HLL's Hot 25 for 5/23/2012

[Michael Pendragon]

Senor - Eartha Kitt
This Guy's in Love With You - Herb Alpert & The Tijuana Brass
Linda - Jim Reeves
Slow Rider - Johnny Cash
Song for You - Michael Buble w/Chris Botti
Tenderly - Johnny Mathis
These Foolish Things (Remind Me of You) - Nat "King" Cole
One-Sided Love Affair - Elvis Presley
Don't Ask Me Why - Eileen Barton
Movin' Out - Billy Joel
Stagger Lee - Dion
If I Could Be With You (One Hour Tonight) - Kay Starr
I Don't Feel Like Singing Anymore - Frankie Laine
I Grabbed for the Engine - The Fontanes
Let's Cry Together - Patti Page
Down Among the Sheltering Palms - Kay Starr
Just for Old Time's Sake - The McGuire Sisters
Anema E Core - Eddie Fisher
Saturday Night (Is the Loneliest Night of the Week) - The King Sisters
So Long (It's Been Good to Know Yuh) - Pete Seeger
There! I've Said It Again - Vaughn Monroe
Something There - Paige Ohara, Robby Benson, & assorted cast members
of Beauty & The Beast
Never Will I Marry - Nancy Wilson & Cannonball Adderly
The Banana Boat Song - Harry Belafonte
The Moment of Truth - Frankie Laine

[RWC]

If you care to, please keep these lists coming, Mike.

Personal experience has shown that at least 50% of your randomly selected,
mostly adult pop, playlist music selections are heads-up keepers - at the very
least for *quality* (non-boring, and better) background music. I thank you for
these playlists.

I've recently embarked on a project to 'database' popular music (including
blues, r&b, rockabilly, & pop folk), focused/concentrating on the years 1930 to
1964 - but I'm also happy to include popular music (incuding reggae, punk, new
wave etc) from 1880 to 2012 along the way :-)

This is a part-time *fun* yet also an excitingly challenging project - plus
there's significant educational benefits for me.

The two stats based messages I've recently posted to this newsgroup are merely
by-products of a greater quest.

There may? be more efficient sources of data, but for fun I've taken the
excellent Roger Ford contests, and Pendragon's wonderful HLL postings, over time
to this Usenet group, as a source of artist name, song title, label details and
year.

Perhaps I could be accused of being (only recently) a Usenet info stalker :-)
- I admit I have in the last week 'databased' all of your playlists, Mike, for
2012. Not just for the project briefly described above, but as I've also
implied above, your lists act as prompts for seeking out some, sometimes a tad
obscure perhaps but nevertheless 'keeper', adult pop music.

Here are some HLL playlist stats:

These are Mike's playlist artists who have contributed 2+ unique recordings
since January. (for instance, Mike's playlists since Jan 2012 have included 60
*different* recordings by Frankie Laine)

BTW The stats below, assuming a truly random, non biased, selection of tracks,
also indicate the relative number of recordings by artist on Mike's Ipod (or
whatever gizmo).
(For instance, it seems Mike, partly renowned for his admirable cd box
collections, has twice as many Sarah Vaughan tracks on his portable music player
as compared to Dinah Shore tracks)

Frankie Laine 60
Elvis Presley 57
Frank Sinatra 42
Connie Francis 26
Teresa Brewer 23
Kay Starr 22
Bing Crosby 21
Pat Boone 19
Peggy Lee 19
Nat 'King' Cole 18
Sarah Vaughan 18
Johnnie Ray 17
Doris Day 15
Tony Bennett 15
Patti Page 14
Joni James 13
Nancy Wilson 11
The Andrews Sisters 11
The Ink Spots 11
Brian Setzer 10
Dinah Washington 10
Ella Fitzgerald 10
Guy Mitchell 10
Julie London 10
Vaughn Monroe 10
Dinah Shore 9
Billie Holiday 8
Crystal Gayle 8
Dean Martin 8
Georgia Gibbs 8
Lena Horne 8
Margaret Whiting 8
Bobby Darin 7
Eddie Fisher 7
Jane Morgan 7
Louis Armstrong 7
Mel Torme 7
Patsy Cline 7
Perry Como 7
The Ames Brothers 7
The McGuire Sisters 7
Eartha Kitt 6
Eddy Arnold 6
Jim Reeves 6
Jo Stafford 6
Kitty Kallen 6
Leon Redbone 6
Vic Damone 6
Vikki Carr 6
Al Jolson 5
Bessie Smith 5
Brian Setzer Orchestra 5
Connie Haines 5
Diana Krall 5
Dion 5
Fats Waller 5
Harry Connick Jr. 5
Janie Grant 5
Louis Jordan & His Tympani 5 5
Nelson Eddy 5
Peter, Paul & Mary 5
Russ Morgan Orchestra 5
The Crew-Cuts 5
The Monkees 5
Tom Lehrer 5
Tony Martin 5
Anita O'Day 4
Billy Joel 4
Bing Crosby & The Andrews Sisters 4
Cab Calloway 4
Chuck Berry 4
Edith Piaf 4
Fred Astaire 4
Gogi Grant 4
June Christy 4
Les Pau l 4
Louis Jordan 4
Mario Lanza 4
Michael Buble 4
Michael Feinstein 4
Phil Harris 4
Ruby Murray 4
Shirley Bassey 4
Spike Jones & His City Slickers 4
The Brothers Four 4
The Carpenters 4
The Doors 4
The Everly Brothers 4
The Platters 4
The Stray Cats 4
Bill Kenny 3
Brenda Lee 3
Cleo Laine 3
Don McLean 3
Dorothy Lamour 3
Eileen Barton 3
Ernest Tubb 3
Fats Waller & His Rhythm 3
Frankie Laine & The Four Lads 3
Gene Autry 3
Gene Pitney 3
George Thorogood 3
Glenn Miller Orchestra, Marion Hutton v. 3
Hank Williams 3
Jane Monheit 3
Jerry Lee Lewis 3
Johnny Desmond 3
Johnny Mathis 3
Judy Garland 3
Kate Smith 3
Keely Smith 3
Les Paul & Mary Ford 3
Louis Prima 3
Marlene Dietrich 3
Patty Andrews 3
Peggy King 3
Pete Seeger 3
Robert Johnson 3
Ruth Wallis 3
Sam Cooke 3
Sinead O'Connor 3
Sonny James 3
Tennessee Ernie Ford 3
The Four Lads 3
The McGuires 3
Vera Lynn 3
Anita Bryant 2
Annette Funicello 2
Bill Haley & His Comets 2
Billy Butterfield Orchestra 2
Billy Eckstine 2
Billy Fury 2
Bing Crosby & The Paul Whiteman Orchestra 2
Buddy Clark 2
Burl Ives 2
Carl Perkins 2
Dave Phillips 2
Doris Day w/Les Brown & His Band of Renown 2
Ed Ames 2
Ella Fitzgerald & Louis Armstrong 2
Ella Fitzgerald w/Chick Webb's Orchestra 2
Fabian 2
Fats Domino 2
Hadda Brooks 2
Howard Keel 2
Jaye P. Morgan 2
Jerry Vale 2
Jo Stafford & Frankie Laine 2
Johnnie Ray & The Four Lads 2
Johnny Cash 2
June Valli 2
Kathryn Grayson & Howard Keel 2
Kathy Kirby 2
Lesley Gore 2
Leslie Uggams, Mitch Miller & The Gang 2
Linda Scott 2
Madonna 2
Marilyn Monroe 2
Marty Robbins 2
Merle Travis 2
Mindy Carson 2
Mitch Miller & The Gang 2
Monica Lewis 2
Neil Sedaka 2
Pat Suzuki 2
Percy Faith Orchestra & Chorus 2
Ricky Nelson 2
Rosemary Clooney 2
Roy Orbison 2
Sammy Davis Jr. 2
Somethin? Smith & The Redheads 2
Sonny Lester Orchestra 2
The Big Bopper 2
The Chiffons 2
The Coasters 2
The Fleetwoods 2
The Four Aces 2
The Four Seasons 2
The Hilltoppers 2
The King Cole Trio 2
The Kingston Trio 2
The Lovin' Spoonful 2
The Shangri-Las 2
The Sons of the Pioneers 2
The Weavers 2
Timi Yuro & Johnnie Ray 2
Tiny Tim 2
Toni Arden 2
Wanda Jackson 2

[RWC]

Rick Schubert would probably, to his credit (true thanks, Rick), suggest I
combine:

Brian Setzer 10
with
Brian Setzer Orchestra 5
==
15

I like to think there aren't any other *material* cases like this.


Geoff

[Michael Pendragon]

Wow! Thanks Geoff.

It's surprising that the breakdown appears to be sticking pretty
closely to the number of tracks per artist I've actually got. I'll
probably do a side by side comparison of the two tonight.

[F R]

Pendragon>

Wow! Thanks Geoff.
It's surprising that the breakdown appears to be sticking pretty closely
to the number of tracks per artist I've actually got. I'll probably do a
side by side comparison of the two tonight.

-----------------------
It shouldn't be surprising in the least Mike. Assuming, as Geoff has,
that your daily dosage of music is truly random, doesn't it stand to
reason that the more tracks you have by an individual artist SHOULD
reflect exactly that when they are randomly selected?

[Michael Pendragon]

On May 24, 1:01 pm, espon...@webtv.net (F R) wrote:
> Pendragon>
> Wow! Thanks Geoff.
> It's surprising that the breakdown appears to be sticking pretty closely
> to the number of tracks per artist I've actually got. I'll probably do a
> side by side comparison of the two tonight.
> -----------------------
> It shouldn't be surprising in the least  Mike. Assuming, as Geoff has,
> that your daily dosage of music is truly random,

Geoff assumes correctly. It's strictly ipod shuffle.

> doesn't it stand to
> reason that the more tracks you have by an individual artist SHOULD
> reflect exactly that when they are randomly selected?

In a mathematically perfect world. But my ipod often seems to be
quite capricious in its selections.

[Bob Roman]

On May 24, 4:07 pm, Michael Pendragon

<michaelmaleficapendra...@gmail.com> wrote:
> On May 24, 1:01 pm, espon...@webtv.net (F R) wrote:
>
> > Pendragon

> > > It's surprising that the breakdown appears to be sticking pretty closely
> > > to the number of tracks per artist I've actually got.

> > -----------------------
> > It shouldn't be surprising in the least  Mike. Assuming, as Geoff has,

> > that your daily dosage of music is truly random, doesn't it stand to

> > reason that the more tracks you have by an individual artist SHOULD
> > reflect exactly that when they are randomly selected?
>
> In a mathematically perfect world.  But my ipod often seems to be
> quite capricious in its selections.

Can you tell us why you find "capricious" representativeness to be
more "surprising" than statistical representativeness?

--
BR

[Michael Pendragon]

No ... because I said I find statistical representation to be more
surprising.

[Bob Roman]

On May 24, 10:50 pm, Michael Pendragon

Statistical representation is inevitable in the long run. I don't
think you understand Frank's point.

--
BR

[Michael Pendragon]

Here's a match-up of the first 100 artists:

8 of the top 10 match, and of those 8 almost all are in the same order
(only Connie and Frank need to trade places).

ipod plays: # of songs
in library:

Frankie Laine 60 Frankie
Laine 1037
Elvis Presley 57 Elvis Presley
674

Frank Sinatra 42 Connie Francis

539
Connie Francis 26 Frank Sinatra
470
Teresa Brewer 23 Kay Starr 398

Kay Starr 22 Bing Crosby

371
Bing Crosby 21 Johnnie Ray 363
Pat Boone 19 Doris Day 349
Peggy Lee 19 Pat Boone 327
Nat 'King' Cole 18 Nat King Cole
316

Teresa Brewer will be bumping Nat out of the top ten in a few days
when a cd set of hers I recently ordered arrives.

17 of the top 20.

Sarah Vaughan 18 Tony Bennett 262
Johnnie Ray 17 Teresa Brewer
257
Doris Day 15 Peggy Lee 254
Tony Bennett 15 Sarah Vaughan 225
Patti Page 14 Patty Andrews/
Andrews Sisters 220
Joni James 13 Ella Fitzgerald
214
Nancy Wilson 11 Brian Setzer/Stray
Cats 210
The Andrews Sisters 11 Patti Page 210
The Ink Spots 11 Georgia Gibbs
199
Brian Setzer 10 Vaughn Monroe 199



Dinah Washington 10 The Andrews Sisters
186
Ella Fitzgerald 10 Billie
Holiday 166
Guy Mitchell 10 Jim
Reeves 153
Julie London 10 Joni
James 153
Vaughn Monroe 10 Judy Garland
151
Dinah Shore 9 Crystal
Gayle 149
Billie Holiday 8 Dion/
Belmonts 147
Crystal Gayle 8 Julie
London 141
Dean Martin 8 The Ink
Spots 136
Georgia Gibbs 8 Jane
Morgan 133

Lena Horne 8 Kitty
Kallen 132
Margaret Whiting 8 Louis Jordan & His
Tympany Five 132
Bobby Darin 7 Glenn
Miller 130
Eddie Fisher 7 Perry
Como 128
Jane Morgan 7 Dean
Martin 127
Louis Armstrong 7 Louis Armstrong 125
Mel Torme 7 Margaret Whiting
123
Patsy Cline 7 Nancy
Wilson 119
Perry Como 7 Fats
Waller 118
The Ames Brothers 7 Jo Stafford
117

The McGuire Sisters 7 Connie Haines
116
Eartha Kitt 6 Johnny
Cash 108
Eddy Arnold 6 Vic Damone
106
Jim Reeves 6 Al
Jolson 104
Jo Stafford 6 The Everly
Brothers 103
Kitty Kallen 6 The Fontane
Sisters 100
Leon Redbone 6 Helen O'Connell 95
Vic Damone 6 The McGuire Sisters
94
Vikki Carr 6 The Four
Lads 93
Al Jolson 5 Guy
Mitchell 92

Bessie Smith 5 Dinah Washington 90
Brian Setzer Orchestra 5 Patsy Cline 90
Connie Haines 5 Lena Horne 89
Diana Krall 5 Dinah
Shore 88 (less than half of Sarah V's - almost a third)
Dion 5 Pete Seeger/
Weavers 86
Fats Waller 5 Ed Ames/Ames Brothers
85
Harry Connick Jr. 5 Diana
Krall 83
Janie Grant 5 Les Paul/Mary Ford
83
Louis Jordan & His Tympani 5 5 Harry James 81
Nelson Eddy 5 Johnny Mathis
81

Peter, Paul & Mary 5 The Ames Brothers 79
Russ Morgan Orchestra 5 Eddie Fisher 77
The Crew-Cuts 5 Bessie Smith
75
The Monkees 5 Guy Lombardo 74
Tom Lehrer 5 Leon Redbone
74
Tony Martin 5 Tom Lehrer
74
Anita O'Day 4 Chuck Berry
72
Billy Joel 4 Tony
Martin 72
Bing Crosby/Andrews Sisters 4 Edith Piaf 71
Cab Calloway 4 Eartha Kitt
70

Chuck Berry 4 The Chordettes
69
Edith Piaf 4 The MGM Studio
Orchestra & Chorus 69
Fred Astaire 4 Tennessee Ernie Ford
69
Gogi Grant 4 Timi
Yuro 69
June Christy 4 Tommy Dorsey
69
Les Pau l 4 Mel
Tormé 68
Louis Jordan 4 The Platters
68
Mario Lanza 4 Jimmy Sacca/
Hilltoppers 67
Michael Buble 4 Jimmy Dorsey
65
Michael Feinstein 4 Rosemary Clooney 64

Phil Harris 4 Anita
O'Day 63
Ruby Murray 4 Mario Lanza 63
Shirley Bassey 4 The Crew-Cuts
62
Spike Jones & His City Slickers 4 Eileen Barton 61
The Brothers Four 4 Jane Monheit 61
The Carpenters 4 Jaye P. Morgan
61
The Doors 4 Michael
Bublé 61
The Everly Brothers 4 Bobby Darin 60
The Platters 4 Toni
Arden 60
The Stray Cats 4 The Hilltoppers
59

Bill Kenny 3 The Stray
Cats 59
Brenda Lee 3 Shirley
Bassey 58
Cleo Laine 3 Russ Morgan
57
Don McLean 3 June
Valli 56
Dorothy Lamour 3 Fabian
54
Eileen Barton 3 Kay
Kyser 53
Ernest Tubb 3 The Kingston Trio 53
Fats Waller & His Rhythm 3 Jerry Vale
52
Frankie Laine & The Four Lads 3 Sammy Davis, Jr. 52
Gene Autry 3 Buddy Clark
50

[Michael Pendragon]

I'm a Mark Twain fan.

[Bob Roman]

On May 24, 11:38 pm, Michael Pendragon

Who isn't? But I don't think you quite understand Twain's point
either.

--
BR

[The Bloomfield Bloviator]

Me

That's one of the reasons that Mike Full-O'Shitzky is so bad with
logic and reasoning (and math). He's too bust shooting his load over
poetry and fantasies he reads in fiction.

[Sharx3335]

"The Bloomfield Bloviator" <Sav...@aol.com> wrote in message
news:4807fac4-5a42-4746...@j25g2000yqn.googlegroups.com...

Most of that porn that YOU whack off to on the Net might as well be
fiction in as much as little a chance you have of having a
relationship with such babes/guys in REAL life!!

[The Bloomfield Bloviator]

On May 25, 3:16 am, "Sharx3335" <sharx...@hotmail.com> wrote:
> "The Bloomfield Bloviator" <Savo...@aol.com> wrote in messagenews:4807fac4-5a42-4746...@j25g2000yqn.googlegroups.com...

Fuck me......I was hoping you had joined Wilhelms.

[Sharx3335]

"The Bloomfield Bloviator" <Sav...@aol.com> wrote in message

news:d265a0c5-8a80-4447...@b21g2000yqn.googlegroups.com...

I suspect that YOU will be going that route, long before me, thanks
to your predilection for low-oxygen masturbatory practices.

[Michael Pendragon]

Either that, or you fail to understand mine.

If you're serious about pursuing the subject, I'll restate my point:

I believe that statistical representation would only be inevitable in
a mathematically perfect world (that is, in theory -- not necessarily
in reality). I believe that far too many extenuating factors intrude
(including, in this particular instance, the method by which an ipod
"arbitrarily" selects songs when in "shuffle play" mode).

[Michael Pendragon]

Damn, Grossberg! Do you have to be coming on to other guys online???

[Bob Roman]

On May 25, 9:36 am, Michael Pendragon

<michaelmaleficapendra...@gmail.com> wrote:

> I believe that statistical representation would only be inevitable in
> a mathematically perfect world (that is, in theory -- not necessarily
> in reality).

Reality does not match a mathematical model when 1) the model is
incorrect or 2) an error term is introduced. In the present case the
model is so fundamental (and demonstrated so regularly) that it cannot
be incorrect. More to the point, in the present case, the model
worked perfectly. There is no evidence of an error term. You are now
arguing not only against basic probability theory but also against
reality as it occurred.

> I believe that far too many extenuating factors intrude
> (including, in this particular instance, the method by which an ipod
> "arbitrarily" selects songs when in "shuffle play" mode).

OK, maybe your theory is correct. Maybe the representative quality of
your lists are not a manifestation of the fundamental mathematical
laws of probability but are instead the product of the whimsical humor
of a mechanical device.

Say hi to the other fairy tale creatures for me.

--
BR

[Michael Pendragon]

On May 25, 10:18 am, Bob Roman <robertjro...@gmail.com> wrote:
> On May 25, 9:36 am, Michael Pendragon
>
> <michaelmaleficapendra...@gmail.com> wrote:
> > I believe that statistical representation would only be inevitable in
> > a mathematically perfect world (that is, in theory -- not necessarily
> > in reality).
>
> Reality does not match a mathematical model when 1) the model is
> incorrect or 2) an error term is introduced.

I admire your faith in mathematics. I'm afraid I'm more of an
agnostic.

> In the present case the
> model is so fundamental (and demonstrated so regularly) that it cannot
> be incorrect.

To which I again cite my alleged misunderstanding of Mr. Twain.

One can provide an unlimited number of statistics to support *any*
given hypothesis. This in no way guarantees that the hypothesis is

correct.

> More to the point, in the present case, the model
> worked perfectly.

No, it didn't. It provided a surprisingly close match for the 10 most
represented artists: 6 of whom appeared in order. However, it is far
from being an exact match, and the differences increase the further
down the chart we go.

This merely shows that the early correspondences are the result of the
disproportionately high numbers of tracks for the most represented
artists. Artists with 600 tracks receive more play than 60.

It fails to show (much less to prove) that randomly generated
playlists will (when examined over a long period of time) ultimately
correspond to the number of tracks/artist.

> There is no evidence of an error term.  You are now
> arguing not only against basic probability theory but also against
> reality as it occurred.

See above.

> > I believe that far too many extenuating factors intrude
> > (including, in this particular instance, the method by which an ipod
> > "arbitrarily" selects songs when in "shuffle play" mode).
>
> OK, maybe your theory is correct.  Maybe the representative quality of
> your lists are not a manifestation of the fundamental mathematical
> laws of probability but are instead the product of the whimsical humor
> of a mechanical device.
>
> Say hi to the other fairy tale creatures for me.

Silly Bob. Mechanical devices of this sort rely upon specific
programs to operate. How random is the operating program of an ipod
shuffle?

[Mark Dintenfass]

In article
<1aefb21b-d317-4ede...@n33g2000vbi.googlegroups.com>,

Michael Pendragon <michaelmalef...@gmail.com> wrote:

> It fails to show (much less to prove) that randomly generated
> playlists will (when examined over a long period of time) ultimately
> correspond to the number of tracks/artist.

IIRC, assuming you have just one playlist, your iPod randomizes the
whole playlist and then plays the result in order. Over a long enough
period of time it will play each track once, and the played list will
exactly correspond exactly to the iPod contents. Sorry, once again, to
disturb your fantasies with a fact.

--
--md
_________
Remove xx's from address to reply

[Travis_Trite]

>
>
> > > I believe that far too many extenuating factors intrude
> > > (including, in this particular instance, the method by which an ipod
> > > "arbitrarily" selects songs when in "shuffle play" mode).
> >
> > OK, maybe your theory is correct. Maybe the representative quality of
> > your lists are not a manifestation of the fundamental mathematical
> > laws of probability but are instead the product of the whimsical humor
> > of a mechanical device.
> >
> > Say hi to the other fairy tale creatures for me.
>
> Silly Bob. Mechanical devices of this sort rely upon specific
> programs to operate. How random is the operating program of an ipod
> shuffle?
>

In water cooler/cocktail party talk about computing, random-number
generation is well-known to be a difficult task. (I haven't understood what
I've read as to WHY that is.)

I've had three mp3-playing devices and the randomizers on all of them will
play certain songs dozens of times, while other songs are played never, or
what I would think of as disproportionately seldom. On the fanciest of the
players I even felt like I was observing "weighting" going on, whereby if I
hit "next track" during the early part of a song, the player would handicap
that song and send it through less often.

--Travis

[Michael Pendragon]

On May 25, 11:27 am, Mark Dintenfass <mdintenf...@xxnew.rr.com> wrote:
> In article
> <1aefb21b-d317-4ede-b8d9-1dd27ba28...@n33g2000vbi.googlegroups.com>,

>
> Michael Pendragon <michaelmaleficapendra...@gmail.com> wrote:
> > It fails to show (much less to prove) that randomly generated
> > playlists will (when examined over a long period of time) ultimately
> > correspond to the number of tracks/artist.
>
> IIRC, assuming you have just one playlist, your iPod randomizes the
> whole playlist and then plays the result in order. Over a long enough
> period of time it will play each track once, and the played list will
> exactly correspond exactly to the iPod contents. Sorry, once again, to
> disturb your fantasies with a fact.

I'm sorry Mark, my use of "randomly generated playlists" was
inaccurate.

I haven't created any playlists.

I was using the term to refer to the 19,000+ tracks collectively. Or,
more specifically, to each of the randomly generated sets of songs
used on my "HLL" lists.

I believe that my ipod randomizes these 19,000+ tracks anew each time
I turn it on. I believe this because it has played certain songs
multiple times over a short period, and others not at all during that
same period. On numerous occasions, it has repeated one of the tracks
(from my 20 track lunch break lists) on the following day. Neither of
these instances would occur if it were running through the entire
library of songs before repeating one.

[Michael Pendragon]

Yes! That's exactly what happens with my ipod. Thanks Travis.

[Mark Dintenfass]

In article
<e9085878-d37e-49da...@3g2000vbx.googlegroups.com>,

Fair enough. With very large libraries, the iPod will generate a new
random list each time you turn it on, and so repeats will occur. If you
kept it on, I believe it would play through the entire list, as I said,
though 19,000 songs may be too large for it to randomize completely.
None of this has anything to do with your skepticism about statistics,
on which matters I would defer to Bob, who actually knows something
(beyond Mark Twain, anyway) about the subject.

[Mark Dintenfass]

In article <jpo8ij$qpl$1...@speranza.aioe.org>, Travis_Trite

<Travis...@gmail.com> wrote:

> On the fanciest of the
> players I even felt like I was observing "weighting" going on, whereby if I
> hit "next track" during the early part of a song, the player would handicap
> that song and send it through less often.

That would be a lot harder to program than a random generator and so I
suspect your observations are incorrect.

[Travis_Trite]

> >
> > > Silly Bob. Mechanical devices of this sort rely upon specific
> > > programs to operate. How random is the operating program of an ipod
> > > shuffle?
> >
> > In water cooler/cocktail party talk about computing, random-number
> > generation is well-known to be a difficult task. (I haven't understood
> > what
> > I've read as to WHY that is.)
> >
> > I've had three mp3-playing devices and the randomizers on all of them
> > will
> > play certain songs dozens of times, while other songs are played never,
> > or
> > what I would think of as disproportionately seldom. On the fanciest of
> > the
> > players I even felt like I was observing "weighting" going on, whereby
> > if I
> > hit "next track" during the early part of a song, the player would
> > handicap
> > that song and send it through less often.
>
> Yes! That's exactly what happens with my ipod. Thanks Travis.

To be fair to Mark D.'s point, I think the higher-end players (my higher-end
device imitates the ipod -- I avoid Apple) do work the way Mark describes.
Whether you've organized playlists or not, the player makes an internal
playlist from whatever is the subsetted area of the player, so it builds its
list from the "Songs" category.

[Travis_Trite]

"Mark Dintenfass" <mdint...@xxnew.rr.com> wrote in message
news:250520121054186510%mdint...@xxnew.rr.com...

> In article <jpo8ij$qpl$1...@speranza.aioe.org>, Travis_Trite
> <Travis...@gmail.com> wrote:
>
>> On the fanciest of the
>> players I even felt like I was observing "weighting" going on, whereby if
>> I
>> hit "next track" during the early part of a song, the player would
>> handicap
>> that song and send it through less often.
>
> That would be a lot harder to program than a random generator and so I
> suspect your observations are incorrect.
>

I'm inclined to agree that my mind is playing tricks on me, but I tried to
research it and I saw some indications that the idea is out there in the
industry. The kind of full-fledged media player one would have on one's
computer, like Itunes or Windows Media Player, do create most-played and
least-played auto-playlists, so the Player program does have the means of
ticking off how many times something goes through.

[Mark Dintenfass]

In article <jpoap3$tg$1...@speranza.aioe.org>, Travis_Trite

True enough. But as I understand probability, randomness does not
produce a perfect run of different songs. It's like flipping a coin and
getting five heads in a row. Probability predicts not only that five
heads in a row can happen, but with enough flips it will happen. And
the chances that it will happen in the first five flips are just as
good as the chances it will happen at flips 95-100. As I recall, back
in the old days when CD player had a random button, some of them were
truly random, choosing a new cut randomly each time. What that meant
was that on a twenty-cut CD, you were likely to start getting repeats
by the 7th or 8th tracks. It used to drive me nuts because our minds
aren't attuned to true randomness and try to impose some kind of sense
and order on things. The same principle on a larger scale is
responsible for most of the ridiculous stuff people believe.

[Michael Pendragon]

On May 25, 12:33 pm, Mark Dintenfass <mdintenf...@xxnew.rr.com> wrote:
> In article <jpoap3$t...@speranza.aioe.org>, Travis_Trite
>
>
>
>
>
>
>
>
>
> <Travis_Tr...@gmail.com> wrote:
> > "Mark Dintenfass" <mdintenf...@xxnew.rr.com> wrote in message
> >news:250520121054186510%mdint...@xxnew.rr.com...
> > > In article <jpo8ij$qp...@speranza.aioe.org>, Travis_Trite

In which case it's fair to say that a truly random set of plays would
not be expected perfecty match up with the number of tracks/artist.

[Michael Pendragon]

On May 25, 11:52 am, Mark Dintenfass <mdintenf...@xxnew.rr.com> wrote:
> In article

> <e9085878-d37e-49da-b0b3-efc6c29ff...@3g2000vbx.googlegroups.com>,

The battery would probably die before it reached 200 songs.

> None of this has anything to do with your skepticism about statistics,
> on which matters I would defer to Bob, who actually knows something
> (beyond Mark Twain, anyway) about the subject.

Certainly it has. In a mathematically perfect world (or one in which
my ipod battery could run 24/7 for 38 consecutive days), the number of
plays per song should (working on the assumption that the ipod can
accurately manage 19,014 files) perfectly match up with the number of
tracks.

My point is that this isn't a mathematically perfect world (and that
my battery only runs for an estimated 6 hours at best), and that as a
result the statistical probabilities would necessarily be compromised.

[F R]

Mike,
Perhaps your random brain cell generator is on "devil's advocate"?
Let's say your ipod is like a slot machine (which it is) as slots also
have a random number generator.
A basic slot machine might pay 5 credits for three cherries, the most
common winning combination as programmed...Frankie Laine or Elvis would
be the cherries.

It pays 25 credits for 3 peaches...Teresa Brewer would be the peaches.

Finally (to keep it simple; slot machines are not always simple), 3
watermelons pay 100 credits...that's Phil Harris.

Obviously, 3 cherries/Frankie Laine will come up much more often than 3
watermelons/Phil Harris.

The RNG is programmed like that on any slot machine.
Similarly, the RNG in your ipod is programmed to get 3 cherries more
often than 3 peaches and much, much more often than 3 watermelons.
It's really that simple, taking into account that it isn't perfectly
reflecting the number of songs by each artist in any one session.

Slots are calculated to pay out certain percentage "in the fullness of
time" as they call it. Same goes for your 19,000 songs.

I don't know how many months or years you've been doing this on a daily
basis, but it's possible you've reached the time requirement, which is
why over a period of time (as Geoff has tracked) it plays very closely
to your entire song list.

[Mark Dintenfass]

In article
<549f272d-1183-4b1e...@eh4g2000vbb.googlegroups.com>,

No. If you randomize an entire playlist and play it through without any
further input, it would perfectly match up. The problem you're
addressing is the result of re-randomizing.

[Michael Pendragon]

On May 25, 1:03 pm, Mark Dintenfass <mdintenf...@xxnew.rr.com> wrote:
> In article

> <549f272d-1183-4b1e-afb9-4671d1d54...@eh4g2000vbb.googlegroups.com>,

Now you're changing the parameters.

Returning to your "twenty-cut CD" analogy (above), you are "likely to
start getting repeats by the 7th or 8th tracks." The CD player in
your analogy is obviously *not* playing all of the tracks on the CD
before repeating some of them.

If the CD player were selecting the tracks on *a truly random basis,*
it's improbable but not inconceivable that the same track could get
repeated 20 times in a row.

[Michael Pendragon]

On May 25, 12:54 pm, espon...@webtv.net (F R) wrote:
> Mike,
> Perhaps your random brain cell generator is on "devil's advocate"?

It's been stuck there since I was 3.

> Let's say your ipod is like a slot machine (which it is) as slots also
> have a random number generator.
> A basic slot machine might pay 5 credits for three cherries, the most
> common winning combination as programmed...Frankie Laine or Elvis would
> be the cherries.
>
> It pays 25 credits for 3 peaches...Teresa Brewer would be the peaches.
>
> Finally (to keep it simple; slot machines are not always simple), 3
> watermelons pay 100 credits...that's Phil Harris.
>
> Obviously, 3 cherries/Frankie Laine will come up much more often than 3
> watermelons/Phil Harris.
>
> The RNG is programmed like that on any slot machine.
> Similarly, the RNG in your ipod is programmed to get 3 cherries more
> often than 3 peaches and much, much more often than 3 watermelons.
> It's really that simple, taking into account that it isn't perfectly
> reflecting the number of songs by each artist in any one session.
>
> Slots are calculated to pay out certain percentage "in the fullness of
> time" as they call it. Same goes for your 19,000 songs.

That makes perfect sense ... and it seems to hold true for the
cherries and peaches. The trouble is with the watermelons. To take
only one example: I've noticed that my ipod has a particular fondness
for "I'd Rather Charleston" by Fred and Adele Astaire. I'd say that
it plays this song on an average of once a month. I average about 4 1-
hour HLL lists per week (16/month). That's about 325 songs/month.
This means that it would take a total of 58.5 months (or nearly 5
years) for my ipod to get through all of my 19,014 tracks. Which
further means that "Charleston" is getting 58.5 times the amount of
plays that it statistically should.

> I don't know how many months or years you've been doing this on a daily
> basis, but it's possible you've reached the time requirement, which is
> why over a period of time (as Geoff has tracked) it plays very closely
> to your entire song list.

It is surprisingly close. :-)

[Mark Dintenfass]

In article
<9299d7aa-9cf7-47b5...@s9g2000vbg.googlegroups.com>,

You're either confused or obfuscating. If you randomize a list of
19,000 tracks and play that list, each track will play exactly once.
That is altogether different from making a random choice from the whole
list for each play, in which case you might indeed get the same cut
twenty times in a row--though the odds of that would be small. You're
just confusing two altogether different forms of random selection.

Your agenda seems to be your notion that the math behind probability
theory is "imperfect," though my own inclination would be to put far
greater trust in math as a way of describing the world accurately than
in your intuitions.

[Mark Dintenfass]

In article
<1576af95-a651-436c...@d6g2000vbe.googlegroups.com>,

Michael Pendragon <michaelmalef...@gmail.com> wrote:

> To take
> only one example: I've noticed that my ipod has a particular fondness
> for "I'd Rather Charleston" by Fred and Adele Astaire. I'd say that
> it plays this song on an average of once a month. I average about 4 1-
> hour HLL lists per week (16/month). That's about 325 songs/month.
> This means that it would take a total of 58.5 months (or nearly 5
> years) for my ipod to get through all of my 19,014 tracks. Which
> further means that "Charleston" is getting 58.5 times the amount of
> plays that it statistically should.

Typical. You start with a subjective set of impressions ( "I'd say that
it plays this song on an average of once a month"), apply imprecise
numbers to them ("I average about 4 1-hour lists a week"), and end up
with a seemingly definitive 58.5, believing you've torn some sort of
hole in the rules of statistical probability. Sheesh.

[The Bloomfield Bloviator]

On May 25, 12:33 pm, Mark Dintenfass <mdintenf...@xxnew.rr.com> wrote:
>

> the chances that it will happen in the first five flips are just as
> good as the chances it will happen at flips 95-100.

Afraid not.

Since 95-100 would be 6 flips rather than 5, there is a better chance
of gettin g 5 heads in a row over the course of 6 flips than over the
course of 5 flips (1-5).

[The Bloomfield Bloviator]

On May 25, 12:54 pm, espon...@webtv.net (F R) wrote:

>

> It pays 25 credits for 3 peaches...Teresa Brewer would be the peaches.

No, Miss Peaches (Etta James) would be the Peaches.

[Bob Roman]

On May 25, 10:53 am, Michael Pendragon

<michaelmaleficapendra...@gmail.com> wrote:
> On May 25, 10:18 am, Bob Roman <robertjro...@gmail.com> wrote:

> > Reality does not match a mathematical model when 1) the model is
> > incorrect or 2) an error term is introduced.
>
> I admire your faith in mathematics.  I'm afraid I'm more of an
> agnostic.

I'm afraid you don't understand the issues well enough to hold a
position one way or the other.

> To which I again cite my alleged misunderstanding of Mr. Twain.
>
> One can provide an unlimited number of statistics to support *any*
> given hypothesis.  This in no way guarantees that the hypothesis is
> correct.

Statistics are misused, both by those who choose to misapply them and
by those who do not understand them. Knowledgeable people can
identify a misused statistic.

None of that, however, applies to the topic that Frank tried to bring
to your attention. Twain's "lies, damned lies, and statistics" quote
points to flaws in the "external validity" of some statistical data
(often the result of a disconnect between a sample and the larger
population). Frank was referring to the "reliability" of probability
theory itself.

Only a fool doubts the reliability of probability theory itself. Mark
Twain was not a fool. That is why I say you misunderstand him.

> No, it didn't.  It provided a surprisingly close match for the 10 most
> represented artists: 6 of whom appeared in order.  However, it is far
> from being an exact match, and the differences increase the further
> down the chart we go.
>
> This merely shows that the early correspondences are the result of the
> disproportionately high numbers of tracks for the most represented
> artists. Artists with 600 tracks receive more play than 60.
>
> It fails to show (much less to prove) that randomly generated
> playlists will (when examined over a long period of time) ultimately
> correspond to the number of tracks/artist.

Now you are misusing statistics. You earlier referred to a
"surprising" correspondence between your population and your sample.
Now you are saying that there is not a perfect correspondence. The
lack of a perfect correspondence is the result of the error term.
There is no reason to believe that the error term will hold steady as
your sample size grows larger. On the contrary, it is inevitable that
the error term will disappear as the sample size grows larger.

--
BR

[The Bloomfield Bloviator]

On May 25, 1:31 pm, Michael Pendragon

<michaelmaleficapendra...@gmail.com> wrote:
>
> It's been stuck there since I was 3.

Just like your shit taste in music.

[Roger Ford]

On Thu, 24 May 2012 21:01:48 -0700 (PDT), Bob Roman
<robert...@gmail.com> wrote:

>On May 24, 11:38=A0pm, Michael Pendragon
><michaelmaleficapendra...@gmail.com> wrote:
>> On May 24, 10:54=A0pm, Bob Roman <robertjro...@gmail.com> wrote:
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> > On May 24, 10:50=A0pm, Michael Pendragon
>>
>> > <michaelmaleficapendra...@gmail.com> wrote:
>> > > On May 24, 8:22=A0pm, Bob Roman <robertjro...@gmail.com> wrote:
>>
>> > > > On May 24, 4:07=A0pm, Michael Pendragon
>>
>> > > > <michaelmaleficapendra...@gmail.com> wrote:

>> > > > > On May 24, 1:01=A0pm, espon...@webtv.net (F R) wrote:
>>
>> > > > > > Pendragon

>> > > > > > > It's surprising that the breakdown appears to be sticking pre=

>tty closely
>> > > > > > > to the number of tracks per artist I've actually got.
>> > > > > > -----------------------

>> > > > > > It shouldn't be surprising in the least =A0Mike. Assuming, as G=
>eoff has,
>> > > > > > that your daily dosage of music is truly random, doesn't it sta=
>nd to
>> > > > > > reason that the more tracks you have by an individual artist SH=

>OULD
>> > > > > > reflect exactly that when they are randomly selected?
>>

>> > > > > In a mathematically perfect world. =A0But my ipod often seems to =

>be
>> > > > > quite capricious in its selections.
>>
>> > > > Can you tell us why you find "capricious" representativeness to be
>> > > > more "surprising" than statistical representativeness?
>>
>> > > No ... because I said I find statistical representation to be more
>> > > surprising.
>>

>> > Statistical representation is inevitable in the long run. =A0I don't

>> > think you understand Frank's point.
>>
>> I'm a Mark Twain fan.
>

>Who isn't? But I don't think you quite understand Twain's point
>either.
>

IIRC wasn't one of Twains more minor accomplishments being born with
Halley's Comet in the sky and dying with Halley's Comet in the sky

ROGER FORD
-----------------------

"Spam Free Zone" - to combat unwanted automatic spamming I have added
an extra "b" in my e-mail address (mari...@bblueyonder.co.uk).
Please delete same before responding.Thank you!

[Mark Dintenfass]

In article <4fbfed99...@news.virginmedia.com>, Roger Ford

<mari...@bblueyonder.co.uk> wrote:


> >
> IIRC wasn't one of Twains more minor accomplishments being born with
> Halley's Comet in the sky and dying with Halley's Comet in the sky

At about the time Bill Haley was born. He eventually acquired multiple
Comets. :-)

[Len Blanks]

"Travis_Trite" <Travis...@gmail.com> writes:

>>
>>
>> > > I believe that far too many extenuating factors intrude
>> > > (including, in this particular instance, the method by which an ipod
>> > > "arbitrarily" selects songs when in "shuffle play" mode).
>> >
>> > OK, maybe your theory is correct. Maybe the representative quality of
>> > your lists are not a manifestation of the fundamental mathematical
>> > laws of probability but are instead the product of the whimsical humor
>> > of a mechanical device.
>> >
>> > Say hi to the other fairy tale creatures for me.
>>
>> Silly Bob. Mechanical devices of this sort rely upon specific
>> programs to operate. How random is the operating program of an ipod
>> shuffle?
>>
>
>
> In water cooler/cocktail party talk about computing, random-number
> generation is well-known to be a difficult task. (I haven't understood what
> I've read as to WHY that is.)

Pseudo-random number generation is quite simple, the algorithms are
quite easy to understand, and programs developed from those algorithms
are short and fast. The problem is that the sequence of numbers
produced is *not* random, hence the 'pseudo' qualifier. For one thing
the sequence is periodic, that is, the sequence is finite, and repeats.
A sequence with a period of three might generate the following:
0.2396108 0.9664275 0.1000241 0.2396108 0.9664275 ... There are
other problems as well, but you get the idea.

If you want true mathematical randomness in a device, software falls
short. A hardware random number generator which samples thermal noise
(from a diode, for example) provides true randomness. I reckon ipods
use software due to cost considerations and the fact that simple task of
shuffling tunes is not critical.

> I've had three mp3-playing devices and the randomizers on all of them will
> play certain songs dozens of times, while other songs are played never, or
> what I would think of as disproportionately seldom. On the fanciest of the
> players I even felt like I was observing "weighting" going on, whereby if I
> hit "next track" during the early part of a song, the player would handicap
> that song and send it through less often.

--

Len

I'd be unworthy of the high trust that's been placed in me if I didn't
do everything in my power to keep our beloved Freedonia at peace with
the world. I'd be only too happy to meet with Ambassador Trentino and
offer him on behalf of my country the right hand of good fellowship.
And I feel sure he will accept this gesture in the spirit of which it is
offered. But suppose he doesn't? A fine thing that'll be. I hold out my
hand and he refuses to accept. That'll add a lot to my prestige, won't
it? Me, the head of a country, snubbed by a foreign ambassador. Who does
he think he is, that he can come here, and make a sap of me in front of
all my people? Think of it! I hold out my hand and that hyena refuses
to accept! Why, the cheap four-flushing swine! He'll never get away with
it, I tell you! He'll never get away with it! -- Rufus T Firefly

[Len Blanks]

Very well said. I used to love the reaction of people at the post office
waiting to buy a lottery ticket when I asked for the winning numbers
for the previous week, and choose them for my entries for that week.
They seemed to believe that the numbers coming up again were nearly
impossible and I explained why the odds on the reused numbers were
the same as any numbers they had chosen.

--
Len

Ειναι ολα Ελληνικα για μενα.

[Mark Dintenfass]

In article <m2wr402...@haruspex.net>, Len Blanks

A very good friend, a math-minded economist, used to buy a single
Superball ticket only when the prize for a dollar ticket, with the odds
of winning set at 177,000,000 to 1, surpassed $177,000,001. The odds,
he would declare ceremoniously, were now in his favor.

[The Bloomfield Bloviator]

On May 25, 8:21 pm, Mark Dintenfass <mdintenf...@xxnew.rr.com> wrote:
> In article <m2wr402bcq....@haruspex.net>, Len Blanks
>
>
>
>
>
> <use...@smashcapitalism.com> wrote:

> > Mark Dintenfass <mdintenf...@xxnew.rr.com> writes:
>
> > > In article <jpoap3$t...@speranza.aioe.org>, Travis_Trite
> > > <Travis_Tr...@gmail.com> wrote:
>

> > >> "Mark Dintenfass" <mdintenf...@xxnew.rr.com> wrote in message
> > >>news:250520121054186510%mdint...@xxnew.rr.com...
> > >> > In article <jpo8ij$qp...@speranza.aioe.org>, Travis_Trite

Not really, because there can be more than one winner.

[Mark Dintenfass]

In article
<c1ea437c-cba6-4074...@e18g2000yqc.googlegroups.com>,

No kidding? I bet he never even thought of that!

[Michael Pendragon]

On May 25, 2:37 pm, Mark Dintenfass <mdintenf...@xxnew.rr.com> wrote:
> In article

> <9299d7aa-9cf7-47b5-b530-76dba9e34...@s9g2000vbg.googlegroups.com>,

Yes.

> That is altogether different from making a random choice from the whole
> list for each play, in which case you might indeed get the same cut
> twenty times in a row--though the odds of that would be small.

Correct again. So far we're on the same page.

> You're
> just confusing two altogether different forms of random selection.

No. I was addressing the second form -- which you'd introduced in the
paragraph I was responding to.

> Your agenda seems to be your notion that the math behind probability
> theory is "imperfect,"

Not at all. I'm saying that the math behind probability theory is
perfect.

My point is that it is only perfect in a ("mathematically perfect")
world that conforms to the conditions that theory is predicated upon.

> though my own inclination would be to put far
> greater trust in math as a way of describing the world accurately than
> in your intuitions.

I'm not going with my intuitions.

I have 38.1 days worth of music on my ipod. If my ipod could play for
38.1 days without a stop, it would play each track once. However, my
ipod cannot play for 38.1 straight days, so any conclusions drawn from
this model, while mathematically correct, have no bearing on how my
ipod actually operates (as we've already agreed, it randomizes its
19,014 tracks anew each time I turn it on).

[Michael Pendragon]

On May 25, 2:45 pm, Mark Dintenfass <mdintenf...@xxnew.rr.com> wrote:
> In article

> <1576af95-a651-436c-a1e6-f8b82ae17...@d6g2000vbe.googlegroups.com>,

>
> Michael Pendragon <michaelmaleficapendra...@gmail.com> wrote:
> >  To take
> > only one example: I've noticed that my ipod has a particular fondness
> > for "I'd Rather Charleston" by Fred and Adele Astaire.  I'd say that
> > it plays this song on an average of once a month.  I average about 4 1-
> > hour HLL lists per week (16/month).  That's about 325 songs/month.
> > This means that it would take a total of 58.5 months (or nearly 5
> > years) for my ipod to get through all of my 19,014 tracks.  Which
> > further means that "Charleston" is getting 58.5 times the amount of
> > plays that it statistically should.
>
> Typical. You start with a subjective set of impressions ( "I'd say that
> it plays this song on an average of once a month"), apply imprecise
> numbers to them ("I average about 4 1-hour lists a week"), and end up
> with a seemingly definitive 58.5, believing you've torn some sort of
> hole in the rules of statistical probability. Sheesh.

I have neither the time nor the inclination to check through my HLL
posts from the past year to confirm exactly how many times
"Charleston" was actually played. Once a month is a relatively fair
estimate.

The 58.5 projection isn't meant to be definitive (it's a rough
projection), nor was it ever presented as such.

And as to my having "torn some sort of hole in the rules of
statistical probability," that's your phrase, not mine. I only mean
for it to show that my ipod, due to the limitations of its battery
life and as used by me in the real world, does not follow the ideal
probability model (of playing each track once before repeating) that
you'd set forth.

[Len Blanks]

I used to play the Lotto weekly in Germany, simply because I feared forgetting
and having the numbers come up and being murdered by my wife.

We were watching the drawing on TV and received a frantic call from one of our
friends who screamed that she had won the jackpot - a few million Marks! The
next day she (and we) learned that she had to share it with a little over 8000
others. The Germans like to play multiple times and after the normal weekly
choices (birthdays and the like), they would choose the 6 numbers randomly,
many times making letters and such from the form. I believe it was a 'T' they
all chose.
--
Len

Dw i'n gallu bwyta gwydr, 'dyw e ddim yn gwneud dolur i mi.

[Michael Pendragon]

Yep.

Btw -- Congratulations on your recent victory in The Miss Ugly-
Dougling Contest.

[Michael Pendragon]

On May 25, 3:07 pm, Bob Roman <robertjro...@gmail.com> wrote:
> On May 25, 10:53 am, Michael Pendragon
>
> <michaelmaleficapendra...@gmail.com> wrote:
> > On May 25, 10:18 am, Bob Roman <robertjro...@gmail.com> wrote:
> > > Reality does not match a mathematical model when 1) the model is
> > > incorrect or 2) an error term is introduced.
>
> > I admire your faith in mathematics.  I'm afraid I'm more of an
> > agnostic.
>
> I'm afraid you don't understand the issues well enough to hold a
> position one way or the other.

A thrown rock can hold the opinion that it's flying.

> > To which I again cite my alleged misunderstanding of Mr. Twain.
>
> > One can provide an unlimited number of statistics to support *any*
> > given hypothesis.  This in no way guarantees that the hypothesis is
> > correct.
>
> Statistics are misused, both by those who choose to misapply them and
> by those who do not understand them.  Knowledgeable people can
> identify a misused statistic.

Knowledgeable people can ... provided that they're able to keep their
personal biases from influencing their perception.

> None of that, however, applies to the topic that Frank tried to bring
> to your attention.  Twain's "lies, damned lies, and statistics" quote
> points to flaws in the "external validity" of some statistical data
> (often the result of a disconnect between a sample and the larger
> population).  Frank was referring to the "reliability" of probability
> theory itself.

Twain's quote (which didn't actually originate with him) is meant to
be a much broader generalization -- although one could certainly apply
it a "disconnect between a sample and the larger population." Twain's
meaning is really much simpler: that one can (mis-)use statistics to
support a false claim.

> Only a fool doubts the reliability of probability theory itself.  Mark
> Twain was not a fool.  That is why I say you misunderstand him.

As usual, you merely *chose* not to understand me.

> > No, it didn't.  It provided a surprisingly close match for the 10 most
> > represented artists: 6 of whom appeared in order.  However, it is far
> > from being an exact match, and the differences increase the further
> > down the chart we go.
>
> > This merely shows that the early correspondences are the result of the
> > disproportionately high numbers of tracks for the most represented
> > artists. Artists with 600 tracks receive more play than 60.
>
> > It fails to show (much less to prove) that randomly generated
> > playlists will (when examined over a long period of time) ultimately
> > correspond to the number of tracks/artist.
>
> Now you are misusing statistics.

No. I'm not using them at all.

> You earlier referred to a
> "surprising" correspondence between your population and your sample.
> Now you are saying that there is not a perfect correspondence.

Yes, Bob. Is this concept really so hard to understand? The
correspondence is surprisingly close, but not perfect.

> The
> lack of a perfect correspondence is the result of the error term.
> There is no reason to believe that the error term will hold steady as
> your sample size grows larger.  On the contrary, it is inevitable that
> the error term will disappear as the sample size grows larger.

It is not inevitable because my ipod re-randomizes with each play. If
my ipod continues to play "I'd Rather Charleston" once a month
(perhaps its operating program has a disproportionate tendency to
select track #10 when arranged by artist), the disparity between the
sample and population would only grow.

[F R]

Pendragon>

Yes, Bob. Is this concept really so hard to understand? The
correspondence is surprisingly close, but not perfect.

--------------------------
Of course not. As I've learned, you have not played your list in the
fullness of time as I first thought. if you play 100 songs a week, your
at 5200 for a year. That's only 25% of your songs and of course you'd
have many random replays.

Let's say you program 100 songs to be played randomly.
90 Elvis
9 Sinatra
1 Anka
After the ipod plays the 100 tunes randomly, it would be against the
odds that you'd get the exact proportion shown above. Perhaps in no span
of 100 numbers do you get that exact breakdown. that's not the fullness
of time, but anecdotal.

But if you randomly play those hundred songs 100,000 times they will be
tiny fractions in the 1000ths away from the projected number. Maybe
you'd get 90,007 Elvis songs, 893 Sinatra and exactly 1000 Anka. Next
time you did it, the numbers would not necessarily match the previous
trial but will be in the same range.

[Bob Roman]

On May 25, 10:54 pm, Michael Pendragon

<michaelmaleficapendra...@gmail.com> wrote:
> On May 25, 3:07 pm, Bob Roman <robertjro...@gmail.com> wrote:

> > None of that, however, applies to the topic that Frank tried to bring
> > to your attention.  Twain's "lies, damned lies, and statistics" quote
> > points to flaws in the "external validity" of some statistical data
> > (often the result of a disconnect between a sample and the larger
> > population).  Frank was referring to the "reliability" of probability
> > theory itself.
>
> Twain's quote (which didn't actually originate with him) is meant to
> be a much broader generalization -- although one could certainly apply
> it a "disconnect between a sample and the larger population."  Twain's
> meaning is really much simpler: that one can (mis-)use statistics to
> support a false claim.

Yes, that's what "external validity" means. It still has nothing to
do with Frank's point.

> > The
> > lack of a perfect correspondence is the result of the error term.
> > There is no reason to believe that the error term will hold steady as
> > your sample size grows larger.  On the contrary, it is inevitable that
> > the error term will disappear as the sample size grows larger.
>
> It is not inevitable because my ipod re-randomizes with each play.  If
> my ipod continues to play "I'd Rather Charleston" once a month
> (perhaps its operating program has a disproportionate tendency to
> select track #10 when arranged by artist), the disparity between the
> sample and population would only grow.

Sampling with replacement or sampling without replacement, the sample
still grows closer to the population the larger it gets.

But enjoy your impishly capricious Ipod. It sounds charming.

--
BR

[Michael Pendragon]

On May 26, 12:54 am, espon...@webtv.net (F R) wrote:
> Pendragon>
> Yes, Bob. Is this concept really so hard to understand? The
> correspondence is surprisingly close, but not perfect.
> --------------------------
> Of course not. As I've learned, you have not played your list in the
> fullness of time as I first thought. if you play 100 songs a week, your
> at 5200 for a year. That's only 25% of your songs and of course you'd
> have many random replays.
>
> Let's say you program 100 songs to be played randomly.
> 90 Elvis
> 9 Sinatra
> 1  Anka
> After the ipod plays the 100 tunes randomly, it would be against the
> odds that you'd get the exact proportion shown above.

No, provided that it played them without stopping, it would play each
song exactly once.

> Perhaps in no span
> of 100 numbers do you get that exact breakdown. that's not the fullness
> of time, but anecdotal.
>
> But if you randomly play those hundred songs 100,000 times they will be
> tiny fractions in the 1000ths away from the projected number. Maybe
> you'd get 90,007 Elvis songs, 893 Sinatra and exactly 1000 Anka. Next
> time you did it, the numbers would not necessarily match the previous
> trial but will be in the same range.

But you're still basing those figures on a presupposition that my ipod
selects the songs randomly (and without resetting itself after each
play) ... and that I'd have a manageable number of song files.

As Len explained, the typical mp3 player program is based on periodic
algorithmic sequences, and thus, not random. Plus, my ipod resets its
randomization after each play (which, regarding the HLL lists) consist
of 20 - 25 songs.
Plus, I add new tracks to my ipod several times a week, which would
again require the ipod to generate a new random playlist.
Plus, even if set my ipod to "shuffle," and left it to play for the
38.1 days (it lists this time) required to play all the songs once,
the battery would die after the first 200 or so -- again forcing the
randomizer to reset.

So, yes, if I only had 100 songs on my ipod, 100,000 random plays
would ultimately provide a more or less accurate statistical
reflection of the number of tracks.

However, your 100 playlist is theoretical, where as my 19,029 (and
counting) playlist is the reality.

[Michael Pendragon]

On May 26, 1:06 am, Bob Roman <robertjro...@gmail.com> wrote:
> On May 25, 10:54 pm, Michael Pendragon
>
> <michaelmaleficapendra...@gmail.com> wrote:
> > On May 25, 3:07 pm, Bob Roman <robertjro...@gmail.com> wrote:
> > > None of that, however, applies to the topic that Frank tried to bring
> > > to your attention.  Twain's "lies, damned lies, and statistics" quote
> > > points to flaws in the "external validity" of some statistical data
> > > (often the result of a disconnect between a sample and the larger
> > > population).  Frank was referring to the "reliability" of probability
> > > theory itself.
>
> > Twain's quote (which didn't actually originate with him) is meant to
> > be a much broader generalization -- although one could certainly apply
> > it a "disconnect between a sample and the larger population."  Twain's
> > meaning is really much simpler: that one can (mis-)use statistics to
> > support a false claim.
>
> Yes, that's what "external validity" means.  It still has nothing to
> do with Frank's point.

It does insofar as it notes that statistics cannot be trusted.

Frank's statement is theoretically correct. But it doesn't apply to
my ipod, because there are extenuating factors connected with my ipod
that do not allow for a true randomization to take place.

I don't mean to imply that Frank is being dishonest. I'm saying that
while his statement is accurate regarding an ipod with 100 tracks, it
is not applicable to one containing 19,029 tracks.

> > > The
> > > lack of a perfect correspondence is the result of the error term.
> > > There is no reason to believe that the error term will hold steady as
> > > your sample size grows larger.  On the contrary, it is inevitable that
> > > the error term will disappear as the sample size grows larger.
>
> > It is not inevitable because my ipod re-randomizes with each play.  If
> > my ipod continues to play "I'd Rather Charleston" once a month
> > (perhaps its operating program has a disproportionate tendency to
> > select track #10 when arranged by artist), the disparity between the
> > sample and population would only grow.
>
> Sampling with replacement or sampling without replacement, the sample
> still grows closer to the population the larger it gets.

No; in fact the opposite occurs. If my ipod's algorithmic sequences
select track #10 on an average of once out of every 300 plays, then
the accumulated plays for track #10 will grow disproportionately
higher the larger the sampling grows.

> But enjoy your impishly capricious Ipod.  It sounds charming.

I do. And it is.

It does a lot of quirky little things with a frequency that I can only
attribute to synchronicity. For example, playing a song from the
soundtrack of Mary Poppins precisely at the moment that I pass the
theater were the stage production of that show is playing. Or playing
Give My Regards to Broadway as I'm walking through Herald Square. Or
by playing a high percentage of country songs on the day I made the
"I'm a Little Bit Country" post.

These meaningful "random" occurrences open up the way for a discussion
on several related topics -- but I'm afraid they're all too
metaphysical for most members of this empirically oriented group.

[F R]

But if you randomly play those hundred songs 100,000 times they will be
tiny fractions in the 1000ths away from the projected number. Maybe
you'd get 90,007 Elvis songs, 893 Sinatra and exactly 1000 Anka. Next
time you did it, the numbers would not necessarily match the previous
trial but will be in the same range.

-----------------
I meant to type 8993 for Sinatra, not 893...but you got the drift of my
point.

Dishonest? Moi? Why would you even think I was being dishonest about
this, MIke? Wrong...perhaps, but dishonest? Nah?

Now it seems to be moot as your RNG is not sophisticated enough to give
true randomness on the one hand (at least as you explain) yet it was
close to dead on when Geoff showed and you confirmed how close,
percentage wise it really was to the number of plays for each artist.

[Bob Roman]

On May 26, 2:28 am, Michael Pendragon

<michaelmaleficapendra...@gmail.com> wrote:

> I don't mean to imply that Frank is being dishonest.  I'm saying that
> while his statement is accurate regarding an ipod with 100 tracks, it
> is not applicable to one containing 19,029 tracks.

> It does a lot of quirky little things with a frequency that I can only
> attribute to synchronicity.  For example, playing a song from the
> soundtrack of Mary Poppins precisely at the moment that I pass the
> theater were the stage production of that show is playing.  Or playing
> Give My Regards to Broadway as I'm walking through Herald Square.  Or
> by playing a high percentage of country songs on the day I made the
> "I'm a Little Bit Country" post.
>
> These meaningful "random" occurrences open up the way for a discussion
> on several related topics -- but I'm afraid they're all too
> metaphysical for most members of this empirically oriented group.

You should do a little reading on the phenomenon of Illusory
Correlation.

When coincidences occur, it is a normal human tendency to look for
patterns or reasons in the chaos. It does not mean they are really
there.

--
BR

[F R]

Bob>

You should do a little reading on the phenomenon of Illusory
Correlation.
When coincidences occur, it is a normal human tendency to look for
patterns or reasons in the chaos. It does not mean they are really
there.

------------------------
I had a really great coincidence just last week, Bob.
I had gone to my library and taken out a few books, one of which was
titled "Last Dance, Last Chance". It's a true crime murder book written
by Ann Rule.

I started reading it for maybe 30 minutes and turned on the TV to see if
would ever stop raining in South Florida but there was "breaking news".
Lo and behold, the breaking news was that Donna Summer had died.

[RWC]

On Fri, 25 May 2012 10:31:34 -0700 (PDT), Michael Pendragon
<michaelmalef...@gmail.com> wrote:

As regards how the iPod shuffle works, you could find some useful stuff here:
http://ipod.about.com/od/advanceditunesuse/a/itunes-random.htm

My Music database WHLL table deliberately does not allow duplicate artist-song
records. But I quickly created an empty copy of this table, minus the
aforementioned restriction, and re-posted (taking a fraction of a second) my
"pasted from HLL messages" running text file of your 2012 playlists (up to May
23).

There are 1387 entries in this new temporary table compared to 1347 in the
original table - so, we have 40 duplicate artist-songs up to May 23, Mike.
These being - using the SQL command (-:

SELECT WHLLID, round((((1387-WHLLID)*134/1387)+10)/7) as week, artist, song
FROM WHLLtemp
WHERE (artist || song) IN
( SELECT (artist || song) FROM WHLLtemp
GROUP BY (artist || song)
HAVING count(*) > 1)
ORDER BY artist, song, week

very approx.
week into 2012
624 11 Arthur Godfrey & Laurie Anders I Like the Wide Open Spaces
542 13 Arthur Godfrey & Laurie Anders I Like the Wide Open Spaces
1,301 2 Bing Crosby That Tumbledown Shack in Athlone
451 14 Bing Crosby That Tumbledown Shack in Athlone
655 11 Don McLean Babylon
434 14 Don McLean Babylon
1,267 3 Doris Day & John Raitt Small Talk
961 7 Doris Day & John Raitt Small Talk
710 10 Eddy Arnold Bouquet of Roses
521 13 Eddy Arnold Bouquet of Roses
686 11 Elvis Presley Don't Cry Daddy
416 14 Elvis Presley Don't Cry Daddy
370 15 Elvis Presley Don't Cry Daddy
991 6 Elvis Presley From a Jack to a King
842 8 Elvis Presley From a Jack to a King
1,004 6 Elvis Presley I'm Leavin'
180 18 Elvis Presley I'm Leavin'
1,080 5 Elvis Presley Miracle of the Rosary
905 8 Elvis Presley Miracle of the Rosary
1,336 2 Frankie Laine Cool Water
1,196 4 Frankie Laine Cool Water
1,046 6 Frankie Laine Cool Water
950 7 Frankie Laine Someday, Sweetheart
887 8 Frankie Laine Someday, Sweetheart
1,271 3 Frankie Laine That Lucky Old Sun (Just Rolls Around Heaven All
Day)
1,115 5 Frankie Laine That Lucky Old Sun (Just Rolls Around Heaven All
Day)
1,168 4 Gene Autry The Yellow Rose of Texas
742 10 Gene Autry The Yellow Rose of Texas
814 9 Guy Mitchell Memory Number One
629 11 Guy Mitchell Memory Number One
1,254 3 Guy Mitchell Rock-a-Billy
11 20 Guy Mitchell Rock-a-Billy
290 16 Harry James Orchestra, Frank Sinatra v. Here Comes the Night
67 19 Harry James Orchestra, Frank Sinatra v. Here Comes the Night
1,290 2 Jerry Lee Lewis Big Blon' Baby
1,264 3 Jerry Lee Lewis Big Blon' Baby
172 18 Jim Reeves I'll Follow You
108 19 Jim Reeves I'll Follow You
1,259 3 Julie London Mad About the Boy
1 20 Julie London Mad About the Boy
685 11 Ken Darby Male Chorus There Is Nothin' Like a Dame
327 16 Ken Darby Male Chorus There Is Nothin' Like a Dame
1,079 5 Louis Jordan & His Tympani 5 If It?s Love You Want Baby,
That?s Me
574 12 Louis Jordan & His Tympani 5 If It?s Love You Want Baby,
That?s Me
230 17 Nat 'King' Cole A Fool Was I
82 19 Nat 'King' Cole A Fool Was I
347 15 Nat 'King' Cole Sweet Lorraine
68 19 Nat 'King' Cole Sweet Lorraine
708 10 Pat Boone Remember You're Mine
22 20 Pat Boone Remember You're Mine
820 9 Perry Como The Most Beautiful Girl
735 10 Perry Como The Most Beautiful Girl
1,045 6 Russ Morgan Orchestra Moonlight Serenade
466 14 Russ Morgan Orchestra Moonlight Serenade
1,213 3 Sarah Vaughan Mean to Me
437 14 Sarah Vaughan Mean to Me
819 9 Shirley Bassey Moonraker
285 16 Shirley Bassey Moonraker
974 7 Sonny James Heaven Says Hello
508 13 Sonny James Heaven Says Hello
500 13 Spike Jones & His City Slickers Tennessee Waltz
133 18 Spike Jones & His City Slickers Tennessee Waltz
1,087 5 Sylvia Syms Some People
86 19 Sylvia Syms Some People
760 10 Teresa Brewer Bell Bottom Blues
116 18 Teresa Brewer Bell Bottom Blues
787 9 Teresa Brewer I Left My Heart in San Francisco
569 12 Teresa Brewer I Left My Heart in San Francisco
636 11 Teresa Brewer More Than You Know
81 19 Teresa Brewer More Than You Know
1,070 5 The Andrews Sisters Tulip Time
345 15 The Andrews Sisters Tulip Time
92 19 The Four Lads Moments to Remember
98 19 The Four Lads Moments to Remember
241 17 The McGuire Sisters Sugartime
5 20 The McGuire Sisters Sugartime
1,364 1 Zayra Alvarez Alma Boricua
536 13 Zayra Alvarez Alma Boricua

I don't have the actual Usenet posting date of each HLL artist-song (since not
relevant to my project) but the sequentially allocated record id above will give
a rough indication since 1 to 1387 represents the date range May 23 going
backwards to Jan 10 (134 days).


Geoff

[Michael Pendragon]

On May 26, 9:10 am, espon...@webtv.net (F R) wrote:
> But if you randomly play those hundred songs 100,000 times they will be
> tiny fractions in the 1000ths away from the projected number. Maybe
> you'd get 90,007 Elvis songs, 893 Sinatra and exactly 1000 Anka. Next
> time you did it, the numbers would not necessarily match the previous
> trial but will be in the same range.
> -----------------
> I meant to type 8993 for Sinatra, not 893...but you got the drift of my
> point.
>
> Dishonest? Moi? Why would you even think I was being dishonest about
> this, MIke? Wrong...perhaps, but dishonest? Nah?

I wouldn't think you were. It was a preemptive measure designed to
stop Bob from accusing me of thinking you were dishonest.

> Now it seems to be moot as your RNG is not sophisticated enough to give
> true randomness on the one hand (at least as you explain) yet it was
> close to dead on when Geoff showed and you confirmed how close,
> percentage wise it really was to the number of plays for each artist.

Which is why I find it surprising.

[Michael Pendragon]

On May 26, 9:56 am, Bob Roman <robertjro...@gmail.com> wrote:
> On May 26, 2:28 am, Michael Pendragon
>
> <michaelmaleficapendra...@gmail.com> wrote:
> > I don't mean to imply that Frank is being dishonest.  I'm saying that
> > while his statement is accurate regarding an ipod with 100 tracks, it
> > is not applicable to one containing 19,029 tracks.
> > It does a lot of quirky little things with a frequency that I can only
> > attribute to synchronicity.  For example, playing a song from the
> > soundtrack of Mary Poppins precisely at the moment that I pass the
> > theater were the stage production of that show is playing.  Or playing
> > Give My Regards to Broadway as I'm walking through Herald Square.  Or
> > by playing a high percentage of country songs on the day I made the
> > "I'm a Little Bit Country" post.
>
> > These meaningful "random" occurrences open up the way for a discussion
> > on several related topics -- but I'm afraid they're all too
> > metaphysical for most members of this empirically oriented group.
>
> You should do a little reading on the phenomenon of Illusory
> Correlation.

I'm familiar with both sides of the argument. In the end, I believe
that "meaningful coincidences" are too numerous to be the mere product
of chance. Furthermore, the concept of Synchronicity is not only
consistent with, but seems like a natural consequence of, both a
Subjective and a Pantheistic universe.

> When coincidences occur, it is a normal human tendency to look for
> patterns or reasons in the chaos.  It does not mean they are really
> there.

True. The unconscious mind also communicates through symbols, so our
entire thought system is based upon finding/creating correspondences
between images, feelings, and, ultimately, thoughts.

However, this tendency would also be necessary for our creation of a
personal (subjective) universe -- especially one that corresponds in
all general particulars to those of every other (sane) member of our
species.

[Michael Pendragon]

On May 26, 3:36 pm, RWC wrote:
> On Fri, 25 May 2012 10:31:34 -0700 (PDT), Michael Pendragon
>

As always, Geoff, your database charts are fascinating (at least for
me).

One thing to note regarding repeats is that I have several copies of
many "greatest hits" songs, as these turn up frequently on cd
collections. Songs like "Sugartime" and "Moments to Remember" are
almost predestined to have multiple copies.

(Once I start running low on free space, I'll remove many of the
copies from the tracklist, but for now I'm leaving them stand. I
figure that it makes sense for a make-believe radio station to play
the big hits more frequently than the lesser known album cuts.)

Some songs, like Frankie Laine's "Cool Water" not only have multiple
copies, but multiple versions (as Mr. Laine recorded several versions
of most of his big hits). So I don't find it at all surprising when
my ipod plays this song.

What makes me think it follows patterns is when it repeats a single
track from an artist I've only got one CD by (like JLL or Syvia
Syms). I mean, if out of a total of 1,387 tracks, the ipod is going
to select 2 by Ms. Syms (who has a total of 15 tracks out of the
19,029 total), the odds of it picking *the same* song by her (or by
JLL, who's got a total of 19 tracks) makes me suspect that it repeats
roughly 1 out of every 20-odd (34.675 according to the stats from your
list) set.

[Bob Roman]

On May 26, 10:58 pm, Michael Pendragon

<michaelmaleficapendra...@gmail.com> wrote:

> I'm familiar with both sides of the argument.  In the end, I believe
> that "meaningful coincidences" are too numerous to be the mere product
> of chance.  Furthermore, the concept of Synchronicity is not only
> consistent with, but seems like a natural consequence of, both a
> Subjective and a Pantheistic universe.
>

> True.  The unconscious mind also communicates through symbols, so our
> entire thought system is based upon finding/creating correspondences
> between images, feelings, and, ultimately, thoughts.
>
> However, this tendency would also be necessary for our creation of a
> personal (subjective) universe -- especially one that corresponds in
> all general particulars to those of every other (sane) member of our
> species.
>
> True. The unconscious mind also communicates through symbols, so our
> entire thought system is based upon finding/creating correspondences
> between images, feelings, and, ultimately, thoughts.
>
> However, this tendency would also be necessary for our creation of a
> personal (subjective) universe -- especially one that corresponds in
> all general particulars to those of every other (sane) member of our
> species.

If everyone indulged such irresponsibly irrational nonsense as the
above we would still be living in the dark ages.

Good luck to you, though.

--
BR

[Bob Roman]

On May 26, 10:44 pm, Michael Pendragon

<michaelmaleficapendra...@gmail.com> wrote:

> I wouldn't think you were.  It was a preemptive measure designed to
> stop Bob from accusing me of thinking you were dishonest.

I never said you thought he was dishonest. I said that you did not
understand his point.

--
BR

[RWC]

On Sat, 26 May 2012 20:29:43 -0700 (PDT), Michael Pendragon
<michaelmalef...@gmail.com> wrote:

>As always, Geoff, your database charts are fascinating (at least for
>me).
>

Thanks Mike. If you're happy enough to accept the veracity of Billboard charts
(is there anything better available for the period 1945 to 1954?) I have a
feature coming up on the popularity of artists 1945 to 1949 - the top 60 -
according to *annual* (top 200 or 300 roughly speaking depending on the year)
Billboard charts - made available by an internet identity known as RomeBoy.

An important (to me) question :-)
I currently perceive the *golden* 'classic pop' era to be from 1945 to 1954.
Of course, styles fade in and fade out over several years - so there's rarely
clear-cut time boundaries for any musical style.

My current thinking goes like this
(which admittedly is partly [only partly] influenced by wishful thinking,
connected to the availability of Billboard information only from 1945 onwards
:-),
before 1945 (the swing band era as distinct from classic pop) the majority (over
50%) of 'hits' were by Orchestra's/Bands/Sextets etc - any vocalist was just
their current songster or songstress who, in terms of star billing status, was
not much more than just another team member of that Orchestra. But from 1945,
the majority of hits now had the vocalist with star billing, with the backing
orchestra being just that. So, in 1941 I presume the buying public were asking
their friends "Have you heard the latest Les Brown record?". But in 1945, again
I presume, it was more likely to be "Have you heard the latest Doris Day
record?"

At the other end of the era, perhaps 1954 was the last year that adult pop
provided the majority of commercial hits? Of course adult pop continued well
beyond 1954, with some great adult pop albums extending into the 60s, but it was
no longer the 'golden' era - if the criteria for 'golden' relates to relative
commercial success.

Now, perhaps this 1945? to 1954? era, as regards classic adult pop can be broken
down into two or more sub-eras. Let's start, perhaps provocatively, by
suggesting 1945-1949 and 1950-1954. Very briefly, 1945-1949 saw, in general, a
traditional laid-back but sometimes jazzy style, while 1950-1954 saw an
increasingly brassy rock-pop style of adult pop. I note Bing Crosby was clearly
falling away in this second period, while Frankie Laine was very much on the
rise :-)

I'm not expecting anyone to agree with the above, somewhat unsophisticated,
assumptions/descriptions. I'm looking to learn from folks opinions - if there
are any :-)

>Some songs, like Frankie Laine's "Cool Water" not only have multiple
>copies, but multiple versions (as Mr. Laine recorded several versions
>of most of his big hits). So I don't find it at all surprising when
>my ipod plays this song.
>

Noted, Mike.

>What makes me think it follows patterns is when it repeats a single
>track from an artist I've only got one CD by (like JLL or Syvia
>Syms).

I think a recent Bob Roman message briefly addressed the common phenomena of
folk seeing sus 'patterns' when a RNG did not agree with their intuitive
expectations. One would think that, for a standard (non special feature)
shuffle, Apple programmers would choose a unique (rarely if ever repeated) seed
for their psuedo-RNG, such as the time in milliseconds for your last scroll.


Geoff

[Michael Pendragon]

On May 27, 1:49 am, RWC wrote:
> On Sat, 26 May 2012 20:29:43 -0700 (PDT), Michael Pendragon
>

That's the basic gist of how I see things ... but, of course, I wasn't
there.

However, I don't think it was that cut and dried.

Many of the "sweet bands" (Russ Morgan, Guy Lombardo, Sammy Kaye)
continued to chart regularly in the late 40s and throughout the 50s
(Les Baxter, Nelson Riddle, Morris Stoloff). Then you had popular
albums by exotica and mood music by Baxter, Paul Weston, and Martin
Denny.

A lot of the films and tv shows from the post-war 40s and throughout
the 50s included bands -- as featured performers.

One of the most eye-opening examples for me was 1949's "Make Believe
Ballroom." The film's plot centered on a radio contest (presumably
fictional) run by popular west coast DJ Al Jarvis (playing himself).
Contestants have to be the first to call in and correctly identify
various "mystery" big bands -- the winner being the person who racks
up the most wins over the course of the contest (which I think ran for
several weeks). The callers are teenagers -- particularly those who
hang out and/or work at a local drive-in fast food joint. They seem
to be able to recognize not only the bandleaders, but many of the
individual musicians in each band after only hearing a few seconds of
the record. I don't know how realistic the film was in that regard,
but it featured many DJs, big bands/musicians, and pop vocalists as
themselves.

My guess would be that the situation was similar to that of the pop/
rock ratio of the mid-to-late 50s, in that the newer style (vocal
oriented pop) outnumbered older style (instrumentals) on the charts,
but that many instrumental hits continued to have a very strong
presence (often placing in the year's top 10). Again, similarly, the
big band hits of the early 50s had moved away from the (older) brash,
jazzy swing style for a more modern (that is, more pop-sounding) one.

So I'm seeing the switch in musical styles (insofar as the charts are
concerned) as a more gradual one, wherein the style of the previous
generation adapts itself to be more compatible with that of the
current generation; and that the two styles quickly meld into what
becomes the signature style of their era. This trend seems to still
be in effect today, although it's harder to track because there are so
many different styles involved.

> At the other end of the era, perhaps 1954 was the last year that adult pop
> provided the majority of commercial hits? Of course adult pop continued well
> beyond 1954, with some great adult pop albums extending into the 60s, but it was
> no longer the 'golden' era - if the criteria for 'golden' relates to relative
> commercial success.

I think that if you're going by chart success, you've got to extend it
to 1956 or 57. Rock 'n' Roll dominated the charts, but Pop still
maintained a very strong presence.

Taking other factors into account, I consider "The Golden Age" to be
from 1946-1964. First, because there were many great Pop albums
produced during this period, many of which made the album charts.
Limiting the era to the singles charts ignores the adult audience
portion of the market, and focuses solely on the tastes of the
teenagers. Second, many Pop stars continued to maintain popularity on
tv shows, in films, and at Las Vegas/concerts. Pop singers, for
instance, ruled Las Vegas until Elvis opened it up for rock 'n' roll
artists in 1969. Doris Day was the top box office female star of the
early-mid 60s, and Frank Sinatra and Dean Martin remained hugely
popular film stars as well. Judy Garland, Dean Martin, Mitch Miller,
Dinah Shore, Perry Como, and many others had popular tv shows and
specials throughout the 50s and well into the 60s. Third, many new
Pop singers like Andy Williams, Johnny Mathis, Anita Bryant, Leslie
Uggams, Steve Lawrence, Eydie Gorme, Shirley Bassey, Gogi Grant, Nancy
Wilson, et al., began their careers in the mid-late 50s or early 60s.
Fourth, many of the top r'n'r singers seemed to focus increasingly on
Pop (and/or heavily Pop-oriented music) after their first few hits/
were generally indistinguishable from Pop: Elvis Presley, Pat Boone,
Connie Francis, The Platters, Bobby Darin, Frankie Avalon, etc.

This isn't to deny the co-existence of "The Rock 'n' Roll Era" (which
I date from 1954-1964). I merely think that musical eras tend to
overlap. And, apart from the more extreme (that is heavily country or
r&b oriented) examples of 50s R'n'R (Early Elvis, Little Richard,
Jerry Lee Lewis), a large portion of the late 50s top 40 hits are Pop
equivalents of what Baxter/Nelson/Stoloff/Denny were to early 40s big
band. That is, they were R'n'R-compatible Pop songs.

> Now, perhaps this 1945? to 1954? era, as regards classic adult pop can be broken
> down into two or more sub-eras. Let's start, perhaps provocatively, by
> suggesting 1945-1949 and 1950-1954.  Very briefly, 1945-1949 saw, in general, a
> traditional laid-back but sometimes jazzy style, while 1950-1954 saw an
> increasingly brassy rock-pop style of adult pop.  I note Bing Crosby was clearly
> falling away in this second period, while Frankie Laine was very much on the
> rise :-)

Definitely. 1950 marks the rise of the "big-voiced" singers who
dominated the early 50s (Laine, Johnnie Ray, Eddie Fisher, Tony
Bennett, Mario Lanza, Teresa Brewer, Kay Starr, Georgia Gibbs, et.
al.). Doris Day, Jo Stafford, Patti Page, Joni James, Rosemary
Clooney and other ladies switched btw the two styles, or else
straddled a very thin line btw them; while Perry Como, Vic Damone,
Tony Martin, Jerry Vale and other male singers mixed a "big voiced"
approach with Frank Sinatra styled crooning (while Sinatra began
alternating crooning standards with swingin' versions of the same).

> I'm not expecting anyone to agree with the above, somewhat unsophisticated,
> assumptions/descriptions.  I'm looking to learn from folks opinions - if there
> are any :-)
>
> >Some songs, like Frankie Laine's "Cool Water" not only have multiple
> >copies, but multiple versions (as Mr. Laine recorded several versions
> >of most of his big hits).  So I don't find it at all surprising when
> >my ipod plays this song.
>
> Noted, Mike.
>
> >What makes me think it follows patterns is when it repeats a single
> >track from an artist I've only got one CD by (like JLL or Syvia
> >Syms).
>
> I think a recent Bob Roman message briefly addressed the common phenomena of
> folk seeing sus 'patterns' when a RNG did not agree with their intuitive
> expectations.  One would think that, for a standard (non special feature)
> shuffle, Apple programmers would choose a unique (rarely if ever repeated) seed
> for their psuedo-RNG, such as the time in milliseconds for your last scroll.

Not sure what you mean here.

[RWC]

On Sun, 27 May 2012 01:52:57 -0700 (PDT), Michael Pendragon
<michaelmalef...@gmail.com> wrote:


>That's the basic gist of how I see things ... but, of course, I wasn't
>there.
>

Nor was I, of course.

>However, I don't think it was that cut and dried.
>
>Many of the "sweet bands" (Russ Morgan, Guy Lombardo, Sammy Kaye)
>continued to chart regularly in the late 40s and throughout the 50s
>(Les Baxter, Nelson Riddle, Morris Stoloff).

Guy Lombardo (Mr New Year's Eve, who sold over 100 million records), for one,
did not change much, it seems, between 1942 and 1949. For instance, compare
"Beale Street Blues" v. Kenny Gardner from '42,
"It's Love, Love, Love", v. Skip Nelson and The Lombardo Trio from '44 and
"Everywhere You Go" v. Lombardo Trio from '49
(while just the year before some 1941 records, such as 'The Band Played On' and
'Intermezzo' are reminiscent of early 30s quaint - gee, what happened?).

>Then you had popular
>albums by exotica and mood music by Baxter, Paul Weston, and Martin
>Denny.
>

I'm a fan of 50s exotica.

>A lot of the films and tv shows from the post-war 40s and throughout
>the 50s included bands -- as featured performers.
>
>One of the most eye-opening examples for me was 1949's "Make Believe
>Ballroom." The film's plot centered on a radio contest (presumably
>fictional) run by popular west coast DJ Al Jarvis (playing himself).
>Contestants have to be the first to call in and correctly identify
>various "mystery" big bands -- the winner being the person who racks
>up the most wins over the course of the contest (which I think ran for
>several weeks). The callers are teenagers -- particularly those who
>hang out and/or work at a local drive-in fast food joint. They seem
>to be able to recognize not only the bandleaders, but many of the
>individual musicians in each band after only hearing a few seconds of
>the record. I don't know how realistic the film was in that regard,
>but it featured many DJs, big bands/musicians, and pop vocalists as
>themselves.

I'd like to see this film!

>
>My guess would be that the situation was similar to that of the pop/
>rock ratio of the mid-to-late 50s, in that the newer style (vocal
>oriented pop) outnumbered older style (instrumentals) on the charts,
>but that many instrumental hits continued to have a very strong
>presence (often placing in the year's top 10).

Across '45 to '49, the two bands with the most success based on chart
performance were the Vaughn Monroe Orchestra (6th place with 35 charting
records), followed by the Sammy Kaye Orchestra (8th with 29 records), then Guy
Lombardo (10th from 30 records), Freddy Martin (12th from 23 records), with four
more instrumental bands in the top 20.

>Again, similarly, the
>big band hits of the early 50s had moved away from the (older) brash,
>jazzy swing style for a more modern (that is, more pop-sounding) one.
>
>So I'm seeing the switch in musical styles (insofar as the charts are
>concerned) as a more gradual one, wherein the style of the previous
>generation adapts itself to be more compatible with that of the
>current generation; and that the two styles quickly meld into what
>becomes the signature style of their era. This trend seems to still
>be in effect today, although it's harder to track because there are so
>many different styles involved.
>
>> At the other end of the era, perhaps 1954 was the last year that adult pop
>> provided the majority of commercial hits? Of course adult pop continued well
>> beyond 1954, with some great adult pop albums extending into the 60s, but it was
>> no longer the 'golden' era - if the criteria for 'golden' relates to relative
>> commercial success.
>
>I think that if you're going by chart success, you've got to extend it
>to 1956 or 57. Rock 'n' Roll dominated the charts, but Pop still
>maintained a very strong presence.
>

In '56 at least 40% of Billboards top 100 seem to be records that I imagine
might have been purchased, or at least enjoyed, by adults 30+?
http://www.bobborst.com/popculture/top-100-songs-of-the-year/?year=1956

Many thanks for your response, Mike. It made for enjoyable and informative
reading. One reason for the sub-era question was to help me decide on a
breakdown for stats presentation, with the idea of seeing the change in
composition of successful artists and type of artist. Just trivia these days
perhaps, but idle curiosity prevails :) and I like much of this music.

>
>> >What makes me think it follows patterns is when it repeats a single
>> >track from an artist I've only got one CD by (like JLL or Syvia
>> >Syms).
>>
>> I think a recent Bob Roman message briefly addressed the common phenomena of
>> folk seeing sus 'patterns' when a RNG did not agree with their intuitive
>> expectations.  One would think that, for a standard (non special feature)
>> shuffle, Apple programmers would choose a unique (rarely if ever repeated) seed
>> for their psuedo-RNG, such as the time in milliseconds for your last scroll.
>
>Not sure what you mean here.

[Michael Pendragon]

On May 27, 9:56 am, RWC wrote:
> On Sun, 27 May 2012 01:52:57 -0700 (PDT), Michael Pendragon
>

> <michaelmaleficapendra...@gmail.com> wrote:
> >That's the basic gist of how I see things ... but, of course, I wasn't
> >there.
>
> Nor was I, of course.
>
> >However, I don't think it was that cut and dried.
>
> >Many of the "sweet bands" (Russ Morgan, Guy Lombardo, Sammy Kaye)
> >continued to chart regularly in the late 40s and throughout the 50s
> >(Les Baxter, Nelson Riddle, Morris Stoloff).
>
> Guy Lombardo (Mr New Year's Eve, who sold over 100 million records), for one,
> did not change much, it seems, between 1942 and 1949. For instance, compare
> "Beale Street Blues" v. Kenny Gardner from '42,
> "It's Love, Love, Love", v. Skip Nelson and The Lombardo Trio from '44 and
> "Everywhere You Go" v. Lombardo Trio from '49
> (while just the year before some 1941 records, such as 'The Band Played On' and
> 'Intermezzo' are reminiscent of early 30s quaint - gee, what happened?).

Guy Lombardo followed the same formula since the 1920s: "Music is easy
to play and easy to listen to if you don't forget the melody and
choose songs people can sing, hum, or whistle." His band was
something of an anomaly in that it always considered "old fashioned"
-- and never wanted to be anything else.

Mitch Miller followed a similar formula with his sing-along records --
with equal success.

Seems like people have a soft spot for good old fashioned tunes.

> >Then you had popular
> >albums by exotica and mood music by Baxter, Paul Weston, and Martin
> >Denny.
>
> I'm a fan of 50s exotica.

> >A lot of the films and tv shows from the post-war 40s and throughout
> >the 50s included bands -- as featured performers.
>
> >One of the most eye-opening examples for me was 1949's "Make Believe
> >Ballroom."  The film's plot centered on a radio contest (presumably
> >fictional) run by popular west coast DJ Al Jarvis (playing himself).
> >Contestants have to be the first to call in and correctly identify
> >various "mystery" big bands -- the winner being the person who racks
> >up the most wins over the course of the contest (which I think ran for
> >several weeks).  The callers are teenagers -- particularly those who
> >hang out and/or work at a local drive-in fast food joint.  They seem
> >to be able to recognize not only the bandleaders, but many of the
> >individual musicians in each band after only hearing a few seconds of
> >the record.  I don't know how realistic the film was in that regard,
> >but it featured many DJs, big bands/musicians, and pop vocalists as
> >themselves.
>
> I'd like to see this film!

As far as I know it's not out on DVD or VHS. I got to see it when a
friend sent me a VHS copy taped from (I think) a British tv airing.
There are a few low quality clips, featuring Frankie Laine, Kay Starr,
and Toni Harper on youtube:

http://www.youtube.com/watch?v=aeI-fpf1WK8
http://www.youtube.com/watch?v=CWx1DriReHc
http://www.youtube.com/watch?v=7iLxxZ1aj4I

> >My guess would be that the situation was similar to that of the pop/
> >rock ratio of the mid-to-late 50s, in that the newer style (vocal
> >oriented pop) outnumbered older style (instrumentals) on the charts,
> >but that many instrumental hits continued to have a very strong
> >presence (often placing in the year's top 10).
>
> Across '45 to '49, the two bands with the most success based on chart
> performance were the Vaughn Monroe Orchestra (6th place with 35 charting
> records), followed by the Sammy Kaye Orchestra (8th with 29 records), then Guy
> Lombardo (10th from 30 records), Freddy Martin (12th from 23 records), with four
> more instrumental bands in the top 20.

I count Vaughn Monroe as a singer unless someone else, like Ziggy
Talent, was handling the vocal.

Thank you, Geoff.

> One reason for the sub-era question was to help me decide on a
> breakdown for stats presentation, with the idea of seeing the change in
> composition of successful artists and type of artist. Just trivia these days
> perhaps, but idle curiosity prevails :) and I like much of this music.

I'm also interested in the sub-era categorization as I think the mid-
era shift from the older to newer sounding style is a key factor in
the development of each era's music. It's pretty straightforward
Darwinism: new musical form appears (upsets the existing status quo),
competes with and slightly out-paces the older musical form, until the
two "mate" (so to speak) -- each taking on characteristics of the
other to create a more homogenized/universal style.

> >> >What makes me think it follows patterns is when it repeats a single
> >> >track from an artist I've only got one CD by (like JLL or Syvia
> >> >Syms).
>
> >> I think a recent Bob Roman message briefly addressed the common phenomena of
> >> folk seeing sus 'patterns' when a RNG did not agree with their intuitive
> >> expectations.  One would think that, for a standard (non special feature)
> >> shuffle, Apple programmers would choose a unique (rarely if ever repeated) seed
> >> for their psuedo-RNG, such as the time in milliseconds for your last scroll.
>
> >Not sure what you mean here.
>
> "When coincidences occur, it is a normal human tendency to look for
> patterns or reasons in the chaos.  It does not mean they are really
> there."

Oh ... another realist.

[RWC]

On Sun, 27 May 2012 19:43:49 -0700 (PDT), Michael Pendragon
<michaelmalef...@gmail.com> wrote:

>I count Vaughn Monroe as a singer unless someone else, like Ziggy
>Talent, was handling the vocal.
>

Just in the slightest chance you have not come across the BigBandLibrary site by
Christopher Popa: http://www.bigbandlibrary.com/vaughnmonroe.html
http://www.bigbandlibrary.com/biographicalsketches.html

[Michael Pendragon]

On May 28, 10:16 am, RWC wrote:
> On Sun, 27 May 2012 19:43:49 -0700 (PDT), Michael Pendragon
>

> <michaelmaleficapendra...@gmail.com> wrote:
> >I count Vaughn Monroe as a singer unless someone else, like Ziggy
> >Talent, was handling the vocal.
>
> Just in the slightest chance you have not come across the BigBandLibrary site by
> Christopher Popa:  http://www.bigbandlibrary.com/vaughnmonroe.htmlhttp://www.bigbandlibrary.com/biographicalsketches.html

Thanks Len.

I have come across this site before, but the Vaughn Monroe page is
always worth a revisit.