[Open Manufacturing] Russia and external software dependencies
Bryan Bishop
I found this to be an excellent example of tech dependency awareness.
In this case, it's only software:
http://linux.slashdot.org/article.pl?sid=09/01/23/1450224
"According to Russian media, the Russian Government is going to
develop a National Operating System (Google translation; Russian
original) to lower its dependencies on foreign software technology
licensing. The Russian plan will base its efforts on Linux and expects
a worldwide impact. Microsoft is also involved in the roundtable
process that led to the recommendation. The Chinese government
successfully lowered its Microsoft licensing costs through an early
investment in a national Linux distribution. I wonder if other large
markets, such as the European Union, will also develop their own Linux
distributions or join in the Russian initiative."
One of the comments:
http://linux.slashdot.org/comments.pl?sid=1102319&cid=26576427
"It's about time! Why any country would voluntarily base their
national security on imported, closed-source, non-free software is
beyond my reasoning. If a country wants to control its infrastructure,
it must use free software. Same goes for us computers users, too, of
course, but the stakes are much higher for a sovereign nation."
Although it's not just about national security, it's also the issue of
just having people who know what they are doing locally. I am sure
there are many Russian F/OSS programmers, but what about countries
that simply don't keep track of their tech dependencies? For instance,
individuals don't know how to manufacture integrated semiconductor
devices [well, some used to, but now it's off-limits as big business].
As that comment points out though, the "stakes" are much higher for a
sovereign nation- though I feel it's just as important for a sovereign
individual, community, or any group.
racas...@gmail.com
I'm always surprised that no other OS has been successul for PCs. Of
course Windows is not the same as when it was 3.0. But I'm always
surprised.
There were Digital Research DR-DOS, PC-DOS from IBM and Compaq DOS
beyond MS-DOS. There was OS/2, which worked like and better than Win
98 when just Win 3.11 existed...
The only risk is that countries without enogh ingrastructure or taking
wrong decisions would be just switching an techno-economical
dependency for a techno-political dependency. And this is not because
of Russia. It could be France, Jamaica or Botswana the originator of
the new product.
Anyway, this is a clear example of the possibilities of a well thought
independece concept.
Regards,
Raúl
Ton Zijlstra
Is independency key here, or merely the absence of dependency (=assymetric power balance). What about interdependency (= symmetric power balance)? Would a certain amount of interdependency not give a higher level of resilience than just independency (as independence also means no additional 'outside' safety net yet additional risks of single point of failure). Assuming resilience is what communities/societies/nation states are after.
Ton
-------------------------------------------
Interdependent Thoughts
Ton Zijlstra
t...@tonzijlstra.eu
+31-6-34489360
http://zylstra.org/blog
-------------------------------------------
Bryan Bishop
> Is independency key here, or merely the absence of dependency (=assymetric
> power balance). What about interdependency (= symmetric power balance)?
> Would a certain amount of interdependency not give a higher level of
> resilience than just independency (as independence also means no additional
> 'outside' safety net yet additional risks of single point of failure).
> Assuming resilience is what communities/societies/nation states are after.
I can't answer your question straight up. What is a dependency?
Essentially, a dependency is something that must occur before
something else can happen. When that something else is increasingly
distancing itself from the 'origin point' of your sphere of influence,
you are increasing the number of steps or the number of things that
must occur before that original dependency is met. When you have it
localized, and it's close within your 'sphere of influence', as in,
the set of things which you can really physically actually act upon,
when that's so: the number of steps for that dependency to be met can
be minimized to your liking. The number of steps necessary to
successfully meet a dependency is the 'distance' from it to you.
Each step is not perfect (we all know this) - all technologies can
break, all of them have problems. As you increase the number of steps,
you're increasing the likelihood of something going wrong on the way
to the end of the chain of steps (since one step must precede the
next). If these steps are just software executing on a processor, the
likelihood of something going wrong is not much, but on the other
hand, what if one of these steps is something about politicians, and
is dependent on the way that somebody else feels? Then you don't
really know the likelihood of if the steps are going to work. I
suppose you could personally draw your own probability distribution
for a certain person between "you" and "success", but really, putting
people in the way just seems absurd. Why would you want to jeopardize
friendship (or, more broadly, fellow people) like that? And anyway,
trying to predict the feelings of somebody else, now that's science
fiction. There are all sorts of rules that societies have built around
to try to make it so that when somebody doesn't feel like 'doing their
job' or making good on a promise, there are actions you can take- like
lawsuits, going to court, appealing to some more authoritative figure,
and so on, but that takes a lot of time that mission critical
applications (like life?) might not have.
I'm reminded of Douglas Adams' Vogons and their legendary bureaucracy.
http://en.wikipedia.org/wiki/Vogon
"""
The Vogons are a fictional alien race from the planet Vogsphere in The
Hitchhiker's Guide to the Galaxy series by Douglas Adams. Vogons are
vaguely humanoid, bulkier than humans and with green skin. They are
employed as the galactic government's bureaucrats. Vogons are
notorious for their aggressive manner and terrible poetry.
In the series it is told that, far back in prehistory, when the first
primeval Vogons crawled out of the sea, the forces of evolution were
so disgusted with them that they never allowed them to evolve again.
Through sheer obstinacy, though, the Vogons survived (partly by
adapting a misplaced, badly malformed, and dyspeptic liver into a
brain). They then emigrated en masse to the Brantisvogon star cluster
(although the film has them staying on Vogsphere), where they form
most of the Galactic bureaucracy, most notably in the Vogon
Constructor Fleets (which, despite their name, patrol the galaxy
demolishing planets).
"""
Vogon surgeons would put out a requisition for a form to request a
scalpel for some general surgical cutting purpose, meanwhile a patient
would be bleeding to death on the table, meanwhile any competent human
surgeon would tend to just grab the scalpel off of a nearby table. The
Vogon paperwork is an example of doing something completely, utterly
senseless, meanwhile the answer is right in front of them. And if they
happened to get an independent thought, they would be "repeatedly
smacked in the face by the paddle creatures under the sand on
Vogsphere."
Anyway, there are a few ways to minimize SPOFs (single points of failure)-
http://en.wikipedia.org/wiki/Single_point_of_failure
(1) reducing complexity
(2) redundancy
(3) diversity
(4) transparency
((I like how the article begins the list with: "The strategy to
prevent from total systems failure is .."; probably deserves a close
treatment with the Systemantics book, constructals, etc.))
In terms of the original topic- Russia and external foreign software
dependency- we see that #3 (diversity) isn't much of an option for
fixing external dependencies. Increasing the total number of external
ways to meet dependencies is like increasing the number of different
forms that Vorgons could possibly file to requisition the scalpel. And
#2 (redundancy) would mean something like the Russians having
contracts with two Microsofts at the same time, and maybe near double
their computer infrastructure, which could be wasteful IMHO. Another
way that this could be implemented would be if Russia had an internal
Microsoft. #1 (reducing complexity) is an interesting idea, but I
suspect that when Microsoft is the only provider of the software, and
in other words there is only one actor in the system, the complexity
isn't significant. And #4 (transparency) is where open source software
would jump up and raise its hand. Russia would become a giant
metaphorical Gentoo operating system. Should the need ever arise, they
could compile everything themselves from scratch. And in the mean
time, external server farms doing the work for them would be fine.
This is taking the superficial view of things of course .. what if
Russia doesn't actually have any semiconductor manufacturing
facilities, no ability to build their own compile farms? I guess they
don't actually need to have compile farms, but they do seem to want
personal computers, which still would require that manufacturing
capacity-- do they have that dependency met, or are they relying on
some broad, generally undefined set of SPOFs?
The reason why I mentioned Systemantics, among other reasons, is
because of a vague memory of the problems with the definition of
system failure or even system success. After all, it is the main book
on the laws of system failure.
(8) The Operational Fallacy: The system itself does not actually do
what it says it is doing.
(12) A complex system cannot be "made" to work. It either works or it doesn't.
(25) The larger the system, the greater the probability of unexpected failure.
(26) "Success" or "Function" in any system may be failure in the
larger or smaller systems to which the system is connected.
Now, to get back to the question of interdependency that you
originally brought up, Ton, I feel that if just Russia is its only
SPOF for itself, then internal improvements can happen at its leisure
and truly, Russia will either live or die. I mean, it will unfold to
the extent that it can unfold and live its story and grow. Growth can
also occur with interdependency. Let's say that Russia doesn't have
the technological dependencies met to create all technologies of human
civilization, and it relies on some external agents (countries,
nations, states, companies, individuals, ...) to meet the dependency.
In this situation, if those externals (the countries, nations, states,
companies, individuals, ...) break, go up in flames, cease to exist,
it's like Russia experiences a *stroke*. In humans, a stroke can cause
the loss of function of important parts of the brain. My grandfather
once had a stroke near the end of his life, perhaps while at a
jewellery fair, and suddenly he could no longer work at the cash
register because he found he could no longer do basic arithmetic on
currency (though otherwise arithmetic was fine- how strange). I wish I
could have been with him at the hospital later that evening. Anyway,
with interdependence, you can be hit by a 'stroke'. I am reminded of
fairytale stories where suddenly the bridge acrosss a canyon goes out,
and townspeople are no longer able to trade, that sort of thing. The
other issue in interdependence is that you can route around damaged
nodes: you can find other ways of getting the same things while still
using external dependencies. Much like on the internet- theoretically,
or maybe even it has already happened, some of the packets that it
took to communicate this email might have been routed under the Black
Sea, or in other words to some place in the world completely most
definitely not in an obvious direct path between our two machines. So
what about that scenario for interdependence? Where you have multiple
suppliers? Where you have multiple possible interchangeable sets of
(external) steps to meet a dependency? Yes, it's certainly better and
to be encouraged than just working with one single supplier, sure. But
this issue of interdependence is more about people than it is about
technology. The technology doesn't need to be so many 'steps' away,
you don't need to put your fellow man or your friends in between you
and what you need to get done ("work"), you don't need some super
magical prediction theory of how people are going to feel, or your
probability distribution for whether or not some person is going to do
what they say- the technology is just going to *work* (unless it fails
for technical reasons). Interdependence between people is probably
necessary and important to personal and societal growth. This also
reminds me of another law of systemantics:
(7) Functionary's Falsity: People in systems do not actually do what
the system says they are doing.
^ this goes back to bureaucracy, laws and such in an attempt to fix
the system when people don't do stuff in current societies. These
law-based systems claim they are solving the problem of people
defaulting on their promises or whatever, but then we see this other
law of systemantics:
(8) The Operational Fallacy: The system itself does not actually do
what it says it is doing.
Open source and other initiatives, like from the free culture scenes,
are reforming parts of those systems about 'promises', licenses and
other deals between members in societies, so that these new
technologies that we have can do what previously we would never have
thought possible. That's why Paul runs around here talking about the
worrisome results of "putting post-scarcity tech in the hands of
scarcity-preoccupied individuals". Not to mention that each of these
people might be in differing stages of personal growth .. like the kid
who might want to make ten thousand stuffed-dolls, or the man who has
lost many of his friends over the years and faces health issues. These
men's health issues are actually why Mensheds Australia exists. They
encourage and support the construction of sheds kind of like techshops
and fablabs, except geared towards enabling men to do something other
than spend countless hours in the pub, simultaneously revising
craftmanship while the recent global tendency has been shifting away
of valuable skills to "the cloud" and leaving these men without roles
in their communities-
http://groups.google.com/group/openmanufacturing/msg/770f9bc0d0a4f364
http://www.mensheds.com.au/
((The "sphere of influence" model is just a metaphor that I am using.
While it would be interesting to see a mathematics of technological
dependencies, that might be some stuff in topology, set theory, graph
theory, or something like that, not quite 3D cartesian spheres. But
remember, the priority is the tech, not the math to describe the
possibilities. The map is not the territory and such.))
Nick Taylor
> I'm always surprised that no other OS has been successul for PCs. Of
> course Windows is not the same as when it was 3.0. But I'm always
> surprised.
Another thing of course is that it's actually quite hard to buy an off-the-shelf PC without windows installed,,, and (more importantly) its quite hard to get a job (in a lot of fields) unless you're adept with MS software.
MS software is designed to make human beings an easily swappable part. Back in the 90s when I worked for Ernst and Young, we (IT people) were euphemistically referred to as "resource"... (not "A resource", just "resource") like there was a lake (or cloud) of us somewhere and we could be deployed and discarded on a whim etc. We were viewed in a way not dissimilar to how cloud-hardware is viewed today. Ephemeral infrastructure.
I think that's something that yer free-market-fundamentalists don't really get. There are powerful systemic drivers for devaluing humans, and any technology (like MS Office or crossbows) that can do this, confers competitive advantage... which is great if you're on top, not so much if you're one of the great-unwashed. I think this is possibly why "labour-saving-devices" don't necessarily improve standards of living. Skilled work becomes less-skilled. Labour-saving = wage-lowering.
Democratisation is often a little more complex than it looks... and the chief danger of The Singularity may not be that "machines don't need us" so much as "employers don't need us"... which for a lot of people is already essentially true. The TV program The Wire (genius) is specifically written about a world where the value of huge chunks of humanity is basically nothing. This isn't tomorrow, this is now.
Back in 1800s America, a slave would cost (in todays money) around... what was it? $30,000? They were valuable. Today you can get on a plane and pick one up in Haiti for about $50.
I think simple technologies for self-sufficiency are critical to our survival to be honest... not just for the vast swathes of people who are already dying from the lack of them, but for everyone.
But I digresss. Back to Linux etc.
Nick