[Pkg-chromium-maint] Bug#581066: Chromium reports wrong UserAgent string

Lincoln de Sousa lincoln at minaslivre.org
Mon May 24 11:26:30 UTC 2010

On Sun, May 23, 2010 at 10:06:39PM -0500, Jonathan Nieder wrote:
> Hi,
> Lincoln de Sousa wrote:
> > The main reason to file this bug was think about the statistics saying
> > that I'm using a browser that I'm not =/.
> Thanks for the explanation.
> >> You can experiment with how sites react to various possibilities with
> >> the --user-agent command-line option (which is undocumented for some
> >> reason).
> >
> > Doing some tests here nothing seems to be different, neither google,
> > youtube and other google services that I was able to test without
> > registering.
> This is good to hear.  Maybe we should just change the Chrome/ to
> Chromium/ and see if anyone screams.

Humm, I can keep looking for any other unwanted side effect.

> >> Or maybe it would be good to fix all sites that care and just declare
> >> that we are a newish browser, like this:
> >> 
> >>  Mozilla/8.0
> >
> > Not a way to go at all =P
> I disagree: once the sites are fixed, this is the right way to go.
> There is no good reason for web pages to require detailed information
> about the browser with each request.  Not even statistics gathering ---
> javascript is a better tool for that.

Thanks for this explanation, now I think I get your point. I do
believe that we can work to improve sites that we can but my concern
is in the sites that we can't fix.

> There has been some discussion recently on the downsides of sending
> too much information about a client to web servers:
>   http://panopticlick.eff.org/browser-uniqueness.pdf

Very nice reference, thanks!

> > (don't think it is actually interesting in this discussion =P)
> I didn’t mean it as a workaround but as a development tool.

Again, now I got what you say =D

Lincoln de Sousa <lincoln at comum.org>

xmpp:lincoln at comum.org

+55 61 8212 9176

More information about the Pkg-chromium-maint mailing list