[nSLUG] Data sharing BOF

Mike Spencer mspencer at tallships.ca
Fri Jun 6 00:31:44 ADT 2014


How'd your BoF meet on data sharing go, Ben?  Anybody take notes and
do a summary of thoughts? 

Johann Tienhaara <jtienhaara at yahoo.com> wrote:

jt> Peer-to-peer data sharing has been a philosophical and, to some
jt> extent, hobby obsession of mine for almost a decade.

In the Old Days of the net, when there were only, say, 10^4 computers
on the net, it was taken for granted that if you wanted to share data,
you ran an ftp server or offered an open telnet.  Later there was
gopher and the Andrew File System. [1]

Knowing that, back circa 1994, I was at a meeting where the MT&T guys
came to give a presentation on the (then nascent and upcoming)
internet service to the home.  I asked them if $I_NOW_FORGET_WHAT that
they described wouldn't interfere with people running mail, web or ftp
servers.  Oh, we don't *want* you to do that!", said they.

Minor epiphany: this new thing wasn't an "information highway", wasn't
reticular, it was an information pipeline, planned to be intrinsically
hierarchical and essentially one-way.  But only people who'd used the
net under ARPA or NSF would realize that this was a major change for
the roll-out of the prole-grade Intertubes.

jt> The time is ripe for us geeks to band together and start
jt> presenting viable peer-to-peer data sharing mechanisms. Many geeks
jt> seems to be upset about the privacy, data ownership, and security
jt> which users gladly and freely hand over to big companies.

Yeah.  I can't figure how any enterprise that takes due diligence
seriously can use the cloud; even less how anyone with a healthy level
of paranoia can put their whole lives on Google's or Facebook's
servers.

I know Ben's remarks (30 May 2014 06:54:47) were a of slightly
different take:

ben> The idea here is that personal improvement by open source software
ben> (not just for Linux, but also Android and other platforms) allows
ben> us to set our own goals, track them, and help each other in their
ben> own tracking regimes by sharing our data and observations with
ben> friends online.

An interesting notion.  I may email knowledgeable friends about how much
boron to give to the turnips or anecdotal info about the hepatic risks
of Tylenol.  I never thought of doing it systematically.

ben> Unfortunately, the popular software that fills this niche today
ben> seems to work by submitting your data to a web site with a (sadly,
ben> fairly standard) privacy policy that both provides no guarantee as
ben> to the privacy of your data, and also claims ownership of the data
ben> you submit.

I think that's just insane.

ben> On and off for months I've had some ideas to address this on the
ben> back burner, and recently I've started doing some tracking of my
ben> own with a non-free app using this kind of website.

Would you like to enumerate two or three of those ideas in more detail
here?  Are you thinking of biometrics like a cardiac monitor that
tracks your EKG all day, reporting by wireless somehow?

Twenty-five years ago, there was a prof at Dal who was part of a study
in which subjects made notes of what they were doing and where they
were every (IIRC) 15 minutes all day. (I never knew just what the
sociologists were going to extract from that.)  Now the NSA is (at
least capable of) doing it for us but we don't get to see the results.
I gather that some people (twits?) now do that with twitter, all
archived under the corporate control and wobbly privacy you mention
above.

More detailed suggestions, please?

- Mike


[1] I once made the mistake of doing a "find / -name $WHATEVER -print"
    on an Andrew-connected system.  That was how I learned what Andrew
    was for. :-o

-- 
Michael Spencer                  Nova Scotia, Canada       .~. 
                                                           /V\ 
mspencer at tallships.ca                                     /( )\
http://home.tallships.ca/mspencer/                        ^^-^^


More information about the nSLUG mailing list