Kelty (2008) Two Bits, Chapter 5
Kelty, C. 2008. Two Bits: The Cultural Significance of Free Software. Durham, N.C.: Duke University Press.
Chapter 5. Conceiving Open Systems
143 ‘open’ most complicated part of Free Software: opposite not ‘closed’ but proprietary software.
143 This chapter: struggle over meaning of ‘open systems’, 1980-1993.
144 Not just technical, also moral components to contest. Moral = ‘imaginations of the proper order of collective political and commercial action’.
144 Open systems have a blind spot: intellectual property. Irresolvable tension between manufacturers promising interoperability and secrecy-ridden intellectual property regime; a tension between ‘incompatible moral-technical orders’.
144 At heart of struggle was standardisation, but never clear how and who would do it
145- Hopelessly Plural
145 Open systems seen as solution to all the legacy of locked-in computer systems. But dream of The Computer never materalised: we have a world of computers: ‘myriad, incompatible, specific machines’.
147 From 1950s to 80s stable marketing strategy: find customers, sell them whole package, charge them a lot. But by 80s computers faster and smaller.
147 Open systems emerged in 80s with promise of interoperability, marketing and PR people spoke of “seamless integration” [this buzzword still around in 2000s, see Harvey, Green et al’s work on ICT projects in Manchester, esp. unattained dream of interoperable museum databases across EU; see also my own research on local e-government in Malaysia, forthcoming].
148 But term open systems “hopelessly plural”, 100+ definitions; means or ends? whose goals? who set them? do state and non-state agents agree on them? [see Roig, this blog, on ‘double openness’ in relation to internet filmmaking].
148 At any rate, openness became a ‘cultural imperative’ [cf. Harvey et al’s ‘imperative to connect’, Manchester ICT projects, this blog]
149 At least one thing clear: opposite of open was ‘proprietary’.
150 Everyone thought open a good idea, but no agreement on *which* open systems.
150 But intellectual property idea of moral order in conflict with that of open systems, yet intellectual property left out of the definitions of open, taken for granted, not challenged.
152 From intellectual property viewpoint, would be mad for company to release source code and let other vendors build on it – what could the company sell then? ‘Open systems did not mean anything like free, open-source, or public-domain computing’.
152 Account of how UNIX entered open systems contest shows tension between open systems and intellectual property
153- Open Systems One: Operating Systems
In 1980 most obvious option for standardised operating system was UNIX, already running on more than one hardware type
153 UNIX wars of mid-1980s: vendors from both sides ganged up to support rival standards.
154 Longhairs vs shorthairs 🙂
155 When UNIX spread and became more fragmented, efforts to standardise it
155 Figuring Out Goes Haywire
… circa 1986-8: four rival standards for open systems
155 UNIX had spread (porting) but also differentiated (forking) – 156 different features added to diff ports
156 Step one: creation of standard that specified minimum set of functions at interface level: ‘interface definition’ – 157 alas two competing definitions emerged
158 ‘Standards’ ended up being products that firms wanted to sell to their industry via a consortium, e.g. interesting Sun ad from 1987 tapping into angst over open systems.
160 Open Source Foundation with some of biggest players: IBM, Digital Equipment, HP…
160 Sun branding itself as open-systems idea originator; promised to free companies from grip of single vendor
162 UNIX wars of late 1980s: all against all, fighting to show customers that they’ve made right choice not of machine or software but of standard.
164 UNIX wars show blind spot of open systems: intellectual property rules at odds with specificities of software; assumption that w/o intellectual property innovation would stop.
164 – Denouement
164-5 Result not a single winner, but ‘reassertion of proprietary computing’. Microsoft one big winner for same reasons open systems failed: ‘intellectual property favored a strong corporate monopoly on a single, branded product over a weak array of “open” and competing components’.
165 Return to IBM-style monopoly but with new monopolist
166 Open Systems Two: Networks
Another crucial part of open-systems story of 80s was efforts to standardise networks, esp. inter-networking protocols.
167 Conflicting social imaginaries of OSI and TCP/IP protocols; across state, uni, industry borders. By 1993 TCP/IP had overtaken OSI cos of (a) availability, (b) modifiability, (c) serendipity, incl “killer app” WWW
168 … familiar C20 battle over gov planning and regulation vs. entrepreneurialism
168- Bootstrapping Networks
From 1970s lots of competing closed networks, with IBM one pioneer
169 Late 70s BBSs appear
169 also in 70s telecomms get in on the action, eg Minitel France experiment
169 in common across experiments: networks piggybacking on existing, state-regulated telecomms; 170 these hampered by antitrust & monopoly laws
170 OSI and TCP/IP epitomised gulf between computing and telecomms worlds; Arpanet was ad hoc and experimental, very different from ISO
171 all agreed standard network protocols were a must; although derided for being slow, bureaucratic etc, the ISO and ITU processes had undoubted legitimacy
172 TCP/IP explicit goal: share computer resources, not necessarily linking two indivs or companies, or to create competitive markets for networks (software)
173 Requests for Comments (RFC) process integral to process of standardising; 174 geeks love its history cos shows clever, ad hoc solutions to coordination problems – muddling through
174- Success as Failure
1985 OSI was an official standard, but there were few implementations; 175 it was in face TCP/IP that turned up in actual systems by late 1980s. Why successful? Again (see above): availability, 176 modifiability, serendipity.
To understand FS practices must understand open systems & openness; 1980s open systems struggles prepared stage for FS, ‘leaving in their wake a partially articulated infrastructure of operating systems, networks, and markets that resulted from figuring out open systems’.
177 TCP/IP protocols success created single standard and entity, the Internet, with own built-in goals mirroring FS’s moral-technical order
178 Constraints to collaborating are in flux, resulting from struggles involving giants like IBM, onetime rookies like Sun, amateurs, academics, geeks, etc.
The creation of a UNIX market failed. The creation of a legitimate international networking standard failed. But they were local failures only. They opened the doors to new forms of commercial practice (exemplified by Netscape and the dotcom boom) and new kinds of politicotechnical fractiousness (ICANN, IPv6, and “net neutrality”). But the blind spot of open systems—intellectual property—at the heart of these failures also provided the impetus for some geeks, entrepreneurs, and lawyers to start figuring out the legal and economic aspects of Free Software, and it initiated a vibrant experimentation with copyright licensing and with forms of innovative coordination and collaboration built on top of the rapidly spreading protocols of the Internet.