[mb-style] RFC: Change BoxSetNameStyle

[Chris B]

On 26/03/2008, Brian Schweitzer <brian.brianschweitzer at gmail.com> wrote:
> On Wed, Mar 26, 2008 at 5:51 AM, Chris B <chris at whenironsattack.com> wrote:
>  > On 26/03/2008, Brian Schweitzer <brian.brianschweitzer at gmail.com> wrote:
>  >  > The only one I can see my agreeing to merge away as is directed by
>  >  these guidelines is #1.  But the wording of the guidelines - esp the
>  >  BSNS version - depending on your interpretation of them, can and is
>  >  being used to get rid of the other types too.
>  >
>  >  so again, reword, not delete...
>  >
>  >
>  >  >  This "but clutter!" argument?  Sorry - if we did things just to get
>  >  >  rid of "clutter", we'd have no bootlegs
>  >
>  >  bootlegs aren't clutter - they have their own section if you want to ignore them
>  >
>  >
>  >  > , we'd not have an ArtistIntent
>  >  >  concept (standardizing just to standardize)
>  >
>  >  ArtistIntent usually just dictates *how* you standardize
>  >
>  >
>  >  > , we'd have no issue
>  >  >  merging Wendy into Walter
>  >
>  >  i don't :) (though i'd merge walter into wendy, rather)
>  >
>  >
>  >  > we'd never bother to want something like
>  >  >  ANV
>  >
>  >  ANV (at discogs) causes no clutter - that's the beauty of it.
>  >
>  >
>  >  , we'd not have transl(iter)ations (they're identical too, even if
>  >  >  the titles are the same), etc.  Life is messy.
>  >
>  >  transliterations are what i feel to be 'necessary clutter'.
>  >
>  >  brian, you seem to be talking about the lack of effect these sets
>  >  would have on 'cluttered' (especially classical) artists. no denying
>  >  that, but then there's not much we can ever do to 'declutter' such
>  >  artists. what i'm saying is that for artist with a limited amount of
>  >  specific studio albums (ie most contemporary artists), our current
>  >  view isn't far off the eventual primary ReleaseGroups view. the only
>  >  'clutter' we have are variants of tracklists, which everyone seems to
>  >  be happy with (until ReleaseGroups get here anyway). there is a value
>  >  in the lookout we have, and IMO allowing BoxSets wholesale is a bad
>  >  precedent that will ultimately end that till the elusive NGS arrives.
>  >
>  >  all of these arguments seem irrelevant as you seem to agree with
>  >  'bundle' type boxsets being clutter, but there you go...
>
>
> Whose current view though?  Frankly, there's plenty of non-classical
>  such box sets in the database, and not all there just because they
>  slipped in; many are there because knowledgable editors added them.
>  Honestly, until this RFP, how many of us even realized there was
>  wording in the guidelines that specifically said to merge away such
>  box sets?
>
>  But yes, I am in favor of merging away only the class Oliver described
>  as "bundles".  A box set has a unique identity, packaging,
>  discographic relevance, etc.  A "bundle" only has an identity at all
>  because someone wrapped a ribbon or plastic wrap around two or three
>  CDs and sold it that way.  To use the Amazon example, a box set has
>  its own ASIN, whereas a bundle, at Amazon, could be any of their
>  suggested "Buy this other CD with this one and save $3!" offers.

actually, that example is totally inappropriate. amazons 'buy this
with this' offers aren't included in any sort of packaging or 'box' at
all. in fact, depending on your luck they might even be sent
separately. all bundles/boxsets will be done at the label's end (even
discogs wouldn't care about 'boxsets' done at the point of sale)

a 'bundle' boxset is something like
http://www.discogs.com/release/1216703 - all preexisting releases, no
special title or theme behind the box, it's just everything they've
previously released compiled into one box. there's a 'name'
('Radiohead') and a cat# but nothing that helps a non-collectors
discography.

>  But the same isn't true when it comes to a box set.  Let's say we have
>  some theoretical box set, with 10 CDs.  The box set has a barcode and
>  catalog number.  Perhaps the CDs have them too, perhaps they don't.
>  But now you go to add release events.  If you add the box set
>  information as an event, you identify that release as that box.  We
>  don't have some overarching box set entity class, where we could link
>  the discs.  We don't have some way to indicate that the box set info
>  there represents only one part of that box.  For all apparent
>  purposes, that event, just like the release title, references the
>  entirety of the release.  And for an unboxed release, this is true.
>  But in the box set, it's not - UPC 1234567890123 with cat # ABC 123,
>  and box release title "The Works of John Smith" uniquely identifies a
>  series of discs.
>
>  What I don't get, though, is why some people here seem to think this
>  is some brand new concept.  If it were, it'd be difficult to find
>  examples.  But take any 10 random artists with more than 1 release in
>  their listing, and you'll likely find at least one with a box set just
>  as we're describing.  ABBA has one.  So do The Doors, The Beatles, Led
>  Zeppelin, Pink Floyd, Tori Amos, Oasis, Steve Reich, Mozart, Bach,
>  Beethoven, Miles Davis, Bob Marley... the list goes on and on, and
>  encompasses basically every single artist with a discography over a
>  certain size.  For those who think these are clutter, ignore them!
>  Noone forces you to add data to or tag to box set listings.  But the
>  view that these box sets should simply be swept away *does* force
>  those who actually find value in complete discographies, including the
>  box sets, to lose that data.  When we have some sort of release
>  grouping, then the "clutter" issue goes away anyhow - not just for
>  boxes, but for all the other variations: the release with 9 versions,
>  the altered tracking between LP/CD/etc, all of them.  So why rush,
>  then, to eliminate legitimate boxes now, only to have to bring that
>  information back at some later point?
>
>  I can only see a few arguments for the "no boxes" concept, yet I can't
>  buy any of them:
>
>  * Less "clutter" - see above
>
>  * Less ARs to do - track merging is coming quite soon, and having the
>  boxes *increases* the chances of people adding ARs to begin with.
>
>  * Larger discographies - So what?  Does it hurt things so badly to
>  have 50 releases in a discography, rather than 40?
>
>  * Redundant discographies - A release isn't redundant if it's not the
>  same.  Just as a translated, transliterated, remastered, just
>  slower/faster enough to be outside +/-5s, etc release is almost the
>  same, but just different enough, the combination of *all*
>  non-performance/composition ARs being different, all event info being
>  different, the title being different, and the release being part of a
>  greater whole, rather than a unique entity unto itself, would all
>  together seem to identify that a box set version of a release is
>  definitely different enough from the unboxed version that it too
>  deserves listing.
>
>  I can't think of any other arguments for why this concept would
>  possibly make sense to anyone.  And none of these would answer the
>  example I provided way back at the start of this, as to why a box set
>  like the Beethoven or Bach or Mozart ones, where we're dealing with a
>  disc that isn't even identified as having ever been released outside
>  of the box...  Investigation into earlier possible releases is fine,
>  if we're looking to set earlier/later release of ARs, or to fill in
>  performer ARs (to either the earlier from the box, or to the box from
>  the earlier), or, when we have track merging, if we're looking to
>  merge tracks.  But noone has yet here provided any one reason why
>  there should be justification that such boxes ought to be disassembled
>  simply because someone actually bothers to - entirely apart from
>  information conveyed by the box - identify each earlier release.  And,
>  also, nowhere in this entire discussion have I yet seen anyone provide
>  a reason why, if we then decide that such "non-earlier-identified box
>  discs" ought to stay, why then we could consider it okay that box disc
>  releases, perfectly legitimate as they already are, ought to be
>  "unboxed", just because someone decides they like the release better
>  if it is unboxed.

we're going over the same points again and again. at this point you
have to reword your RFC to specifically address non-bundles, or accept
that we don't all agree on this issue.