathetic and disengaged psyche of life through controlled voyeurism. More than extrapolating the fraudulence of GUI-based interaction to prove the cynicism of the industrial elite, Stephenson sees the GUI as an example of the consumer's general desire to ignore anything not filtered to him, through a medium designed and controlled by an entity that he may have erroneously placed his trust in, to avoid the need to personally judge and decide.

By now Stephenson has already described the OS choices of 1999 with car metaphors, so we know that Macs are stylish sedans with sealed bonnets, Windows is an over-priced, but hugely popular family station wagon, BeOS a Batmobile and Linux a free tank. It's no surprise then that he introduces the latter two with enthusiasm as he documents his move back to the command line after losing faith in his GUI-only Mac. He wanted to pop the bonnet again, and revels in the open source culture of hands-on computing, open bug-tracking and technical freedom. This may well be the point at which Stephenson suddenly looked back and saw the years of encroaching abstraction and consequently declining user abilities and waning understanding.

That the least technically sound OS options have become the most expensive and most popular is explained as a result of the historical business decisions of Apple and Microsoft to become hardware and software manufacturers respectively, and their need to swaddle their products, obfuscating their workings to create something salable. However, Stephenson surmises that Apple, as a hardware company, will lose out to the open architecture hardware that MS Windows runs on, and that Microsoft will lose out to Linux because operating systems are naturally required to be free (as in freedom). With optimism that a new age is beginning, where users demand openness and embrace the available power of Batmobiles and free tanks, Stephenson closes with a succinct summary of his position, that "if you don't like having choices made for you, you should start making your own."

Critique
It needs to be understood from the outset that Neal Stephenson isn't presenting In the Beginning as a history of modern computing, a technical guide to Linux or even an expostulation for the use of open source software. Instead it's a very personal account of his own experience of computing and how he sees this discipline going wrong. As such he makes, and is required to make, no effort to prove his case by citation and instead relies on the reader's trust in his authority and understanding of his logic to be persuaded. For example, he makes no mention of the invention of the GUI paradigm at Xerox PARC. This doesn't alter his argument, and he doesn't explicitly and incorrectly credit Apple with the first GUI, he simply mentions the Macintosh as his first encounter with it.

His anecdotal sampling of readily identifiable examples from the non-computing world illustrates his belief in the systematic decline of modern consumers' ability to think for themselves. The popularity of fake Disney-style worlds in preference to adventure travel, and factory souvenirs over genuine African art is used as an argument to prove a craving for "mediated experiences", which Stephenson likens to the desire to be coddled by a GUI. He also introduces HG Wells' Morlocks and Eloi to explain the relationship between technical workhorses and aloof users, highlighting the perversity that the latter are in the majority in contrast to the original Wellsian world. Then there's the Hole Hawg, a monster power drill that makes all others pale by comparison, used to represent UNIX. Like a populist politician, Stephenson isn't trying to complicate his arguments with exhaustive research or endless historical examples, he simply uses an easily imagined occurrence to illustrate what he's sure is a general point.

The Interface Culture chapter is particularly beautiful; a poignantly personal article in its own right during which Stephenson reveals the nucleus of his distaste for the effete and facile society he sees evolving. He gives himself room to explain the failings as he sees them, before logically tying the development of "feckless human being[s]" to the GUI-driven world where Morlocks present Eloi with only the most simple of choices. It's in this chapter that Stephenson's need for command line interaction is indelibly linked to the world at large, but he lightens what could degenerate into a diatribe by clearly conceding the misery associated with the Platonic rule of intellectuals.

Stephenson also spends a full, and relatively long, chapter looking into the contrasting ways in which the commercial marketing machines and open communities deal with bugs. This should be essential reading for every Microsoft software user, not because great hordes of Eloi will want to read technical documentation about the shortcomings of their tools, or to shock them with the abundant faults in the kit they've so expensively licensed. This chapter should be read to force users to accept that a company that wants to sell them something is going to lie to them about its product's quality, while a community that wants help to build something needs an honest exchange of findings. Again, Stephenson doesn't see a need to cite published evidence of his position, he gives us some examples from his personal experience and asserts that, "Joint stock corporations ... are good at many things. Admitting failure is not one of them."

It's probably fair to say that Stephenson has no great love of MS Windows and wouldn't encourage its use by Eloi or Morlock, but he is no stereotypical Microsoft-basher either. He seems to like the behemoth's application software and predicts, or at least hopes for, the day when this becomes its focus, leaving operating systems to those who make them truly open. As with his care not to deride corporations in general, this might be an attempt to appear balanced, but is more likely a piece of genuine praise, since Stephenson doesn't seem generally concerned to court favour from anyone, the reader included. His prediction that Apple's attempt to create a hardware monopoly and Microsoft's reliance on selling operating systems as its core business would eventually be their undoing has logic, and some of us live in hope, but to date these companies are still succeeding without changing these strategies.

In the Beginning is great reading for the anti-capitalist, Linux-using Morlock, and is probably as good for those who score two out of those three, but Stephenson isn't just trying to preach to the converted. Constantly stepping outside the computing world and directly using metaphors like the already mentioned OS-as-car, combined with Stephenson's personal narrative, makes the philosophy that he expounds not only understandable to non-technicals but also applicable to the modern consumer world in general. He acknowledges that he could be seen as a "snotty intellectual throwing a tantrum", but at the same time explains the need, as he sees it, for his personal world-view to be understood before the cultural aspect of Microsoft and Apple domination can be made sense of. He's not a political revolutionary and isn't calling for the overthrow of the world as we know it, but attempts in this book to open the eyes of the docile majority to the need he sees for hands on interaction in place of passive compliance.

In keeping with the informality and evangelism of the essay, the complete text is offered online to be freely downloaded (2). It comes in this form as an ASCII text file, archived in both Windows and Mac OS formats, leaving no excuse for the online community to ignore it.

Conclusion
In the Beginning was the Command Line is a manifesto. It's a line in the sand that Neal Stephenson drew in 1999, asking his readers to join him in as a Morlock or condemn themselves to sheepish Eloism. His position is simply that as consumers, modern westerners have become accustomed to, and even demand, that their interactions with the world be filtered through some kind of interface that reduces the need for them to think or judge or experience life for themselves, and that their decisions are increasingly made for them in advance by detached technocrats. Stephenson sees an escape from this by a move back to the command line; literally in the OS world and as a metaphor for direct interaction.

Linux and BeOS are discussed as the panacea to the commercially controlled and cynically protected operating system industry of 1999. The user's return to the command line will begin his journey to real involvement and understanding, and exemplify the failure of Microsoft and Apple to cage what should naturally be free and simplify what is intrinsically complex.

The circumstances have only slightly changed since then, with Apple embracing UNIX and once more endearing itself to Stephenson (3), as well as reviving its fortunes with the iPod portable music player. Just as Apple repackaged the Xerox GUI and sold it as its own, it now has a whole new market of consumers that thinks it invented the Walkman. The hardware company hasn't yet met its demise, neither has MS Windows yet lost its mass appeal, and BeOS is dead - though its fans may enjoy Haiku (http://haiku-os.org). The age of open source operating systems naturally replacing the incumbent closed ones hasn't dawned, but Linux has steadily increased in popularity, with even Michael Dell in its user base (4) and Dell itself about to launch a range of PCs with Ubuntu Linux pre-installed (5). Apple came to the command line party late and shunned Linux for a UNIX variant that it could still hide the code for, but it exploits plenty of open source projects - including Apache, OpenLDAP, Samba and Postfix - that allow users to dirty their hands beneath the GUI.

One day readers might regard In the Beginning as an historic document from a time before users wised-up to the need to educate themselves about the technologies that were running their lives, when point-and-click technicians called themselves Systems Engineers, and when the need for a company in Seattle to generate some cash for shareholders pushed worthless face-lifts onto any business that needed to buy a new PC, but the only alternative they could think of was an expensive, pretty toy for 'creatives'. That day hasn't come, but it still might and this gives continuing value to Neal Stephenson's 1999 commentary. Government policy makers and university strategists (assuming that these groups are independent of commerce) should read it so that business leaders, who don't care anyway, might one day at least be influenced by it by proxy. Stephenson's Eloi should read it and take time to assess who is directing their lives, and his Morlocks and would-be Morlocks should read it to grasp the importance of philosophy and policy in technology. What Stephenson reminds us is that as a society swamped in a culture of marketing-driven pseudo expertise and half-understood buzzwords, where brand familiarity replaces true comprehension and homogeneous thought is the profiteer's nirvana, it's up to those who call themselves experts and professionals to truly be such, and to provide real leadership in their domains. Exclusively learning the methodology and product range of one entity, no matter its market dominance, is, far from education, nothing more than a branch of consumerism.

Neal Stephenson paints a sobering picture of a user booting a machine to see a single button labelled 'LIVE' as the ultimate expression of a culture of deference. We're nearly there, but there is still time to pull back from the brink.


References
1. I haven't found the alleged original slashdot article. Though it is often referenced in forums about this book, it may be a piece of 'Net mythology.

2. Neal Stephenson (1999). In the beginning was the command line. Retrieved on 27 March 2007 from http://www.cryptonomicon.com/beginning.html

3. Neal Stephenson responds with wit and humour (20 October 2004). Retrieved on 27 March 2007 from http://interviews.slashdot.org/article.pl?sid=04/10/20/1518217

4. Michael's computers (2007). Retrieved on 17 May 2007 from http://www1.ap.dell.com/content/topics/global.aspx/corp/biographies/en/msd_computers

5. Dell to offer Ubuntu 7.04 (1 May 2007). Retrieved on 17 May 2007 from http://direct2dell.com/one2one/archive/2007/05/01/13147.aspx