The amount of technological voodoo in web-related things makes me
hesitant to respond here.  However, here goes.  At a very quick glance,
I notice a few things.

First, the googerteller app seems like it does not intelligently
differentiate between kinds of web requests.  Innocuous cross-origin GET
requests without cookies or added query args will be treated the same as
other requests.

Additionally, it doesn't appear as if the app differentiates between IP's.
Can someone verify that the IP's in this one guy's minimal github
application are legitimately ones that collect data? The link from their
github page to the google list of IP's seems to indicate the list is too
large and will make a lot of web requests audible that are not
necessarily google related.

For example, there are many mom and pop websites without analytics or
telemetry that are hosted by google services (eg. virtual machines in
the cloud) and would show up under those IP ranges, afaict.  Other
popular services like cloudflare or whatnot might be using google
infrastructure underneath as well.  Assuming https termination, is that
not okay?

Second, the only google request I see when loading  are for the Roboto font and material design

I don't see cookies or anything else being sent with these requests to

You can verify this yourself if you open the developer tools in your
browser (safari, chrome, and firefox all have them now). Go to the
'network' tab and load the map application.

However, I do see something that isn't picked up by the googerteller
app. The map app is storing a google analytics cookie and sending it on
requests to

You can see the cookie in the browser developer tools under
'application->cookies' (the google analytics cookies all start with _GA)
and see them being sent under the 'network' tab. Again, these are only
being sent to and not google directly.  What the app
does behind the scene with that info is hidden from me, but I imagine it
is not passing it further along to google; in which case, it is no more
damaging than standard web logs. (note: in a private tab, I no longer
see this cookie, so maybe app had analytics once but got rid of them

An added note: Material design is google's design language
and something a LOT of folks use to make designing web applications
easier.  Using something like that means you don't have to waste time
thinking about hard issues like layout and various animated widgets, not
to mention bigger, more difficult, topics like accessibility in design.
Google's massive and well-paid team has put a lot of these thoughts into
usable design and code for you. I personally don't use it, but
understand why someone would. 

This is a cool project and it doesn't look like is sending any _harmful_ data to google directly.
(Please correct me if wrong here)

Without detracting from the conversation about the mapping project
itself (I think it deserves its own above the app layer), I would love
to hear more from nettime if there is a different assessment or more
enlightened ways to think about hosting your data and applications on
the web without "feeding the beast".  Is it as easy as avoiding to use
google analytics, or does one need to consider more?  What else?  Can we
accept the usage of google's open source artifacts, of which there are
many, as a trade-off to more easily develop applications?  Is there a
way of talking about these issues that simultaneously respects the
complexity and elasticity of the technology, but doesn't trail too far
into the weeds?


#  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:
#  archive: contact:
#  @nettime_bot tweets mail w/ sender unless #ANON is in Subject:

Reply via email to