[NetBehaviour] Know Your Filesystem (and how it affects you)
aymeric at kuri.mu
Wed Oct 28 11:03:14 CET 2015
Thanks Dave for the text and cheers to Furtherfield for publishing it as
well as hosting this discussion.
I'm very much looking forward to tomorrow's event. Actually, some of the
points discussed in this thread so far, relate quite closely to the
topic of my presentation for RWX, namely how nowadays modes of
production and social structures are sandboxed both at a technological
and juridical level.
Some of my comments below are derived from this idea:
Rob Myers said :
> Xanadu was started a decade before the Star project, and Computer
> Lib/Dream Machines was published in 74. There may be something to the
> idea of epochs, or at least eras. :-)
> UNIX's file/directory system is no more "natural" than DOS/Windows'
> version with the slashes going the other way. They and the desktop
> file/folder metaphors contrast with other historical filesystems: BeOS's
> database, VMS's versioned file system, the Lisa's search system, the
> original Mac's flat list of files. It's more weird that UNIX won than
> that other systems do it differently.
There has been a Cambrian explosion of OS research during the 60s
up to the early 80s, but the rise of the computer industry made it so
that such diversity was only seen as a pool of competitive products from
which a few were meant to survive. The Unix-like operating systems
won unintentionally this race because they had a simple way to implement
portability, time-sharing, process interoperability and showed promises
of standardisation (the latter which led to a big failure during the
so-called Unix wars), but most importantly Unix benefited from a
proto-free software distribution model in its early days, and this
benefited a lot to AT&T to turn it into a commercial product once the
telco, liberated from its monopoly position, was able to sell its UNIX
branded software, after a decade of development and distribution as part
of a "fellowship" (term employed by Unix authors themselves). Within
this fellowship the social structure linked to the production of
software was closely informed and reflected from news ideas in computer
science, most notably the notions of library and utility (that were
initially explored in Project MAC during the mid-60s), and became quite
explicit with Unix, where cooperation and collaboration followed similar
patterns of modularity and reuse in the couples software-software,
human-software, and human-human.
If Bell Labs had tried to turn Unix into a commercial product right away
in the early 70s and if its development had depended exclusively on its
immediate commercial success, I am not sure if Unix would have been so
widely present nowadays. Maybe there is a parallel universe where smart
phones are not Unix based (like Android, iOS) but instead Lisp machines?
Anyway, from a technical perspective, as the preface of the UNIX-haters
handbook says, from the top of my head, Unix won because it was "good
enough", and that's also why an OS like Plan 9 which was supposed to
replace Unix eventually never got adopted because Unix like systems were
doing just fine with the odd duct tape fix every now and then.
In fact it is quite interesting to realise that much of the smartness
Dave Young said :
> > I find the use of the term "smart operating systems" strange.
> The term 'smart phone', like 'cloud computing', is the produce of the
> dark arts of corporate tech marketing, happily echoed by the likes of
> The Verge, Engadget, Guardian Tech, etc.
... is built on top of development approaches where it is quite common
to avoid qualifying tools as "good" but as "sucking less" than others,
and where the playful cleverness of the hacker cohabits with crude hacks
and piles of temporary fixes, and where ultimately, methods such as Keep
It Simple Stupid (KISS) contrasts with whatever smartness is added
during the marketing stage. To some extend an automagical tech slogan
like "it just works!" could easily be reversed to another interpretation
as something that in fact *merely and barely* works.
So indeed it should not be a surprise that todays computational culture
is reduced to the making and polishing of fanciful user interfaces and
user experience, UI and UX, and ...
> I think we are really losing the entitlements that come with
> user-agency and tool-ownership as a consequence of these
> 'smart operating systems' and their reluctance to share their
> dirty laundry (filesystems, background processes,
> data-caching, and so on) with us - should we ask them to.
... as a strategy to hide the dodgy business behind these facades.
In that regard the way Android permissions and sandboxing are
implemented on top of the Unix user/group system is quite illustrative
of both the crudeness and flimsiness of these things and how the notion
of home folders and user files has became completely irrelevant in the
age of smart phones being in fact dumb terminals for remote services.
And of course this very crudeness and flimsiness are reflected in the
way Android/Google has treated free and open source software projets,
that is to say as a pool of disposable code and ideas that can be used
to quickly test and assemble new products and services. This attitude
also feeds back into the community of hacking prosumers surrounding
these new systems where the pimping of ROMs and rooting the latest
gadget seems to be the only thing that matters.
But this is not so much different from the Unix landscape I quickly
described above, in both case these techno-legal construction allows the
formation of sandboxes in which certain lifestyles and beliefs can be
structured and articulated.
So in that sense ...
Dave Young said :
> > As we lose the ability to exercise our freedom we also lose the
> > awareness that we can?
> I think this is true, but also the incredibly powerful sense of
> seduction around these devices means that people may not actually -want-
> freedom in the first place. Many would rather a beautiful object in rose
> gold with the latest hype feature (fingerprint sensors, 4k selfie
marc garrett said :
> If we were to take a Evgeny Morozov stance - in his eyes we are prisoners
> of these laptops, tablets, phones; via surveillance, self tracking, and
> gamification, and we are drowning under the weight of acceleration, through
> technological determined domination.
... is all correct.
By loosing touch with the underlying of these systems we are left in
a position where we are simply performing their function without giving
much thought about it, and we forget who we are, where were are, what we
want, what we can ... We forget that these are sandboxes. We just wait
for the next bleep, popup, and notification, which together form a new
notation system for composing neoliberal popping choreographies.
However, all that said, I also would like to point out that there is
another side of the story that is too often ignored in tech/net
criticism. For instance while metaphors like the cloud are obviously
problematic when seen from the viewpoint of manipulative and deceptive
obfuscating strategies, they also have an important value as the
blackboxing of the computational, networked and storage infrastructures
into one cloud, which permits the expression of other ideas once
detached from their subjacent complexity. In that interpretation the
cloud is not so much a metaphor but more of a symbol or abstraction
operating at a higher level of technological articulations.
As such, this also means that the question of affordances need also to
be put in the perspective of the context in which these systems are
used and expressed. Maybe from the artificial viewpoint of the software
itself, an app where its data are supposedly transparently synched to
the cloud can be described as alienating and opaque, but it should also
be analysed from a broader angle, at a higher level in the context of
the life of its user, and what function or value the latter is attaching
to it. So back to the idea of sandboxing, it should not be taken for
granted that the lifestyle and culture that emerge from it is fake,
inauthentic or meaningless for those that inhabits it.
More information about the NetBehaviour