NNSquad - Network Neutrality Squad
[ NNSquad ] Re: Privacy Questions About the New "Google Cloud Print" Service
|
You're right -- I did go into rant mode as I was rushing between
things and may be overreacting but there is a real issue here. I do recognize
that cities do have a problem with budgets and will seek help from wherever
they can. It’s a mixed bag – I cite Skype as a positive
example of a solution that doesn’t depend on the network but it too
depends on its central directory service rather than a user-distributed
approach. I say user-distributed because the Skype controlled directory is probably
distributed to but not in a way that the user controls. My real frustration is the lack of effort to find fully decentralized
solutions. I don’t necessarily expect this from the current players because
they don’t have the incentive to find an altruistic solution (though I
can argue that they would benefit from a larger market). This isn’t morally
evil – just a response to their local incentives. I’ve noted the problems
with GENI focusing on solving problems solely within the network – it fits
the research paradigm that network people should research networks even if the
Internet should be seen as empowering applications rather than the network
itself. What concerns me is that interim solutions can breed
complacency that makes it more difficult to provide more powerful solutions. It’s
the problem we see with firewalls – they give us a sense of local
security but don’t give us end-to-end connectivity. The risk is that
challenging complacency may indeed come across as a rant because, after all, the
solution does work within a comfort zone. It’s only when things go wrong that
the problems come to the force. The reason that I’m so aware of the DNS problem is
that I had to think about how to define the relationship between elements in one’s
home – a light switch and a light fixture – in a way that works
even if you move the elements outside the home. Solutions that require going
out of the home such as looking up the information in the DNS don’t “just
work”. And, in the end (or at the end-points) you need solutions that don’t
depend upon a network operator. -----Original Message----- [ Please see my comments below -- Lauren ] I agree that this is the wrong direction - it takes the
idea of "Networks as a Service" to the next level by making us more
dependent upon a centralized service. It is the anti-Internet making us more dependent
upon a middle. Instead we should have a protocols that allow us to have
a direct relationship between end points. The problem with NATs is
the protocol and instead of adding another layer of dependency we need
protocols which support such direct relationships. As you note what happens in an emergency when I need to
print and can't get out of the local network? Imagine if you can't turn on a
light in your house without consulting a central server like the DNS. It's a
bug not a feature. Same goes for Schlage which wants to wants to charge me
$13 month for their smart door likes - I can't change who can open the door
without going through their central service. Is Google joining Cisco and the carriers in making us
increasingly dependent upon their omnipresence and benevolence so they can
monetize us? Sure seems that way. Communities that available themselves of Google Apps should
be very afraid. [ Bob, With all due respect, your
reply in this case has pushed the needle somewhat into the
"rant" zone. Google explicitly has solicited comments on
their cloud printing plans -- which to me says that they are not
engraved in stone. I believe that a reasonable case (as I
noted in my original message) can be made that in spite of the
considerable benefits that cloud printing can bring, there is also a
need for some level of native basic print capabilities for a
number of reasons. You might consider waiting
until Google responds on these issues before launching headfirst
into an "Evil Google/Cisco" tirade. As for monetization, keep
in mind that Google also explicitly assumes that they will not
be the only organization providing cloud print
processing. You could presumably set one up, if you wish. But I would
argue that irrespective of who runs the cloud printing operation, there
are still considerations that make native printing an
important design element as a fall-back or "high security"
choice in various situations and for suitably capable devices. Your trying to scare
communities from Google Apps is increasingly nonsensical
given the real world of municipal governments and budgets
these days. Even large cities like L.A. have proven themselves
incapable of reliably operating their local IT infrastructures
with available budgets and personnel, and the privacy and
security risks in that situation are pretty awful. Local doesn't
mean panacea. So both with the printing
situation in particular and IT more generally, it's a matter
of ideally not being locked wholly into narrow (often
emotionally-laden) pardigms, but rather having the appropriate options
available for each given situation. Easier said than done, of course.
-- Lauren Weinstein
NNSquad Moderator ] - - - -----Original Message----- From:
nnsquad-bounces+nnsquad=bobf.frankston.com@nnsquad.org [mailto:nnsquad-bounces+nnsquad=bobf.frankston.com@nnsquad.org]
On Behalf Of Lauren Weinstein Sent: Friday, April 16, 2010 16:34 To: nnsquad@nnsquad.org Subject: [ NNSquad ] Privacy Questions About the New
"Google Cloud Print" Service Privacy
Questions About the New "Google Cloud Print" Service
http://lauren.vortex.com/archive/000707.html Greetings. Google Labs has just published
preliminary specifications (and an associated call for comments and feedback) for a
"print anywhere" service that they are implementing, called
"Google Cloud Print" ( http://bit.ly/955NSl [ Google Code ] ). The project has the very laudable goal of reducing OS
print stack complexities and enabling the ability for users of the
Google Chrome OS (and potentially other systems) to send print jobs to
any cooperating printer, anywhere in the world. Google
hopes that printer manufacturers (or third parties) will implement
supporting protocols in printer firmware, though for now proxy software will
be used to bridge between users and hardware. It is not immediately clear to me from the available
documents to what extent this model would ultimately be extended to the
Google Chrome browser. While the goal of "universal printing" is
wonderful, there are some tricky non-technical issues that immediately come to
mind. For example, the Google documentation states: "Google Chrome OS will use Google Cloud
Print for all printing. There is no print stack and there are
no printer drivers on Google Chrome OS!" This has two immediate and obvious implications. It
would appear that if you don't have an Internet connection at any given
time, you would seemingly have no way to print under this model.
And perhaps of even greater concern to some individuals and organizations,
any document that you wished to print would need to transmitted off of
the local system for cloud processing before it could even print on
a printer attached to the same local computer. While Google explicitly assumes that various
organizations will implement their own cloud processing services that meet
the required specs, at least initially (and presumably for most users
for some time to come) Google itself would be the likely cloud print processor. The willingness of individuals to send sensitive print
jobs through a remote processing point, simply to have them end up back
on the printer sitting on the desk next to them, seems
problematic in at least some cases. Firms or organizations with
ongoing security concerns and related policies (law firms, law
enforcement, other government agencies, and so on) may also balk at this
model (or simply be prevented from using it due to privacy and/or security
regulations) -- even if high-grade encryption protects
the documents in transit and the cloud processing service promises to
delete all associated data immediately after print processing. Google's Chromium OS holds a great deal of promise, and I
have very high hopes for its wide deployment and success. But
my gut feeling is that any OS or system that depends solely on remote
processing for local printer operations will find its adoption hobbled
in many environments by the kinds of concerns discussed above. While I most definitely understand the technical
attraction of reducing local print processing complexity, I would urge reconsideration of the assumption that a 100% reliance
(as I read the documents, anyway) on cloud-based printing can be an
adequate substitute for at least basic local printing capabilities
that do not depend on Internet connections and cloud services to
operate. --Lauren-- Lauren Weinstein lauren@vortex.com Tel: +1 (818) 225-2800 http://www.pfir.org/lauren Co-Founder, PFIR - People For Internet Responsibility -
http://www.pfir.org Co-Founder, NNSquad - Network Neutrality Squad -
http://www.nnsquad.org Founder, GCTIP - Global Coalition for Transparent Internet Performance -
http://www.gctip.org Founder, PRIVACY Forum - http://www.vortex.com Member, ACM Committee on Computers and Public Policy Lauren's Blog: http://lauren.vortex.com Twitter: https://twitter.com/laurenweinstein |