No Subject

"Take That Bass Out of Your Voice.  You Talk to Me in Treble"18


For Kittler19, for Heim, and other commentators, the keyboard profoundly
effects the users access to flow.  However, several more years into the
phenomenon, not only has the processuality of flow been broken up into
more and more granules, more features, on the one hand, on the other, the
more long-term acclimatisation to word processing means that this
seemingly magical release of language gained by the removal of the
struggle against materials becomes normalised in comparison to its
reception by these users.  Flow, as in that revealed by an extended
exposure photograph of moving water simply becomes a form, a range of
potential paths already traced out.

What presents itself as a zone of plenitude is in fact an immensely
complicated lock with hundreds of tumblers and latches, variations and
categories.  At the same time, it is too easy to allow the luxury of
assuming determinism.  There is no such intensity, no implicit speed or
brutality to the intersection at which the user is placed by the
feature-vectors of Word. Neither is there the languidity, coolness or a
sense of polysensual opening up of process.  Rather there is a range of
positions of balance, which is not to say of comfort, but of a neutrality
a thermodynamic flatness that always has embedded within it a number of
calls to order

The perfect text for Word is not the cutted, pasted and folded deck of
samples, nor the synthesising torrent of a mind welling up into its
perfect receptacle but something like Pride and Prejudice.  A novel sewn
together by an insect plugged into a perfunctory AI and a relational
database.  It gathers its objects, makes lists, describes them, runs
protocols and observes their correlation with a limited variety of
perceptual screens.  Taking advantage of integration with Excel, the
writing of realist novels could be taken to an even higher level.  Once
the relevant data of income, beauty, ownership of land or title,
intelligence, relationship to trade or domestic skill, familial position
has been entered, the spreadsheet can run endlessly variable simulations.


Just as tools become tasks formulated around structured choices rather
than material potentialities, language becomes information.  What is
excised from language in the mode of information is noise - what Lecercle
calls the 'remainder'20.  Everything that is not only the 'junk' of
language, its gibberish, but also what animates it.

Word combines both the receptive, control automating mode of software
typical of an automated production line, a traffic control centre, a
security system - systems designed to maintain a homeostatic, conditioned
level of specific flows - and the menu based channelling of behaviour
typical of the personal computer.  The discussion of the interface
focussed mainly on the latter.  It is quite how language is articulated as
flow that is of concern here.

There is an argument that as texts in word processing are divisible and
combinable their 'contents' are eroded.  "Texts are provided with a
pointed structure.  The argument is structured in advance and divided into
separate subjects, items and paragraphs.  Items can be added or deleted
later on which may result in some loss of the course of the argument"21
Contents in this sense means their relationship to truth, how closely
their moral infrastructure is matched by the formal hygiene of their
construction.  This is an argument more around 'brainstorming'22 or
idea-processing softwares such as Inspiration, where a missing or out of
sequence cell in a flow chart can have 'significant' effects, rather than
word processing as exemplified in Word but essentially this argument is
one with planned structure in writing (something contradicted by the
monotonous ticking off of 'issues in cyberculture' by the book which makes
it - itself a waste product of a modularised education system).  Where
this effect is produced is at the level of screen size actually precluding
an overview of the document.

	At a 'lower' level than that of content, documents are also
marked.=20 It has become a commonplace that all speech acts are as much
verifiable by their circumstances as by what they actually 'say'.  At the
level of import filters there is a politics of control over standards that
determines whether speech can even be enacted.  Open a PC Word file in a
Mac and find where it sits on the hard disk of the author.  At other
times, texts produced in competing word processors or versions of Word
that have 'surpassed' the one you are stuck with are rendered completely
unreadable.  How Word interacts with and sometimes countermands the local
Operating System can also produce a kind of moir=E9 pattern of competing
systems of control:  use Word on a Mac and find that if you save to the
desktop the program will pedantically rearrange your files in the order of
last use, no matter if you are attempting to spatially order them by a
different means.  The way, like all programs produced by Microsoft, Word
is, to put it politely, optimised to work in amalgamation with their other
products.

In terms of how language is inflected, Word still has some way to go
before it fully automates flow or even composition.  Whilst newer versions
of the program are capable of assessing whether a word needs a space
inserting before or after it when it is being pasted into another sequence
of characters it still does not recognise modes within language on the
basis of semantics.  For instance, although in AutoCorrect the first
letter of a new line can be automatically capitalised, when it is cut and
pasted into another section of text the capitalisation of the letter is
not reversed. All that is seen is the occurrence of an event, a carriage
return, the change in semantic mode is not recognised.  This is of course
what allows for the relative flexibility of use of the software and why it
makes a strict division between itself and the document.  Where it most
visibly does not however is where that perfectly isomorphic machine
Microsoft English is liberated from the clutches of its miserable captors,
the users.


stunt-double for spelling


The dictionary is always a revival of language.  The Oxford English
Dictionary's attempt to surpass this condition by establishing itself
online, by having weather maps of linguistic norms and clusters updated
second by second as they breed, die and spill across the globe. (The OED
will soon no doubt begin to become available as an online facility built
into word processors as all text gradually becomes melded into one
document.  Not only would they be able to provide exactly up to date
corrections of meaning and spelling by this means, but also be able to
scan every document submitted for any savoury traces of neologism, or
unlogged usage).  Even in its networked state, the dictionary can only
break from its passive position of recording and instantly updating
nostalgia to impose a future on language by throwing etymology into
reverse, tearing the roots of words up from the mud of the past and into
the speech of the present.

	Fearing such a hermeneutic catastrophe, the dictionary must always
remain the patient recorder of language.  When it puffs its little
bookshelf chest up to establish a normative function is when it looses and
imposes, not its past upon the present, but its intended, already lost to
language, future of residues.  In Microsoft Word this looser is always
grappling for some purchase on any piece of text that passes it by before
it is eventually swamped and drowned.  This is not of course to say that
as a knowing, personalised zombie it cannot have its uses for minor
services of correction.

	You have to go through several layers of interface to switch off
'Grammar' and 'Spelling'.  Where things get odder and more interesting
though are in the user dictionary and in the setting for the personalised
'AutoCorrect'. User dictionaries are not repressed by the language of the
dictionary in the same way that if you wear a stereo you can spend enough
time in a crowded train, headbanging, farting and beating your fingers on
things or in the street yapping and gesticulating into your hands-free
mobile without the constraint of the audial conditioning of your physical
location.  This is of course the same domain that an enlightened Kant
allows for reason within discourses circumscribed by the securing rule of
the prince rather than that of the headset.  Room to manouevre.  Sort of.

	In order to escape this position, users incorporate the dictionary
and the thesaurus into every grain of their text, running verbs,
prepositions, nouns, the lot, through the mill of red and green lines.

Every possible utterance becomes a combination on look-up table results.  
Global English, fuelled by digitisation and a ready supply of standard
components simply becomes a permutation of all potential combinations.

User dictionaries are full of skirmishes action against 'imperialist'
English, but with only this 'standard' to refer to they cut loose from
conforming to a more 'localised' set of rules (say, English English
English) unless they have already been learned, or unless another device
such as a dictionary becomes a part of the writing sequence.  In this way,
given no direct resource to the production of a local field of variation,
they cut loose even from dialect.  The divergence from the normalised
language of the world is potentially immensified as these pockets of
mistake, homonymy, dyslexia, takes their time to inbreed and thicken their
juices before lurching out into the supposedly clean white substance of
language under telecommutation.=20 One spells with correctness, a variant
of it (derived from some minoritarian, dialectal or other process of
variation) or neglects to spell at all. 


At the same time another private mutation is being hatched. Microsoft's
recent alliance with Real Names23 is an attempt at a rigidification of
language to the point at which it becomes solely data.  What is hoped by
this scheme, (wherein words that are also brand names typed into the
search portal of Microsoft Network or the location window of Internet
Explorer lead directly to the correctly denoted site) is of course that
there is no longer any ground, but only a continuous and spreading act of
freezing.  When the meaning of a string of characters can be bought and
locked into place this is the thermodynamics of language reduced to a
single cryogenic chamber in which the corpses of Ford and Disney are in
omnipotent storage.  For linguists, a dream, a few years on there will be
nothing left to study, nothing except to go round with clipboards ticking
off the packets of data as they pass the test


Other aspects of the impact of the net on Microsoft applications are
readily visible within Word.  The provision for HTML code generation is
clear, but there are also quite substantial beginnings of a change in use
and in the construction of the user.  From the beginning, Word came with a
single license for a single machine.  This has not changed.=20 What has is
the implication of a single author for each document.=20 Whilst it was
assumed from the beginning that several users would be authoring their own
documents using the same copy of the software one after another, the
massive involvement of the nets in the production of texts now produces
side-effects in the construction of Word.

	The economic foundation for this is the hope that by making a
networked work group accept Word as a tool, the people who they in turn
have to work with will also find it necessary in order to access the
document formats or features that have become incorporated into their
procedures - for instance the layers of stages in which versions of a
document can be 'signed off' by managers.  A slightly more straightforward
way of enforcing this 'usefulness' is by configuring the import filters
and the formats in which files can be saved or 'Sent to' in order that
Word effectively works to disable its competition. Reinforcement of use
is thus provided positively by features such as the ability to Track
Changes and negatively by its selective ability to deal with non-native
file formats.

	Word demonstrably succeeds though in those places when it makes
available the qualities of perception of data specific, at least at speed,
only to computers.  Where the production of a text is spread over various
authors, locations or times the ability to deal with various versions, to
Track Changes for instance, allows the computer to begin to take a part in
the processing of text in a manner which maximises its own propensities
without beginning to format the user in a non-interrogable manner.  The
multiple and shared versioning of documents is one level at which the nets
impact on Word.  The multiple versioning of language is another,
exemplified by the incorporation of Unicode into Office 2000 allowing the
program to begin to be used more fully across a massively expanded, though
not comprehensively resolved range of character systems.  The historical
genesis of computing in the West still leaves this material-semiotic
lockdown - the way in which text is still ordered from right to left for
instance - remains for instance as one of the most substantial legacy
problems to be dealt with.

	Another, perhaps more promising, way in which Word is opened up to
the net is again at a level producible only in the conjunction of
computers with networks.  Macro viruses such as Melissa and the Love Bug
exploit the monocultural strategy of Microsoft, in its Operating System,
in its applications as a vector for its own multiplication.  Absolute
standardisation, whilst infinitely smooth in its topological isomorphism
has the quality of breeding its own infinitely fast massively predatory
nomads.  Equally, if the strategy towards the language of the nets is
really as crude as that being put into place with Real Names its clean
world of unambiguous meaning is inevitably going to collapsed into the
subterranean plague pits of language, of allegory, of puns, of a thousand
mirrored misuses.


Typing with a raised fist


Office's capabilities remain very much the types of application that
defined and set the pace for the first era of personal computing.  They
have largely stayed in the same state but metastasised functions.  They
have been impacted, but not fundamentally changed by the nets and by the
multiple production of documents, rather than retaining a singular mode of
authorship.  And they have produced a sensorium that whilst it is a
near-disaster in terms of design and overloaded with the material-semiotic
infrastructure of business is still of course incapable of determining the
uses to which it is put.  That at least is probably some sort of
usefulness.

	What remains here is to ask what can be learnt from Word in terms
of the production of software which moves beyond the limits of the
increasingly closed models produced by an increasingly small number of
corporations for a vastly increasing number of users.

Is there the possibility of producing a software which aids and encourages
'autonomous work' in the terms that Gorz suggests?  That is, which
reiterates the process of becoming autonomous at a more profound level of
producing the amalgamation which the work progresses through, or whether
it is possible to develop software in mutuality with specific or numerous
drives and tendencies that somehow maximise escape from, not always simply
repurpose, the codifications of programs such as Word?  At the same time,
whilst autonomy is nice as a policy document, as a flag of convenience, it
exists a little too much on one plane, that of work, to fully accommodate
the sheer bandwidth of behaviour of life, of language.  What at the very
least is also needed to be those moments when what we know to be true,
what our certainties are about software, are lost.  Is it possible to
produce a software that joins with language in throwing up both moments of
realisation and for the user, devices by which it can become strange from
itself?

	The argument against this possibility is that software is somehow
neutral.  There is no need for a drive towards a reinvention of software
or what is sometimes awkwardly called a 'radical' software because at
bottom, all software is simply just algorithms24.  This is both
compellingly optimistic, in that it refuses to get locked into
determinism, but also a little more innocent than strictly necessary.=20
Mathematics is always situated, always developed in some sort of tune with
its moment in time.  Ancient Greek mathematics clearly developed in
relation to geometry, measurement and building. Why is there so much
attention put into producing algorithms today?  In order to run vast
amounts of changing data through sequences of uniform and repeatable
processes.

	This is not to say of course that uses of certain forms of
numerical technology are determined by inherent relations of power.  It is
again what they are amalgamated with that is of substantial importance.  
Not everything produced in Word is a CV.  However, acknowledging this
should not become a way of blocking processes of reinvention -
particularly not as a way of slipping a more thoroughgoing conformism
through as a way of seeming to, at least partially, confront the
problematics of software production.

As we have seen, software is reduced too often into being simply a tool
for the achievement of pre-existing neutrally-formulated tasks.=20 Culture
becomes an engineering problem.

	Where this is perhaps most telling is not in the area of
proprietary software, but in those areas of software production that have
made an important break with it.  Free, or Open Source software,
particularly that available under the GNU Public License, works because it
operates via of one of the most useful socio-technical inventions of
recent times25.  Those active in free software are increasingly aware that
it needs to develop beyond its core constituency of software written by
engineers to be used by engineers.26 This model has of course given us
some phenomenal successes but this process of openness needs itself to be
opened up. Just as Microsoft trapped Apple into its financial domain by
refusing to produce Office for their OS unless Explorer was bundled as
standard, proprietary software traps supposedly 'free' programmers into
their imaginal space by convincing them that they will have no users
unless they conform to what is already known, what is already done.

	The surplus rationalism that has given us tools for tasks may be
free in some senses but its springs a trap upon itself every time it stays
too happy to accept access to code without access to its conceptual
infrastructure.  Just as you can only fully reveal a phenomena if it is
already disappearing, Free Software is too content with simply
reverse-engineering or mimicking the cramped sensoriums of proprietary
software.  Copying Microsoft Word feature by feature and opening up the
source code is not freedom.  Mimesis is misery.

	Whilst there is at least the beginnings of a move to collaborate
with designers and other users to produce replica DTP packages and artists
to clone Photoshop - and thus pull the practice of engineers into relation
with currents it has been excluded from - there is a need to go further.  
Where it seems open or free approaches are most fruitful at present is in
small software, making specific interventions to precise technical,
economic and social problematics.  Particular bottlenecks to the
distributed circulation of information are broken through, often of course
using simple surplus rationalism as their crowbar.  Despite the legal
problems it encountered, the recent open DVD player, DeCSS, is a strong
example.  This strategy of focussing on precise technical conflicts needs
now to be intensified by drawing in antagonisms from supposedly separate
fields.  Geek drives to innovation must, as awkwardly and confusingly as
it will happen, be coupled with the drive to make language, to cut the
word up, open, and into process.




Matthew Fuller is a member of the speculative software group I/O/D
(http://bak.spc.org/iod/) and a regular collaborator with Mongrel
(http://www.mongrelx.org.uk/). Shake Editions
(http://www.shake-editions.com) have recently published his novel
'ATM'.  <bigger> This text accompanies 'A Song for Occupations' a large
wall-mounted map of the interaction sequences of microsoft
word</bigger>



1   The HTML generated automatically by Word is of such low quality,
full of extraneous code, that Dreamweaver 3 for instance makes a
feature of being able to automatically reformat it into useable HTML.

2 Gilles Deleuze and Felix Guattari, A Thousand Plateaus, trans. Brian
Massumi, Athlone, London 1988, p.90

3 For a good account of this earlier stage see: EDP Analyzer, The
Experience of Word Processing, in Tom Forester ed. Blackwell, Oxford 1980
p.232-244

4 James Gleick, Chasing Bugs in the Electronic Village,
http://www.around.com

5 John Hewitt, Good Design in the Market Place: The Rise of Habitat Man,
The Oxford Art Journal, Vol. 10 no2, Oxford 1987, p.28-42

6 Andr=E9 Gorz, Paths to Paradise - on the Liberation from Work, trans.
Malcolm Imrie, Pluto Press, London 1985, p.64 (He continues... ...By its
very nature autonomous behaviour cannot be explained sociologically, of
course it always occurs within a socially determined field, with socially
pre-given instruments. But both are reshaped in unforeseen ways to fit the
requirements of a personal venture)

7 Michel Foucault cited in Richard Sennett, The Corrosion of Character,
 W.W. Norton, New York 1998 p.13

8 Heim, p.10 This is a preamble to Heim's attack on McLuhan for getting
wasted in the volcano of the random fragmenting world.  See Also, Gary
Genosko, McLuhan and Baudrillard: masters of implosion, Routledge, London
1999

9 cited in Wendy Goldman Ruhm, The Microsoft Files: the secret case
against Bill Gates, Times Business, New York, 1998, p.256

10 Heidegger, cited in Heim, p. 195

11 Edwards, The Closed World: computers and the politics of discourse in
cold war america, MIT Press, Cambridge, 1996, p.147

12 Heim, p.92

13 Heidegger, The Question Concerning Technology

14 Alan Coooper, The Inmates are Running the Asylum: why high-tech
products drive us crazy and how to restore the sanity, Sams Publishing,
Indianapolis, 1999, p.106-107

15 Tufte, Envisioning Information, Graphics Press, Cheshire, Connecticut,
1990, p.89

16 Edward Tufte, p.53

17 Malcolm McCullough, Abstracting Craft: the practiced digital hand, MIT
Press, Cambridge, 1998, p.80

18 Reflection Eternal, feat. Bahamdia, Chaos , SoundBombing volume two,
Rawkus Records, New York, 1999

19 Discourse Networks 1800/1900, trans. Michael Metteer with Chris
Cullens, Stanford University Press, Stanford, 1990

20 Lecercle, The Violence of Language, Routledge

21 Jan van Dijk, The Network Society, trans. Leontine Spoorenberg, Sage,
London, 1999, p175

22 Heim specifies a number of early word-processing tools (ie
=46ramemaker) in which this capacity was still present.

23 =20 http://www.microsoft.com/presspass/features/2000/03-14realnames.asp

24 For a version of this argument in Linguistic terms see, J.Stalin,
Marxism and Problems of Linguistics at http://www.marxists.com

25 see the Free Software Foundation Website:  http://www.fsf.org

26 Matthew Mastracci, LInux Office Showdown (Oart 1: Applixware 5.0 M1 vs
KOffice pre-beta) at http://slashdot.org/features/00/02/22/0850251.shtml

with an excellent following thread on free word processors and office
suites including various short suggestions for a conceptual as well as
technical opening up of this area



#  distributed via <nettime>: no commercial use without permission
#  <nettime> is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: majordomo@bbs.thing.net and "info nettime-l" in the msg body
#  archive: http://www.nettime.org contact: nettime@bbs.thing.net