Blind voting

[Steve Wyles]

  The recent server release has totally removed the blind voting element.

  This means that voters can now be easily influenced by previous votes
rather than voting purely based on their own experience and merit of the edit
being made.

  Although, it was previouly possible to find out the vote by using the
show edit page, current votes were always hidden in the voting iframe and
within the search edit pages until the editor had actually voted.

  Allegedly there was some discussion on IRC, but it should have been
discussed wider in the community before making this change, hence this email.

http://bugs.musicbrainz.org/ticket/1825 is the related bug report.

Steve (inhouseuk)

_______________________________________________
Musicbrainz-experts mailing list
Musicbrainz-experts@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-experts

[DonRedman]

On Sun, 16 Jul 2006 16:00:43 +0200, Steve Wyles wrote:

>   The recent server release has totally removed the blind voting element.
>
>   This means that voters can now be easily influenced by previous votes  
> rather than voting purely based on their own experience and merit of the  
> edit being made.
>
>   Although, it was previouly possible to find out the vote by using the  
> show edit page, current votes were always hidden in the voting iframe  
> and within the search edit pages until the editor had actually voted.
>
>   Allegedly there was some discussion on IRC, but it should have been  
> discussed wider in the community before making this change, hence this  
> email.

I do not think IRC is the right place to make such a decision. The people  
who are on IRC are only a subset of the MB community and there are both  
old-timers and active newbies whose feedback you will never get if you  
have such a debate on IRC only.

I remember that there was a very long debate about how to handle display  
of cast votes a long time ago. IIRC it was when we were introducing the  
vote count on the edit details page. The decision was that it was ok on  
the details page but should not show on the voting overview nor in the  
iframe.
This way people can have an _intential_ look at the previous votes, but  
this is not given as an initial clue. I do not see what has changed in MB  
that this would not make sense any more.

In general I would like a policy about not making community decisions on  
IRC (unless this is announced on a mailinglist or something like that).

Note that this is not intended as criticism to what happened here. Keschte  
has changed many subtle details on the server. I can understand very well  
if he just asked for a small feedback on IRC and then went on with his  
work.

   DonRedman

--
Words that are written in CamelCase refer to WikiDocs,
the MusicBrainz documentation system.
Go to http://musicbrainz.org/doc/<SomeTerm>
(you might need to transform the term to singular)

_______________________________________________
Musicbrainz-experts mailing list
Musicbrainz-experts@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-experts

[keschte]

> I remember that there was a very long debate about how to handle display
> of cast votes a long time ago. IIRC it was when we were introducing the
> vote count on the edit details page. The decision was that it was ok on
> the details page but should not show on the voting overview nor in the
> iframe.

This was probably a time when there weren't 6000 open edits in the
system at all time. Any feature that allows to more quickly assess the
state of an edit is a good feature, i think - even if it might be a
bit populistic, it is _not_ such a big deal if an edit would be failed
because of the influence of another vote.

sometimes users only leave notes and do not vote, in other cases they
vote but do not leave messages, which is bad given the CodeOfConduct.
it's very clear than we usually miss those cases when skipping over
the list of open edits, unless you go and open each and everyone to
see if they have negative votes unless you actually voted (i do not
revisit edits i voted on, unless notes are added to them).

in my opinion, the advantages outweight the ethical aspect of showing
the vote tally on edits a user hasn't voted on.

> Note that this is not intended as criticism to what happened here. Keschte
> has changed many subtle details on the server. I can understand very well
> if he just asked for a small feedback on IRC and then went on with his
> work.

thanks.

_______________________________________________
Musicbrainz-experts mailing list
Musicbrainz-experts@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-experts

[DonRedman]

On Mon, 17 Jul 2006 09:33:56 +0200, Stefan Kestenholz wrote:

>> I remember that there was a very long debate about how to handle display
>> of cast votes a long time ago. IIRC it was when we were introducing the
>> vote count on the edit details page. The decision was that it was ok on
>> the details page but should not show on the voting overview nor in the
>> iframe.
>
> This was probably a time when there weren't 6000 open edits in the
> system at all time. Any feature that allows to more quickly assess the
> state of an edit is a good feature, i think - even if it might be a
> bit populistic, it is _not_ such a big deal if an edit would be failed
> because of the influence of another vote.

Well I strongly disagree. What I say is that there was a community  
decision after a long debate. And there were a lot of open edits then  
(even if not 6000).

What I say is that neither your personal design decision nor a discussion  
on IRC should override such a community decision. Another debate with new  
consensus can.

In fact these details can have a huge impact on the entire dynamics of  
MusicBrainz (this is the sociologue speaking again). The way people vote,  
the way open edits stay in the system form a rather delicate cybernetic  
system. And the question which information is available where plays an  
important role in the resulting dynamics.

> in my opinion, the advantages outweight the ethical aspect of showing
> the vote tally on edits a user hasn't voted on.

Um sorry, but ethical aspects are never outweighted by practical  
advantages (e.g. you do not outweight the ethical aspect of not giving our  
user's email addresses away with the practical advantage of $100.000).

Again. As of now I consider this a bug that should be fixed. If you want  
to change this, you are free to start a discussion and ask for oppions  
(currently you have inhouseuk's and mine).

   DonRedman

--
Words that are written in CamelCase refer to WikiDocs,
the MusicBrainz documentation system.
Go to http://musicbrainz.org/doc/<SomeTerm>
(you might need to transform the term to singular)

_______________________________________________
Musicbrainz-experts mailing list
Musicbrainz-experts@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-experts

[keschte]

> Again. As of now I consider this a bug that should be fixed. If you want
> to change this, you are free to start a discussion and ask for oppions
> (currently you have inhouseuk's and mine).

Ok, I don't feel as strongly about, and I'm sorry, I did not know about
this, and found this a rather annoying thing, having to open the detail view
for cases where I was not sure if I should follow the example of other users
or just press on "abs". This is what a majority of users do, I assume.



_______________________________________________
Musicbrainz-experts mailing list
Musicbrainz-experts@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-experts

[Steve Wyles]

On Tue, 18 Jul 2006, Stefan Kestenholz wrote:

>> Again. As of now I consider this a bug that should be fixed. If you want
>> to change this, you are free to start a discussion and ask for oppions
>> (currently you have inhouseuk's and mine).
>
> Ok, I don't feel as strongly about, and I'm sorry, I did not know about
> this, and found this a rather annoying thing, having to open the detail view
> for cases where I was not sure if I should follow the example of other users
> or just press on "abs". This is what a majority of users do, I assume.

The answer here, you should never allow yourself to be influenced by the
way that other people vote.

You should vote purely on your experience and the merit of the edit in
question.

Dealing with in any other way has the same impact as rigging a
presidential election!

Steve

_______________________________________________
Musicbrainz-experts mailing list
Musicbrainz-experts@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-experts

[keschte]

> You should vote purely on your experience and the merit of the edit in
> question.
>
> Dealing with in any other way has the same impact as rigging a
> presidential election!

If the debate was really that long when this was decided (must have been
before I've gotten serious with MB) then certainly other people must have
felt otherwise.

To be influenced by other peoples opinion is a very natural thing, you can't
know about all the stuff that has been put up to a vote. But the system
demands that people vote on edits, and lots of edits are applied after 1 or
2 weeks without much proof if they were valid or not. So, on obscure
artists, if there was one or two yes votes, I'd be glad to give it a final
push, such that it gets applied rather than staying open and collecting
numerous abstain votes, but I'd never open the detail page to see if there
actually are votes or not.




_______________________________________________
Musicbrainz-experts mailing list
Musicbrainz-experts@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-experts

[keschte]

> To be influenced by other peoples opinion is a very natural thing, you can't
> know about all the stuff that has been put up to a vote. But the system
> demands that people vote on edits, and lots of edits are applied after 1 or
> 2 weeks without much proof if they were valid or not. So, on obscure
> artists, if there was one or two yes votes, I'd be glad to give it a final
> push, such that it gets applied rather than staying open and collecting
> numerous abstain votes, but I'd never open the detail page to see if there
> actually are votes or not.

can i expect a response on this, or is the subject closed?

_______________________________________________
Musicbrainz-experts mailing list
Musicbrainz-experts@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-experts

[Steve Wyles]

>> To be influenced by other peoples opinion is a very natural thing, you
>> can't
>> know about all the stuff that has been put up to a vote. But the system
>> demands that people vote on edits, and lots of edits are applied after 1 or
>> 2 weeks without much proof if they were valid or not. So, on obscure
>> artists, if there was one or two yes votes, I'd be glad to give it a final
>> push, such that it gets applied rather than staying open and collecting
>> numerous abstain votes, but I'd never open the detail page to see if there
>> actually are votes or not.
>
> can i expect a response on this, or is the subject closed?

I totally agree with Don's remarks in:

http://lists.musicbrainz.org/pipermail/musicbrainz-experts/2006-July/000494.html

It should be set back to the old behaviour at the moment.

Any decision to remove blind voting needs to be discussed community wide
and the impact fully assessed.

I believe the ticket is still open http://bugs.musicbrainz.org/ticket/1825

Steve

_______________________________________________
Musicbrainz-experts mailing list
Musicbrainz-experts@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-experts