You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-dev@lucene.apache.org by Ryan McKinley <ry...@gmail.com> on 2008/12/04 17:17:51 UTC

logo contest

We have discussed with the apache PRC (public relations committee),  
and they agree that the top choice in the logo contest should be  
disqualified for its similarity to the solaris logo.

Given the rules agreed upon in http://wiki.apache.org/solr/ 
LogoContest, the next step is for Solr committers to use the results  
of the community poll to decide what the official logo should be.

I posted the results here:
  http://people.apache.org/~ryan/solr-logo-results.html
If we count a vote that came in 12 hours late, the results are quite  
different:
  http://people.apache.org/~ryan/solr-logo-results-late.html

Using the direct scoring method agreed upon, the logo with the most  
points is:
  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png

However, it is tough to gauge the real intent/preference since the  
vote totals are so low.

I see two options:

1.  Have solr committers vote to accept:
  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png

2. Have a 'runoff' poll with the top contenders:
https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg
https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg


Following the rules strictly points to option #1, but I think option  
#2 may better reflect the original intent of the community poll.

Personally, I am happy with any of these options (and logos); I just  
want to make sure we have a process that everyone feels is/was fair.

ryan






Re: logo contest

Posted by Ryan McKinley <ry...@gmail.com>.
> I see two options:
> 1.  Have solr committers vote to accept:
>  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
> 2. Have a 'runoff' poll with the top contenders:
> https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
> https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
> https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg
> https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg
>
> Following the rules strictly points to option #1, but I think option #2 may
> better reflect the original intent of the community poll.
> Personally, I am happy with any of these options (and logos); I just want to
> make sure we have a process that everyone feels is/was fair.


I'll add that i have a slight preference for option #1 since it would
get this process done with sooner :)

ryan

Re: logo contest

Posted by Erik Hatcher <er...@ehatchersolutions.com>.
I'm with Noble.  #1 for me as well for the sake of making a decision  
and running with a quality logo sooner rather than later.

ObBiasTransparency: Steve Stedman,the designer of sslogo* submissions,  
is a good friend of mine.  Awesome dude.  He does really nice work.

	Erik

On Dec 4, 2008, at 11:36 AM, Noble Paul നോബിള്‍  
नोब्ळ् wrote:

> I prefertaking option #1 and not delaying this any further
>
> On Thu, Dec 4, 2008 at 9:57 PM, Mark Miller <ma...@gmail.com>  
> wrote:
>> Hoss may lay down the rules on us, but if he doesn't (or if hes in  
>> a good
>> mood today), +1 on the runoff vote.
>>
>> Ryan McKinley wrote:
>>>
>>> We have discussed with the apache PRC (public relations  
>>> committee), and
>>> they agree that the top choice in the logo contest should be  
>>> disqualified
>>> for its similarity to the solaris logo.
>>>
>>> Given the rules agreed upon in http://wiki.apache.org/solr/LogoContest 
>>> ,
>>> the next step is for Solr committers to use the results of the  
>>> community
>>> poll to decide what the official logo should be.
>>>
>>> I posted the results here:
>>> http://people.apache.org/~ryan/solr-logo-results.html
>>> If we count a vote that came in 12 hours late, the results are quite
>>> different:
>>> http://people.apache.org/~ryan/solr-logo-results-late.html
>>>
>>> Using the direct scoring method agreed upon, the logo with the  
>>> most points
>>> is:
>>>
>>> https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
>>>
>>> However, it is tough to gauge the real intent/preference since the  
>>> vote
>>> totals are so low.
>>>
>>> I see two options:
>>>
>>> 1.  Have solr committers vote to accept:
>>>
>>> https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
>>>
>>> 2. Have a 'runoff' poll with the top contenders:
>>>
>>> https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
>>>
>>> https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
>>>
>>> https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg
>>>
>>> https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg
>>>
>>>
>>> Following the rules strictly points to option #1, but I think  
>>> option #2
>>> may better reflect the original intent of the community poll.
>>>
>>> Personally, I am happy with any of these options (and logos); I  
>>> just want
>>> to make sure we have a process that everyone feels is/was fair.
>>>
>>> ryan
>>>
>>>
>>>
>>>
>>>
>>>
>>
>>
>
>
>
> -- 
> --Noble Paul


Re: logo contest

Posted by Noble Paul നോബിള്‍ नोब्ळ् <no...@gmail.com>.
I prefertaking option #1 and not delaying this any further

On Thu, Dec 4, 2008 at 9:57 PM, Mark Miller <ma...@gmail.com> wrote:
> Hoss may lay down the rules on us, but if he doesn't (or if hes in a good
> mood today), +1 on the runoff vote.
>
> Ryan McKinley wrote:
>>
>> We have discussed with the apache PRC (public relations committee), and
>> they agree that the top choice in the logo contest should be disqualified
>> for its similarity to the solaris logo.
>>
>> Given the rules agreed upon in http://wiki.apache.org/solr/LogoContest,
>> the next step is for Solr committers to use the results of the community
>> poll to decide what the official logo should be.
>>
>> I posted the results here:
>>  http://people.apache.org/~ryan/solr-logo-results.html
>> If we count a vote that came in 12 hours late, the results are quite
>> different:
>>  http://people.apache.org/~ryan/solr-logo-results-late.html
>>
>> Using the direct scoring method agreed upon, the logo with the most points
>> is:
>>
>>  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
>>
>> However, it is tough to gauge the real intent/preference since the vote
>> totals are so low.
>>
>> I see two options:
>>
>> 1.  Have solr committers vote to accept:
>>
>>  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
>>
>> 2. Have a 'runoff' poll with the top contenders:
>>
>> https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
>>
>> https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
>>
>> https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg
>>
>> https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg
>>
>>
>> Following the rules strictly points to option #1, but I think option #2
>> may better reflect the original intent of the community poll.
>>
>> Personally, I am happy with any of these options (and logos); I just want
>> to make sure we have a process that everyone feels is/was fair.
>>
>> ryan
>>
>>
>>
>>
>>
>>
>
>



-- 
--Noble Paul

Re: logo contest

Posted by Chris Hostetter <ho...@fucit.org>.
: Hoss may lay down the rules on us, but if he doesn't (or if hes in a good mood

For the record: i'm (almost) always i na good mood -- it's just hard to 
tell because i spell like an angry unibomber wanna-be and i have a moral 
objection to using emoticons so my email based sarcasim is very, very, 
dry.


-Hoss


Re: logo contest

Posted by Mark Miller <ma...@gmail.com>.
Hoss may lay down the rules on us, but if he doesn't (or if hes in a 
good mood today), +1 on the runoff vote.

Ryan McKinley wrote:
> We have discussed with the apache PRC (public relations committee), 
> and they agree that the top choice in the logo contest should be 
> disqualified for its similarity to the solaris logo.
>
> Given the rules agreed upon in 
> http://wiki.apache.org/solr/LogoContest, the next step is for Solr 
> committers to use the results of the community poll to decide what the 
> official logo should be.
>
> I posted the results here:
>  http://people.apache.org/~ryan/solr-logo-results.html
> If we count a vote that came in 12 hours late, the results are quite 
> different:
>  http://people.apache.org/~ryan/solr-logo-results-late.html
>
> Using the direct scoring method agreed upon, the logo with the most 
> points is:
>  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png 
>
>
> However, it is tough to gauge the real intent/preference since the 
> vote totals are so low.
>
> I see two options:
>
> 1.  Have solr committers vote to accept:
>  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png 
>
>
> 2. Have a 'runoff' poll with the top contenders:
> https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png 
>
> https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg 
>
> https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg 
>
> https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg 
>
>
>
> Following the rules strictly points to option #1, but I think option 
> #2 may better reflect the original intent of the community poll.
>
> Personally, I am happy with any of these options (and logos); I just 
> want to make sure we have a process that everyone feels is/was fair.
>
> ryan
>
>
>
>
>
>


Re: logo contest

Posted by Andrzej Bialecki <ab...@getopt.org>.
Ryan McKinley wrote:

> I see two options:
> 
> 1.  Have solr committers vote to accept:
>  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png 
> 
> 
> 2. Have a 'runoff' poll with the top contenders:
> https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png 
> 
> https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg 
> 
> https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg 
> 
> https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg 
> 
> 
> 
> Following the rules strictly points to option #1, but I think option #2 
> may better reflect the original intent of the community poll.

I prefer option #2 as well.



-- 
Best regards,
Andrzej Bialecki     <><
  ___. ___ ___ ___ _ _   __________________________________
[__ || __|__/|__||\/|  Information Retrieval, Semantic Web
___|||__||  \|  ||  |  Embedded Unix, System Integration
http://www.sigram.com  Contact: info at sigram dot com


Re: logo contest

Posted by Ryan McKinley <ry...@gmail.com>.
On Dec 10, 2008, at 3:41 PM, Yonik Seeley wrote:

> On Wed, Dec 10, 2008 at 3:30 PM, Mike Klaas <mi...@gmail.com>  
> wrote:
>>
>> On 8-Dec-08, at 10:47 AM, Mike Klaas wrote:
>>
>>> On 7-Dec-08, at 7:40 PM, Chris Hostetter wrote:
>>>
>>>>
>>>> : I would personally prefer more of an "elimination"-style vote  
>>>> (i.e.,
>>>> STV).
>>>>
>>>> Ah... yeah, that seems like it would be a more fair way to deal  
>>>> with
>>>> things then my suggestion, and it doesn't violate the spirt of the
>>>> rules as original outlined (it's still a vote of ranked  
>>>> preferences).
>>>> Are
>>>> you volunteering to do the vote counting Mike?
>>>
>>> Sure thing.
>>
>> I take it that there are no objections?  If so, I'll call a vote by  
>> the end
>> of the week.
>
> +1
> I just wish we had used this method with the community vote.
>
> I guess as a committer I should try and figure out what order the
> community would have voted and do that.
>

Yes -- I am only comfortable with a 'committer only' vote if it is  
clear we are all operating to figure out what the first community vote  
intended -- not to select our own preferences.  (unless it is an  
explicit veto)

ryan



Re: logo contest

Posted by Chris Hostetter <ho...@fucit.org>.
: There you go.  clear as mud!

: As I see it, we either call a new community poll -- nothing in the rules says
: we can't.  or vote on:

FWIW: My opinion hasn't really changed since i said...

>> committers should cast their votes as they feel appropriate to best 
>> serve the interests of the community -- it's not really different then 
>> voting on an implementation approach for a feature, or what logging 
>> framework to use, or a decisison to switch from java 1.5 to 1.6 ... we 
>> have to make a subjective decision based on the feedback we've observed 
>> from the community as a whole (with solr-logo-results.html serving as 
>> our cliff notes) 

... i suggest we move forward with a straight STV/IRV based vote of 
committers (if anyone was going to VETO a submission i'm sure they would 
have done it by now) and let individual committers rank things the way 
they feel best represents the wishes of community.  since we have to 
subjectively interpret the community vote, let's apply individual 
subjective interpretations and aggregate them (in line with the orriginal 
contest rules) instead of picking one arbitrary subjective interpretation 
and then voting +1.

If we can't trust ourselves to represent the community fairly on picking 
a logo, we have no business committing any code.




-Hoss


Re: logo contest

Posted by Ryan McKinley <ry...@gmail.com>.
I added some more math to the script I used to generate the results  
page...  and what do you know: no official winner by any of the  
methods!  (assuming my math is correct)

check:
http://people.apache.org/~ryan/solr-logo-results.html
for a variety of attempts.

The problem with the STV method is that it continues until you meed a  
quota: (votes/seats+1)+1, in our case (19/2)+1=10.  Nothing ever gets  
to 10, so you will notice that stv==votes for everything!

So the next option is to try instant runoff voting:
http://en.wikipedia.org/wiki/Instant-runoff_voting

in that method, the losers of each round is tossed out and the vote  
continues until there is a majority.  Again, we never reach a  
majority, but the last two logos to lose are:
eliminating (4) https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
eliminating (4) https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png

Other things to perhaps consider.  Using the "scoring" method,
https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
is the winner.

The STV or vote counting method puts
https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
or
https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_b_red.jpg
at the top.

If we only count the first round of voting
https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
wins followed by:
https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_b_red.jpg

There you go.  clear as mud!

As I see it, we either call a new community poll -- nothing in the  
rules says we can't.  or vote on:

1. https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
last eliminated by instant runoff and it won the 'soring' method

2. https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
last eliminated by instant runoff and *may* have won if the other  
similar logos were not involved.

ryan

here is the hacked up code if you are curious...



     String[][] v = new String[][] {
       {
        "https://issues.apache.org/jira/secure/attachment/12394267/apache_solr_c_blue.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394265/apache_solr_b_blue.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394263/apache_solr_a_blue.jpg 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394366/solr3_maho.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394218/solr-solid.png 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394218/solr-solid.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394376/solr_sp.png 
",
        "https://issues.apache.org/jira/secure/attachment/12393951/sslogo-solr-classic.png 
",
        "https://issues.apache.org/jira/secure/attachment/12391946/apache_solr_burning.png 
",
        "https://issues.apache.org/jira/secure/attachment/12392306/apache_solr_sun.png 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394267/apache_solr_c_blue.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg 
",

        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394366/solr3_maho.png 
",
        "https://issues.apache.org/jira/secure/attachment/12393936/logo_remake.jpg 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394475/solr2_maho-vote.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394263/apache_solr_a_blue.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394353/solr.s5.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394376/solr_sp.png 
",
        "https://issues.apache.org/jira/secure/attachment/12393936/logo_remake.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394353/solr.s5.jpg 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394353/solr.s5.jpg 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394314/apache_soir_001.jpg 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394165/solr-logo.png 
",
        "https://issues.apache.org/jira/secure/attachment/12391987/solr.png 
",
        "https://issues.apache.org/jira/secure/attachment/12392306/apache_solr_sun.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394376/solr_sp.png 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394165/solr-logo.png 
",
        "https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg 
",
        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
       },
       {
        "https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg 
",
       },
       { /
         "https://issues.apache.org/jira/secure/attachment/12394218/solr-solid.png 
",
         "https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png 
",
         "https://issues.apache.org/jira/secure/attachment/12391946/apache_solr_burning.png 
",
         "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
       },
       {
         "https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png 
",
         "https://issues.apache.org/jira/secure/attachment/12394475/solr2_maho-vote.png 
",
         "https://issues.apache.org/jira/secure/attachment/12393995/sslogo-solr-70s.png 
"
       },
       {
         "https://issues.apache.org/jira/secure/attachment/12394350/solr.s4.jpg 
",
         "https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg 
",
         "https://issues.apache.org/jira/secure/attachment/12393995/sslogo-solr-70s.png 
",
         "https://issues.apache.org/jira/secure/attachment/12393936/logo_remake.jpg 
",
       },
       {
         "https://issues.apache.org/jira/secure/attachment/12394165/solr-logo.png 
",
         "https://issues.apache.org/jira/secure/attachment/12394475/solr2_maho-vote.png 
",
         "https://issues.apache.org/jira/secure/attachment/12394350/solr.s4.jpg 
",
         "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
         "https://issues.apache.org/jira/secure/attachment/12394314/apache_soir_001.jpg 
",
       },
       {
         "https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg 
",
         "https://issues.apache.org/jira/secure/attachment/12394350/solr.s4.jpg 
",
         "https://issues.apache.org/jira/secure/attachment/12394376/solr_sp.png 
",
         "https://issues.apache.org/jira/secure/attachment/12394267/apache_solr_c_blue.jpg 
"
       },
// LATE VOTE
//      {
//        "https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg 
",
//        "https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
",
//        "https://issues.apache.org/jira/secure/attachment/12394267/apache_solr_c_blue.jpg 
"
//      }
     };

     Map<String,Entry> votes = new HashMap<String, Entry>();

     for( String[] cast : v ) {
       int value = 5;
       for( String u : cast ) {
         Entry entry = votes.get( u );
         if( entry == null ) {
           entry = new Entry();
           entry.url = u;
           votes.put( u, entry );
         }
         entry.votes++;
         entry.score += value--;
       }
     }

     // go through each vote and remove:
     //

     // Toss out any votes for: solr2_maho_impression.png
     // but make the next vote #1
     List<List<String>> normalized = new ArrayList<List<String>>();
     for( String[] cast : v ) {
       List<String> pref = new ArrayList<String>();
       for( String u : cast ) {
         if( !"https://issues.apache.org/jira/secure/attachment/12394282/solr2_maho_impression.png 
".equals( u ) ) {
           pref.add( u );
         }
       }
       if( !pref.isEmpty() ) {
         normalized.add( pref );
       }
     }

     List<Entry> past = new ArrayList<Entry>();
     int quota = (normalized.size()/2)+1;
     for( int round=0; round<7&&past.isEmpty(); round++ ){
      // System.out.println( "round:"+round );
       for( List<String> pref : normalized ) {
         if( pref.size() > round ) {
           Entry e = votes.get( pref.get( round ) ); // should not be  
null
           e.stv++;
           if( round == 0 ) {
             e.first = e.stv;
           }
         }
       }

       // check if anyone is past the quota
       for( Entry e : votes.values() ) {
         if( e.stv >= quota ) {
           past.add( e );
         }
       }
     }

     List<Entry> entries = new LinkedList<Entry>();
     entries.addAll( votes.values() );

     PrintStream out = System.out;
     out.println( "<html><body>" );

     Collections.sort( entries, new PropertyComparator( "votes", true,  
false ) );
     out.println( "<h1>Sorted by number of votes</h1>" );
     writeTable( entries, out );

     Collections.sort( entries, new PropertyComparator( "score", true,  
false ) );
     out.println( "<h1>Sorted by score</h1>" );
     writeTable( entries, out );

     Collections.sort( entries, new PropertyComparator( "stv", true,  
false ) );
     out.println( "<h1>Sorted by Single transferable vote.  Quota:  
"+quota+"</h1>" );
     String u = "http://en.wikipedia.org/wiki/Single_transferable_vote";
     out.println( String.format("<a href=\"%s\">%s</a>",u,u) );
     writeTable( entries, out );

     Collections.sort( entries, new PropertyComparator( "first", true,  
false ) );
     out.println( "<h1>Sorted by first (excluding  
solr2_maho_impression.png)</h1>" );
     writeTable( entries, out );

     // Now use Instant-runoff voting
     Set<String> eliminated = new HashSet<String>();
     int majority = normalized.size() / 2;
     out.println( "<h1>Instant-runoff voting</h1>" );
     out.println( "<pre>" );
     Entry winner = null;
     for( int round=1; winner == null; round++ ) {
       Map<String,Entry> inRunning = new HashMap<String, Entry>();
       for( Entry e : votes.values() )  {
         if( !eliminated.contains( e.url ) ) {
           e.votes = 0;
           inRunning.put( e.url, e );
         }
       }
       out.println( "round: "+round );
       out.println( "entries: "+inRunning.size() );
       if( inRunning.isEmpty() ) {
         out.println( "finished with no winner." );
         break;
       }

       // Give each person a vote for the top remaing canidate
       for( List<String> pref : normalized ) {
         for( String p : pref ) {
           Entry e = inRunning.get( p );
           if( e != null ) {
             e.votes++;
             break;
           }
         }
       }

       // Check if anyone has a majority
       for( Entry e : inRunning.values() ) {
         if( e.votes > majority ) {
           winner = e;  // can't have two with a majority
         }
       }

       // remove the lowest score
       entries.clear();
       entries.addAll( inRunning.values() );
       Collections.sort( entries, new PropertyComparator( "votes",  
true, true ) );
       int lowestscore = entries.get( 0 ).votes;
       for( Entry e : entries ) {
         if( e.votes == lowestscore ) {
           eliminated.add( e.url );
           out.println( "eliminating ("+lowestscore+") "+e.url );
         }
       }
     }
     out.println("</pre>");

     out.println( "</body></html>" );
   }

   void writeTable( Collection<Entry> entries, PrintStream out ) {
     out.println( "<table border=1 cellpadding=20>" );
     for( Entry entry : entries ) {
       out.println( "<tr valign=\"top\">" );
       out.println( String.format( "<td>votes: %d<br/>score: %d<br/ 
 >stv: %d<br/>first: %d",
             entry.votes, entry.score, entry.stv, entry.first  ) );
       out.println( String.format( "<td><img src=\"%s\" /></td>",  
entry.url  ) );
       out.println( "</tr>" );
     }
     out.println( "</table>" );
     out.println( "<br/><br/>" );
   }



On Dec 10, 2008, at 4:59 PM, Mike Klaas wrote:

>
> On 10-Dec-08, at 12:41 PM, Yonik Seeley wrote:
>>>>
>>>> Sure thing.
>>>
>>> I take it that there are no objections?  If so, I'll call a vote  
>>> by the end
>>> of the week.
>>
>> +1
>> I just wish we had used this method with the community vote.
>>
>> I guess as a committer I should try and figure out what order the
>> community would have voted and do that.
>
> I could run the results of the community vote interpreted as STV, if  
> that would help (it'll be a few days, though).
>
> -Mike


Re: logo contest

Posted by Mike Klaas <mi...@gmail.com>.
On 10-Dec-08, at 12:41 PM, Yonik Seeley wrote:
>>>
>>> Sure thing.
>>
>> I take it that there are no objections?  If so, I'll call a vote by  
>> the end
>> of the week.
>
> +1
> I just wish we had used this method with the community vote.
>
> I guess as a committer I should try and figure out what order the
> community would have voted and do that.

I could run the results of the community vote interpreted as STV, if  
that would help (it'll be a few days, though).

-Mike

Re: logo contest

Posted by Yonik Seeley <yo...@apache.org>.
On Wed, Dec 10, 2008 at 3:30 PM, Mike Klaas <mi...@gmail.com> wrote:
>
> On 8-Dec-08, at 10:47 AM, Mike Klaas wrote:
>
>> On 7-Dec-08, at 7:40 PM, Chris Hostetter wrote:
>>
>>>
>>> : I would personally prefer more of an "elimination"-style vote (i.e.,
>>> STV).
>>>
>>> Ah... yeah, that seems like it would be a more fair way to deal with
>>> things then my suggestion, and it doesn't violate the spirt of the
>>> rules as original outlined (it's still a vote of ranked preferences).
>>>  Are
>>> you volunteering to do the vote counting Mike?
>>
>> Sure thing.
>
> I take it that there are no objections?  If so, I'll call a vote by the end
> of the week.

+1
I just wish we had used this method with the community vote.

I guess as a committer I should try and figure out what order the
community would have voted and do that.

-Yonik

Re: logo contest

Posted by Mike Klaas <mi...@gmail.com>.
On 8-Dec-08, at 10:47 AM, Mike Klaas wrote:

> On 7-Dec-08, at 7:40 PM, Chris Hostetter wrote:
>
>>
>> : I would personally prefer more of an "elimination"-style vote  
>> (i.e., STV).
>>
>> Ah... yeah, that seems like it would be a more fair way to deal with
>> things then my suggestion, and it doesn't violate the spirt of the
>> rules as original outlined (it's still a vote of ranked  
>> preferences).  Are
>> you volunteering to do the vote counting Mike?
>
> Sure thing.

I take it that there are no objections?  If so, I'll call a vote by  
the end of the week.

cheers,
-Mike

Re: logo contest

Posted by Mike Klaas <mi...@gmail.com>.
On 7-Dec-08, at 7:40 PM, Chris Hostetter wrote:

>
> : I would personally prefer more of an "elimination"-style vote  
> (i.e., STV).
>
> Ah... yeah, that seems like it would be a more fair way to deal with
> things then my suggestion, and it doesn't violate the spirt of the
> rules as original outlined (it's still a vote of ranked  
> preferences).  Are
> you volunteering to do the vote counting Mike?

Sure thing.

-Mike

Re: logo contest

Posted by Chris Hostetter <ho...@fucit.org>.
: I would personally prefer more of an "elimination"-style vote (i.e., STV).

Ah... yeah, that seems like it would be a more fair way to deal with 
things then my suggestion, and it doesn't violate the spirt of the 
rules as original outlined (it's still a vote of ranked preferences).  Are 
you volunteering to do the vote counting Mike?

Anyone have objections to using STV?

Background Info...
http://en.wikipedia.org/wiki/Single_transferable_vote
http://en.wikipedia.org/wiki/Instant-runoff_voting





-Hoss


Re: logo contest

Posted by Mike Klaas <mi...@gmail.com>.
On 4-Dec-08, at 2:33 PM, Chris Hostetter wrote:

>
> : Being the likely two candidates for winning.  My guess is that
> : narrowing to the two most popular options first would make #2 the
> : winner, while voting on the top 10 (w/o any strategy for winning)
> : would make #1 the winner.
>
> limiting to only voting for the top 2 seems unrepresentative since  
> more
> then one apache_solr_c_red.jpg variant tied for 2nd.
>
> : fun, fun.  So people who want one of these options to win should  
> vote
> : only for that option, really.
>
> Perhaps instead of just ranking top 5, we should ask committers to
> rank all of the choices on the final ballot to eliminate the
> "strategy" factor you are refering to ... i think we can trust all
> committers to understand this, but if someone botches it (or refuses?)
> we'll just shift the number of points each item earns down by the
> appropraite number (so if you want your 1st rank to earn 10
> points, you must list all 10, if you only list 4 then your top  
> ranked item
> only earns 4 points)

Eliminating strategic voting merely biases the outcome toward the logo  
without the vote splitting problem.  That is no solution.
It is better to allow strategic voting, as that is the only way for  
voters to express certain preferences in this system.

I would personally prefer more of an "elimination"-style vote (i.e.,  
STV).  Each voter lists the logos they prefer, in order.  The logos  
are ranked by first place votes.  The last in the rank is eliminated  
from the contest, and anyone who had that logo as their first-place  
vote has their vote transferred to the next logo on the list, if any.   
Iterate until two logos remain.  There is no danger of vote-splitting  
and the outcome maximizes global welfare in terms of binary  
preferences (well, probably not, due to Arrow's theorem, but it does a  
good job regardless).

-Mike

Re: logo contest

Posted by Chris Hostetter <ho...@fucit.org>.
: Being the likely two candidates for winning.  My guess is that
: narrowing to the two most popular options first would make #2 the
: winner, while voting on the top 10 (w/o any strategy for winning)
: would make #1 the winner.

limiting to only voting for the top 2 seems unrepresentative since more   
then one apache_solr_c_red.jpg variant tied for 2nd.
 
: fun, fun.  So people who want one of these options to win should vote
: only for that option, really.

Perhaps instead of just ranking top 5, we should ask committers to
rank all of the choices on the final ballot to eliminate the 
"strategy" factor you are refering to ... i think we can trust all 
committers to understand this, but if someone botches it (or refuses?) 
we'll just shift the number of points each item earns down by the 
appropraite number (so if you want your 1st rank to earn 10 
points, you must list all 10, if you only list 4 then your top ranked item 
only earns 4 points)
 
that won't violate anything in the rules as orriginally spelled out, and 
should help take into account the variant score dilution. (even though i 
don't think we should be overly accomidating this seems fair)

-Hoss


Re: logo contest

Posted by Chris Hostetter <ho...@fucit.org>.
: All apache_solr_c_red.jpg flavoured logos have a total score of 94.
: That should be taken into account IMHO and we should reduce the number
: of choices for these ones.

To re-iterate a comment I made in SOLR-84 that "wouldn't be fair to the 
people who have been submitting ideas and then retracting them and 
resubmitting variations based on feedback."

People were told many, MANY, times throughout the process that submiting 
multiple "variant" entries would risk diluting the votes.  One of the 
purposes of the long period for submissions was to give people time to 
post ideas, get feedback, and then tweak submissions and people who did 
that shouldn't be "excluded" from the final vote for following the rules. 
(sslogo-solr-finder2.0.png is a prime example of this)








-Hoss


Re: logo contest

Posted by Guillaume Smet <gu...@gmail.com>.
On Thu, Dec 4, 2008 at 7:34 PM, Yonik Seeley <yo...@apache.org> wrote:
> The methodology will very likely determine the outcome here, with
>
> https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
> https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg
>
> Being the likely two candidates for winning.  My guess is that
> narrowing to the two most popular options first would make #2 the
> winner, while voting on the top 10 (w/o any strategy for winning)
> would make #1 the winner.

+1.

All apache_solr_c_red.jpg flavoured logos have a total score of 94.
That should be taken into account IMHO and we should reduce the number
of choices for these ones.

-- 
Guillaume

Re: logo contest

Posted by Yonik Seeley <yo...@apache.org>.
The methodology will very likely determine the outcome here, with

https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg

Being the likely two candidates for winning.  My guess is that
narrowing to the two most popular options first would make #2 the
winner, while voting on the top 10 (w/o any strategy for winning)
would make #1 the winner.

fun, fun.  So people who want one of these options to win should vote
only for that option, really.

-Yonik


the two most popular would make the second option win, while expanding
it would make fir

On Thu, Dec 4, 2008 at 1:16 PM, Chris Hostetter
<ho...@fucit.org> wrote:
>
> : 1.  Have solr committers vote to accept:
> :  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
>
> The process as outlined on the wiki was that the commiters should have a
> ranked prefrence vote, after considering the point totals from the first
> vote. (with the added caveat that a -1 veto needs to be allowed since it's
> a vote to commit a change to the project)
>
> Considering the community prefrences expressed, I suggest that the
> committers hold a vote of the "high scoring" entries.  Picking a score
> of 10 as the cut off, that would give us 10 entries to vote on....
>
> https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
> https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg
> https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
> https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg
> https://issues.apache.org/jira/secure/attachment/12394165/solr-logo.png
> https://issues.apache.org/jira/secure/attachment/12394376/solr_sp.png
> https://issues.apache.org/jira/secure/attachment/12394267/apache_solr_c_blue.jpg
> https://issues.apache.org/jira/secure/attachment/12394475/solr2_maho-vote.png
> https://issues.apache.org/jira/secure/attachment/12394350/solr.s4.jpg
> https://issues.apache.org/jira/secure/attachment/12394218/solr-solid.png
>
> (given the distribution of scores, 10 just seems like a natural cutoff)
>
>
> -Hoss
>
>

Re: logo contest

Posted by Chris Hostetter <ho...@fucit.org>.
: right, but what should be the role of committers voting in the second round?
: Is it:
: 
: 1. Rank the entries the committers like best
:  or
: 2. Rank the entries the committers think best represent the community
: preferences.
: 
: My understanding of the purpose of the second round is to interpret the
: results of the community poll and cast a binding VOTE.  I think we should

committers should cast their votes as they feel appropriate to best serve 
the interests of the community -- it's not really different then voting on 
an implementation approach for a feature, or what logging framework to 
use, or a decisison to switch from java 1.5 to 1.6 ... we have to make a 
subjective decision based on the feedback we've observed from the 
community as a whole (with solr-logo-results.html serving as our cliff 
notes)


-Hoss


Re: logo contest

Posted by Ryan McKinley <ry...@gmail.com>.
On Dec 4, 2008, at 1:16 PM, Chris Hostetter wrote:

>
> : 1.  Have solr committers vote to accept:
> :  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
>
> The process as outlined on the wiki was that the commiters should  
> have a
> ranked prefrence vote, after considering the point totals from the  
> first
> vote. (with the added caveat that a -1 veto needs to be allowed  
> since it's
> a vote to commit a change to the project)
>
> Considering the community prefrences expressed, I suggest that the
> committers hold a vote of the "high scoring" entries.  Picking a score
> of 10 as the cut off, that would give us 10 entries to vote on....

right, but what should be the role of committers voting in the second  
round? Is it:

1. Rank the entries the committers like best
  or
2. Rank the entries the committers think best represent the community  
preferences.

My understanding of the purpose of the second round is to interpret  
the results of the community poll and cast a binding VOTE.  I think we  
should either have committers vote on the community intent is or re- 
run the poll with the full community, since deciphering the #2 choice  
is unclear.

As yonik said "fun fun fun"

ryan



Re: logo contest

Posted by Chris Hostetter <ho...@fucit.org>.
: 1.  Have solr committers vote to accept:
:  https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png

The process as outlined on the wiki was that the commiters should have a 
ranked prefrence vote, after considering the point totals from the first 
vote. (with the added caveat that a -1 veto needs to be allowed since it's 
a vote to commit a change to the project)

Considering the community prefrences expressed, I suggest that the 
committers hold a vote of the "high scoring" entries.  Picking a score 
of 10 as the cut off, that would give us 10 entries to vote on....

https://issues.apache.org/jira/secure/attachment/12394070/sslogo-solr-finder2.0.png
https://issues.apache.org/jira/secure/attachment/12394266/apache_solr_b_red.jpg
https://issues.apache.org/jira/secure/attachment/12394264/apache_solr_a_red.jpg
https://issues.apache.org/jira/secure/attachment/12394268/apache_solr_c_red.jpg
https://issues.apache.org/jira/secure/attachment/12394165/solr-logo.png
https://issues.apache.org/jira/secure/attachment/12394376/solr_sp.png
https://issues.apache.org/jira/secure/attachment/12394267/apache_solr_c_blue.jpg
https://issues.apache.org/jira/secure/attachment/12394475/solr2_maho-vote.png
https://issues.apache.org/jira/secure/attachment/12394350/solr.s4.jpg
https://issues.apache.org/jira/secure/attachment/12394218/solr-solid.png

(given the distribution of scores, 10 just seems like a natural cutoff)


-Hoss