[Noisebridge-discuss] [drama] New logo source?

[Jonathan Lassoff]

Excerpts from Kelly's message of Sun Jan 10 19:21:37 -0800 2010:
> Oh wow, that's awesome.  Is it not like... dynamic? Because I only get
> the CCC logo on the first page, although I do get the standard logo on
> several internal pages.
> 
> I like the idea of a randomized selector. Can you post how to add to the pool?

I believe Jacob, Leif, and Andy have access to this host.

I wonder how we could allow people to post images securely, without
compromising the securty of the host. Judging from the URLs for the
images being served now, it looks like they're just being placed in the
root directory for the web server.

One fair way I could see approaching this would be to have users upload
their alternate logo through the normal Mediawiki upload form, add it to
the Identity article, and then poke some form or something to add the
new image's URL to the rotation.

That being said, I think having more than one logo on the main, front
page is a bad idea. I think it dilutes the good brand Noisebridge has
built for itself so far.
Re-designing the logo is one thing, but saying, "all 50 of these images
are all official logos" is another, in my opinion.

* * *

I was curious how random the images are, so I made a small script to
loop over fetching the front page and couting which logo is presented. I
ran this loop 1000 times, twice. One a few minutes after the other. I
didn't want to hammer the server any harder, so it's hard to say if only
2000 tightly-spaced samples are statistically significant, but I'll let
you draw your own conclusions:

Run 1:

{:"background-image: url(https://www.noisebridge.net/NB-logo-red-black-med.png);"=>331.0, :"background-image: url(https://www.noisebridge.net/wunderground-noisebridge.png);"=>518.0, :"background-image: url(https://www.noisebridge.net/Pissingponynb-131.png);"=>151.0}
fetched 1000.0 times
got background-image: url(https://www.noisebridge.net/wunderground-noisebridge.png); 51.8 percent of the times
got background-image: url(https://www.noisebridge.net/NB-logo-red-black-med.png); 33.1 percent of the times
got background-image: url(https://www.noisebridge.net/Pissingponynb-131.png); 15.1 percent of the times

Run 2:

{:"background-image: url(https://www.noisebridge.net/wunderground-noisebridge.png);"=>522.0, :"background-image: url(https://www.noisebridge.net/Pissingponynb-131.png);"=>156.0, :"background-image: url(https://www.noisebridge.net/NB-logo-red-black-med.png);"=>322.0}
fetched 1000.0 times
got background-image: url(https://www.noisebridge.net/wunderground-noisebridge.png); 52.2 percent of the times
got background-image: url(https://www.noisebridge.net/NB-logo-red-black-med.png); 32.2 percent of the times
got background-image: url(https://www.noisebridge.net/Pissingponynb-131.png); 15.6 percent of the times

And the script:

==========================================================
#!/usr/bin/env ruby

require 'rubygems'
require 'hpricot'
require 'open-uri'

logos = {}
fetches = 0.to_f

1000.times do
        url = open('https://www.noisebridge.net/wiki/Index')
        if page=url.read then
                fetches += 1
                puts "got #{page.length} bytes"
                document = Hpricot(page)
                logo_path = document.at("#p-logo/a")["style"].to_sym
                if logos[logo_path] then
                        logos[logo_path] += 1
                else
                        logos[logo_path] = 1.to_f
                end 
        end 
end

p logos
puts "fetched #{fetches} times"
logos.each do |key,val|
        puts "got #{key} #{val/fetches*100} percent of the times"
end
==========================================================

Cheers,
jof