Step and half brother data. Do we really want this?

classic Classic list List threaded Threaded
11 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Step and half brother data. Do we really want this?

Nicolás Tamargo de Eguren
Hi! 

Ages ago (before my style time) an RFC passed to implement "step" and "half" attributes for the sibling relationship, and "step" for the parent/child one. This was never implemented, and there's a ticket for it still (well, 5...) at http://tickets.musicbrainz.org/browse/STYLE-9

I personally feel this is overkill and have no interest in it, but do other people feel this is useful?

--
Nicolás Tamargo de Eguren

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

Rachel Dwight

On Feb 2, 2015, at 6:50 AM, Nicolás Tamargo de Eguren <[hidden email]> wrote:

Hi! 

Ages ago (before my style time) an RFC passed to implement "step" and "half" attributes for the sibling relationship, and "step" for the parent/child one. This was never implemented, and there's a ticket for it still (well, 5...) at http://tickets.musicbrainz.org/browse/STYLE-9

I personally feel this is overkill and have no interest in it, but do other people feel this is useful?

I do; I even had plans to resurrect that proposal at one point.


--
Nicolás Tamargo de Eguren
_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

Nicolás Tamargo de Eguren
On Mon, Feb 2, 2015 at 2:52 PM, Rachel Dwight <[hidden email]> wrote:

On Feb 2, 2015, at 6:50 AM, Nicolás Tamargo de Eguren <[hidden email]> wrote:

Hi! 

Ages ago (before my style time) an RFC passed to implement "step" and "half" attributes for the sibling relationship, and "step" for the parent/child one. This was never implemented, and there's a ticket for it still (well, 5...) at http://tickets.musicbrainz.org/browse/STYLE-9

I personally feel this is overkill and have no interest in it, but do other people feel this is useful?

I do; I even had plans to resurrect that proposal at one point.

Hmm, ok. It's easy to add, FWIW, so if one more person shows any interest I guess I'll put it in.

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

tommycrock
Although I've not come across a need for it while editing, as a step-brother  with step-parents they're (obviously) a different kind of relationship but an important one. So I wouldn't use brother to mean step-brother but think we should be able to represent it. I think it would be odd to limit the relationships to biological ones.

On 2 February 2015 at 13:06, Nicolás Tamargo de Eguren <[hidden email]> wrote:
On Mon, Feb 2, 2015 at 2:52 PM, Rachel Dwight <[hidden email]> wrote:

On Feb 2, 2015, at 6:50 AM, Nicolás Tamargo de Eguren <[hidden email]> wrote:

Hi! 

Ages ago (before my style time) an RFC passed to implement "step" and "half" attributes for the sibling relationship, and "step" for the parent/child one. This was never implemented, and there's a ticket for it still (well, 5...) at http://tickets.musicbrainz.org/browse/STYLE-9

I personally feel this is overkill and have no interest in it, but do other people feel this is useful?

I do; I even had plans to resurrect that proposal at one point.

Hmm, ok. It's easy to add, FWIW, so if one more person shows any interest I guess I'll put it in.

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

swisschris
I think with Nicolás that this is overkill. I vividly remember the editor who tried to add relationships for even the remotest theoretically imaginable stuff – not only for family relationships, but for the vocal tree (http://wiki.musicbrainz.org/Proposal:Advanced_Vocal_Tree) and the like. IMO all this will lead to cluttering the UIs with stuff barely ever needed making editing even more difficult for new editors. If we start this, we'll have Blood Brothers, Brothers in Arms and of course all kind of adoptions (I'm not making this up, this was seriously discussed at the time). Why can't we just use annotations for the rare cases where such a thing occurs (and is relevant for the DB). Or let's do it like for the instrument tree, where we use annotations until at least 5 actually existing (and relevant) credits can be shown ;-)

On Mon, Feb 2, 2015 at 3:44 PM, Tom Crocker <[hidden email]> wrote:
Although I've not come across a need for it while editing, as a step-brother  with step-parents they're (obviously) a different kind of relationship but an important one. So I wouldn't use brother to mean step-brother but think we should be able to represent it. I think it would be odd to limit the relationships to biological ones.

On 2 February 2015 at 13:06, Nicolás Tamargo de Eguren <[hidden email]> wrote:
On Mon, Feb 2, 2015 at 2:52 PM, Rachel Dwight <[hidden email]> wrote:

On Feb 2, 2015, at 6:50 AM, Nicolás Tamargo de Eguren <[hidden email]> wrote:

Hi! 

Ages ago (before my style time) an RFC passed to implement "step" and "half" attributes for the sibling relationship, and "step" for the parent/child one. This was never implemented, and there's a ticket for it still (well, 5...) at http://tickets.musicbrainz.org/browse/STYLE-9

I personally feel this is overkill and have no interest in it, but do other people feel this is useful?

I do; I even had plans to resurrect that proposal at one point.

Hmm, ok. It's easy to add, FWIW, so if one more person shows any interest I guess I'll put it in.

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

Robert Bihlmeyer
In reply to this post by Nicolás Tamargo de Eguren
Hi,

Am 2015-02-02 um 13:50 schrieb Nicolás Tamargo de Eguren:
> Ages ago (before my style time) an RFC passed to implement "step" and
> "half" attributes for the sibling relationship, and "step" for the
> parent/child one.
a question I had on my mind for some time, that could have some impact
on this: what is the policy on adding people that have no reasonable
place in our database (i.e. never performed in any role related to
music), but are useful as "relationship proxies"?

The case I came across was that a man and his grandfather are both
musicians in MB, but the man's father is not. The grandfather
relationship cannot (at the moment) be entered directly, but could be
constructed by adding the father (not otherwise useful to us) and
relating the other two to him via parent-child.

This would also make half-brother relationships superflous: you could
just link both brothers to their shared parent. If X is the step-parent
of Y, you could replace that with a marriage and a (normal) parent link
via a new person Z.

Of course, some intelligence in software would be needed to show the
transitive relationships. But that would be nice, anyway, so that "don't
make relationship clusters" makes sense.

--
Robbe

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

tommycrock


On 10 Feb 2015 21:42, "Robert Bihlmeyer" <[hidden email]> wrote:
>
> Hi,
>
> Am 2015-02-02 um 13:50 schrieb Nicolás Tamargo de Eguren:
> > Ages ago (before my style time) an RFC passed to implement "step" and
> > "half" attributes for the sibling relationship, and "step" for the
> > parent/child one.
> a question I had on my mind for some time, that could have some impact
> on this: what is the policy on adding people that have no reasonable
> place in our database (i.e. never performed in any role related to
> music), but are useful as "relationship proxies"?
>
> The case I came across was that a man and his grandfather are both
> musicians in MB, but the man's father is not. The grandfather
> relationship cannot (at the moment) be entered directly, but could be
> constructed by adding the father (not otherwise useful to us) and
> relating the other two to him via parent-child.

Makes sense to me

>
> This would also make half-brother relationships superflous: you could
> just link both brothers to their shared parent. If X is the step-parent
> of Y, you could replace that with a marriage and a (normal) parent link
> via a new person Z.

Not just half brother, all sibling relationships, assuming you have enough information about who the parents are ...

>
> Of course, some intelligence in software would be needed to show the
> transitive relationships. But that would be nice, anyway, so that "don't
> make relationship clusters" makes sense.
>
> --
> Robbe
>
> _______________________________________________
> MusicBrainz-style mailing list
> [hidden email]
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

Ulrich Klauer
In reply to this post by Robert Bihlmeyer
Robert Bihlmeyer wrote:

> a question I had on my mind for some time, that could have some impact
> on this: what is the policy on adding people that have no reasonable
> place in our database (i.e. never performed in any role related to
> music), but are useful as "relationship proxies"?

I don't think there is an official style guideline about this (though  
I may be mistaken). I'd be fine with it, though. - There was some  
related discussion in the companion thread on the forums:  
http://forums.musicbrainz.org/viewtopic.php?id=5510

Ulrich (chirlu)


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

Robert Bihlmeyer
In reply to this post by tommycrock
Am 2015-02-10 um 23:37 schrieb Tom Crocker:
> On 10 Feb 2015 21:42, "Robert Bihlmeyer" <[hidden email]> wrote:
>
>> This would also make half-brother relationships superflous: you could
>> just link both brothers to their shared parent. [...]
> Not just half brother, all sibling relationships, assuming you have enough
> information about who the parents are ...
>
Hmm, you could go full-monty and not only open the gates for
non-musician-parents, but nameless ones as well.
Gary McGuy has parents Ebeneezer McGuy and Anonymous Female #38475
Andy McGuy has parents Ebeneezer McGuy and Anonymous Female #837212
Therefore they are step-brothers.

This would potentially triple the number of persons in the database.

--
Robbe

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

Robert Bihlmeyer
In reply to this post by Robert Bihlmeyer
Am 2015-02-10 um 22:42 schrieb Robert Bihlmeyer:
> Of course, some intelligence in software would be needed to show the
> transitive relationships. But that would be nice, anyway, so that "don't
> make relationship clusters" makes sense.
>
Regarding grandparent relationships, we have a fair number (more than
100, IIRC) of them already in the database. Of those that I
spot-checked, all three (child, parent, grandparent) were MB-relevant.
The ones I still remember are Leopold, Wolfgang Amadeus, and Franz Xaver
Wolfgang Mozart, as well as Benito Mussolini with his son and granddaughter.

Better software support that showed me that Leopold is the grandfather
of Franz Xaver without me resorting to SQL would be desirable.

(There are also two families, where four generations are in our database...)

--
Robbe


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Step and half brother data. Do we really want this?

Ulrich Klauer
In reply to this post by Robert Bihlmeyer
Robert Bihlmeyer wrote:

> Gary McGuy has parents Ebeneezer McGuy and Anonymous Female #38475
> Andy McGuy has parents Ebeneezer McGuy and Anonymous Female #837212
> Therefore they are step-brothers.

Half, not step.

> This would potentially triple the number of persons in the database.

As your own example shows, it would at most inflate the number by a  
factor of 2.5, not 3 (three parents for two half-siblings). Assuming  
every artist actually has a half-sibling artist, and exactly one,  
because otherwise the parents/children ratio gets lower; and further  
assuming none of the parents is an MB artist already. So, slightly  
unrealistic.

Ulrich


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Loading...