What is the current trend in public interest for string theory?

[marcus]

Informed public interest in scientific research is part of what keeps the enterprise healthy, and merits some attention. As an index of public interest in string, expressed in the book market, I have been tracking the "string Dow"

This is the average at noon (mountain time) of the standings of the five most popular string books on the Amazon physics bestseller list. I have recorded this each day and have averaged the series over several 12-day intervals.

WHAT DO YOU PREDICT THE AVERAGE WILL BE FOR THE MONTH OF JANUARY?
Here's the record.
"String Dow" averaged over successive twelve day intervals:

17-31 oct average 20.5
01-13 nov average 20.4
14-25 nov average 23.8
26 nov-07 dec average 25.4
08-19 dec average 15.6
20-31 dec average 20.3"String Dow" so far in January:

01 jan Monday noon* 19.8
02 jan Tuesday noon 23.0
03 jan Wednesday noon 24.4
04 jan Thursday noon 31.4
05 jan Friday noon 35.8
06 jan Saturday noon 52.4

*mountain time

If the string Dow index stays in the range 15-25, I would call that STEADY popularity. But if the average rank goes below 15, that would signify a big GAIN in public interest in string. On the other hand, going over 25 would seem to indicate a loss of interest, at least if the shift is sustained over time.

 

[marcus]

Here is the list:
https://www.amazon.com/gp/bestsellers/books/14545/ref=pd_ts_b_ldr/102-1454980-4598542&tag=pfamazon01-20

The list is 100 titles long, and in my experience when I check at noon there are always 5 or more stringy books on it. So I just take the first 5 that show up and average their ranks.

If it should turn out some time that the fifth book is not in the top 100, I would arbitrarily assign it rank 100, and put a + sign by the average, but hopefully that will not be necessary.

The five most popular stringy books are usually two by Greene, one or two by Kaku, and one or two by Randall. Sometimes a third Greene book (the hardbound edition of one of the other two) will show up. Sometimes Susskind's book is among the five most popular. I don't count the Smolin and Woit books as stringy since although they discuss string research they are more on the critical side.

[EDIT, 6 JAN] If you go to the list right at the moment (Saturday 12:01 PM mountain) you will find the sum is 262, so the index is 52.4
09 Greene elegant
29 Greene fabric
66 Kaku parallel
67 Randall warped (hardbound)
91 Randall warped

This is in popularity "nosedive" territory. Could be just a temporary excursion. It this shift holds then 6 January 2007 could be remembered as "The Day the Public Stopped Buying String" :-) or some such. The change is remarkable. In mid-December this index was often 10 or even less (after Lisa Randall's interview on Charlie Rose TV) and now it is up in the 40s and 50s.

[EDIT, 6 JAN] just as a check, I also looked at 7:05 PM mountain today, and found the sum 260, so the index at that hour was 52.0.
14 Greene fabric
15 Greene elegant
51 Kaku parallel
80 Randall warped
100 Greene fabric (hardbound)

 

[marcus]

"String Dow" so far in January:

01 jan Monday noon* 19.8
02 jan Tuesday noon 23.0
03 jan Wednesday noon 24.4
04 jan Thursday noon 31.4
05 jan Friday noon 35.8
06 jan Saturday noon 52.4
07 jan Sunday noon 18.8
08 jan Monday noon 36.6

*mountain time

Tracking this index is a bumpy ride.
Sunday it was back in the "steady public interest" range, but on Monday it had returned to "nosedive".
Hard to guess what the average for January will be. In a couple of days I will post an average for the first ten days of the month.

 

[marcus]

Correction, the standings changed right at the hour, so the figure for Monday noon is 42.4 instead of what I had written at first.

"String Dow" so far in January:

01 jan Monday noon* 19.8
02 jan Tuesday noon 23.0
03 jan Wednesday noon 24.4
04 jan Thursday noon 31.4
05 jan Friday noon 35.8
06 jan Saturday noon 52.4
07 jan Sunday noon 18.8
08 jan Monday noon 42.4

*mountain time

the running total for the past eight days, in case anyone is curious, is 248 making the average 31.

 

[william donnelly]

Wouldn't it make more sense to add up the sales of the books?

At the very least I would take a harmonic mean. Otherwise if a book goes from 40->42 (no big deal) this is a more significant event than a book going from 2->1 (a very big deal).

 

[marcus]

Wouldn't it make more sense to add up the sales of the books?

At the very least I would take a harmonic mean. Otherwise if a book goes from 40->42 (no big deal) this is a more significant event than a book going from 2->1 (a very big deal).

One can make arguments either way, William. As it happens my DATA from 17 October up to present consists of arithmetic averages. I couldn't go back and convert them to harmonic means even if I wanted to!
So those are the observations that we are working with, and predicting.

Notice that in the October-December period where i have observations, the index stayed fairly stable in the range 15 - 25, mostly around 20.

This is even though the set of books changed considerably (it differs from day to day which the top five stringy books are) and the standings of individual books jumped around quite a bit. The fact that you are always taking the 5 lowest numbers, to average, tends to stabilize things quite a bit.

what one is keeping track of, basically, is the footprint in the bestsell list of WHATEVER are the five most popular stringy books at the moment. One watches where that footprint is in the context of the whole list of 100 bestsellers.

One wants an arithmetic mean because one wants to see if the center of gravity of the most popular five shifts any.

Oh BTW, numbers of copies sold is not information that the publishers make available (except in certain cases, after the fact). Since we don't know numbers sold, we can keep track of position in the bestsell list.

Thanks for your comment I am glad to see that someone is thinking about this!

 

[marcus]

an illustration

William I should show you by example what I mean.

Prompted by your post, I just checked the amazon physics list at 3:15 mountain time (8 January 2007) and found the most popular five stringy books had these ranks

39 Greene fabric
43 Greene elegant
46 Randall warped (hardbound)
52 Kaku parallel
89 Randall warped

keep in mind that for a considerable part of last month the average standing was around TEN which is to say that many of the standings were one digit numbers. What we see is that the footprint of this bunch of titles has SHIFTED, from around 10 or 20 to (at least for the moment) around 40 or 50.

since publishers do not give us "number of copies sold" information we are not measuring that or trying to predict it. We are looking at where is the center of gravity of the bunch of most popular string books and we are looking to see if that SHIFTS. That is why we should be using a LINEAR AVERAGE as I am using.

I would be delighted if you would try your hand at forecasting the average for January

the observation just made leads to a sum of 269 and an average standing (the "String Dow" index) of 53.8
but that is a 3 PM reading. I only record one reading each day, at noon (mountain time.)
In case anyone is curious and wants to check out the index at other times of day, here again is the link to the physics bestseller list:
https://www.amazon.com/gp/bestsellers/books/14545/ref=pd_ts_b_ldr/102-1454980-4598542&tag=pfamazon01-20

===EDIT (UPDATE)===

as of 4:30 PM mountain time the String Dow was 65.0
today could be a candidate for "The Day the Public Stopped Buying String"
the contrast with just a month ago is remarkable and things seem to be happening fast.
Perhaps we will see a swingback tomorrow.

When some book's standing gets to be more than 100, I just truncate and call it 100+. (by looking at an alternative list one can work it out but it's unwanted bother). The standings at 4:30 of the five most popular were
48 elegant
52 fabric
53 warped (hard)
72 parallel
100+ warped
the sum being 325 and the average standing 65.0+
===================

christo, welcome to this little forecasting game. The object is to predict the January average objectively. I assume you aren't letting your personal opinion of string researchers and whether or not you think they have good prospects for making progress influence your judgement about what the book-buying public is going to do. These are separate issues. Thanks for contributing your prediction!

It is pretty dynamic isn't it? I have really no idea how the month is going to turn out...or even the next two days!

Except for the fact that Smolin's book "The Trouble with Physics" continues to have a fairly high standing on the list, I have no inkling as to what is driving these changes in book-buyer behavior.
=========UPDATE=======
I checked again at 10:05 am mountain time this morning Tuesday 9 Jan. The index was still in the "nosedive" range. It is quite strange.

30 elegant
45 fabric
46 warped (hard)
80 parallel
84 warped
sum is 285 and average is 57.0

These numbers are way off from what it was like last month, or at any time last fall. It really does begin to seem as if the public has, at least momentarily, stopped buying string pop literature.

======THE NOON DOW======
Today at noon mountain time (Tuesday 9 January) it turned out to be 55.8 so the January record is

01 jan Monday noon 19.8
02 jan Tuesday noon 23.0
03 jan Wednesday noon 24.4
04 jan Thursday noon 31.4
05 jan Friday noon 35.8
06 jan Saturday noon 52.4
07 jan Sunday noon 18.8
08 jan Monday noon 42.4
09 jan Tuesday noon 55.8

the running total for the past nine days, in case anyone is curious, is 303.8 making the nine-day average 33.8.

 

[Chronos]

It's all about the $$ in my mind. The masses vote their pocket book. How do the numbers stack up by that yardstick? I think ST is dangling like the sword of damocles these days . . . an ironic analogy if ever there was one.

 

[marcus]

Chronos, you couldn't be talking about that sword suspended by a piece of string now, could you?

In case anyone is curious and wants to check out the index for themselves, here again is the link to the physics bestseller list:
https://www.amazon.com/gp/bestsellers/books/14545/ref=pd_ts_b_ldr/102-1454980-4598542&tag=pfamazon01-20

======UPDATE ON THE NOON DOW======
Today at noon mountain time (Wednesday 10 January) it turned out to be 64.4 so the January record is now

01 jan Monday noon 19.8
02 jan Tuesday noon 23.0
03 jan Wednesday noon 24.4
04 jan Thursday noon 31.4
05 jan Friday noon 35.8
06 jan Saturday noon 52.4
07 jan Sunday noon 18.8
08 jan Monday noon 42.4
09 jan Tuesday noon 55.8
10 jan Wednesday noon 64.4

the running total for the past ten days, in case anyone is curious, is 368.2 making the ten-day average 36.8.

11 jan Thursday noon 68.6

 

[marcus]

Wouldn't it make more sense to add up the sales of the books?
...

constructive suggestion. I would do it if the information was available. Searched several months ago for booksales figures. closely guarded by the publishers

however Amazon does provide an hourly OVERALL SALES RANK for each book, among all the books it sells of all types. We might get some additional information by averaging THAT for the five most popular stringy books.
I will give this a try! We will need to watch it for a few days to get an idea of benchmark-size and range of variation.

average sales rank at noon mountain time:
11 jan Thursday noon 5756
12 jan Friday noon 5685

======update on String Dow========

17 oct - 31 dec average 21.0
01 - 10 jan average 36.8

11 jan Thursday noon 68.6
12 jan Friday noon 53.0

 

[marcus]

======update on String Dow========
the "string Dow" is the average standing of the five most popular string books in the physics bestseller list----it is like the sales rank but within the restricted category of physics:

17 oct - 31 dec average 21.0
01 - 10 jan average 36.8

11 jan Thursday noon 68.6
12 jan Friday noon 53.0
13 jan Saturday noon 44.2
14 jan Sunday noon 36.8

the challenge was to predict the average for the month of January, and in particular to guess which range it will be in out of these four alternatives:
under 15
15-25
25-35
over 35.

===update on SP (string popularity) Index====
this is the average overall sales rank (among all books sold) of the five most popular stringy books
recorded at noon mountain time:

11 jan Thursday noon 5756
12 jan Friday noon 5685
13 jan Saturday noon 4203
14 jan Sunday noon 4204

 

[marcus]

======update on String Dow========
the "string Dow" is the average standing of the five most popular string books in the physics bestseller list----it is like the sales rank but within the restricted category of physics:

17 oct - 31 dec average 21.0
01 - 10 jan average 36.8

11 jan Thursday noon 68.6
12 jan Friday noon 53.0
13 jan Saturday noon 44.2
14 jan Sunday noon 36.8
15 jan Monday noon 28.4
16 jan Tuesday noon 47.6
17 jan Wedneday noon 61.4

the challenge was to predict the average for the month of January, and in particular to guess which range it will be in out of these four alternatives:
under 15
15-25
25-35
over 35.

===update on SP (string popularity) Index====
this is the average overall sales rank (among all books sold) of the five most popular stringy books---here compared with the sales rank of Smolin's TWP.
All ranks recorded at noon mountain time:

11 jan Thursday noon 5756 (1604) ratio 3.6
12 jan Friday noon 5685 (2159) ratio 2.6
13 jan Saturday noon 4203 (936) ratio 4.5
14 jan Sunday noon 4204 (1070) ratio 3.9
15 jan Monday noon 3497 (1231) ratio 2.8
16 jan Tuesday noon 5487.8 (530) ratio 10.4
17 jan Wednesday noon 6076.0 (872) ratio 7.0

 

[Chronos]

It's like a desert here, marcus. Where did everyone go?

 

[marcus]

Hi Chronos,
posting may be light for the moment but I think people are checking in and reading threads as usual. They'll be posting again too, as prolifically as ever, I expect :-)

We are in a kind of revolutionary situation as regards fundamental physics, I think. You, as another watcher, probably have ideas about this. We could try to characterize the shifts in perception going on (among string theorists, in the broader physics community, among science journalists and even possibly the public. It is a complex of changes, not quick to describe.)

But that is more something to do in the long run, perhaps. You have just reminded me that I haven't updated this record for a couple of days!

======update on String Dow========
the "string Dow" is the average standing of the five most popular string books in the physics bestseller list----it is an average sales rank but within the restricted category of physics:

17 oct - 31 dec average 21.0
01 - 10 jan average 36.8

11 jan Thursday noon 68.6
12 jan Friday noon 53.0
13 jan Saturday noon 44.2
14 jan Sunday noon 36.8
15 jan Monday noon 28.4
16 jan Tuesday noon 47.6
17 jan Wedneday noon 61.4
18 jan Thursday noon 63.6
19 jan Friday noon 52.2
20 jan Saturday noon 49.4

the challenge was to predict the average for the month of January, and in particular to guess which range it will be in out of these four alternatives:
under 15
15-25
25-35
over 35.

The running total for the first ten days was 368.2, so we ought to be able to total up for the first 19 or 20 days and get a provisional average to see how it's going. Actually you may have the best guess, Chronos, judging by how it looks at the moment. We'll see. Hmmm. I get that the total for the first 20 days of this month is 873.4 and the running average is 43.7.

this is incredibly different from what it used to be up until recently. My impression was before October 2006 it was typically in the teens, and then when I started to make regular observations at the same time every day it was normally around 20. So 20 is my baseline and something like 43 is way way off.
I guess it is in "nosedive" range :-)

But the month isn't over yet and the next 10 readings can pull the average back some. And we can always check in February.

 

[marcus]

In case anyone wants to study this in detail, here is an expanded record:
this is an index of public interest I call the "Stringy Dow Jones"---the average standings of the 5 most popular stringy books on the Amazon physics bestseller list
https://www.amazon.com/gp/bestsellers/books/14545/ref=pd_ts_b_ldr/102-1454980-4598542&tag=pfamazon01-20

My impression from sporadic samples in September and early October is that it was typically in the teens. In mid October I began recording noon readings, and averaging them over successive twelve day intervals:

17-31 oct average 20.5
01-13 nov average 20.4
14-25 nov average 23.8
26 nov-07 dec average 25.4
08-19 dec average 15.6
20-31 dec average 20.3

The average for the entire period (mid October thru December) was 21.0.

After that there seemed to be an abrupt change.
The average for 01-20 January is 43.7 which breaks down into
successive ten day intervals as follows:
01-10 jan average 36.8
11-20 jan average 50.5

In case anyone is interested in still further detail here is "String Dow" so far in January:

01 jan Monday noon* 19.8
02 jan Tuesday noon 23.0
03 jan Wednesday noon 24.4
04 jan Thursday noon 31.4
05 jan Friday noon 35.8
06 jan Saturday noon 52.4
07 jan Sunday noon 18.8
08 jan Monday noon 42.4
09 jan Tuesday noon 55.8
10 jan Wednesday noon 64.4

ten-day running total 368.2 and ten-day avg. 36.8

11 jan Thursday noon 68.6
12 jan Friday noon 53.0
13 jan Saturday noon 44.2
14 jan Sunday noon 36.8
15 jan Monday noon 28.4
16 jan Tuesday noon 47.6
17 jan Wednesday noon 61.4
18 jan Thursday noon 63.6
19 jan Friday noon 52.2
20 jan Saturday noon 49.4

ten-day total 505.2 and average 50.5

21 jan Sunday noon 37.6
22 jan Monday noon 44.2
23 jan Tuesday noon 54.0
24 jan Wednesday noon 58.6

*noon mountain time.

 

[marcus]

We have a winner!

In case anyone is interested in still further detail here is "String Dow" so far in January:

01 jan Monday noon* 19.8
02 jan Tuesday noon 23.0
03 jan Wednesday noon 24.4
04 jan Thursday noon 31.4
05 jan Friday noon 35.8
06 jan Saturday noon 52.4
07 jan Sunday noon 18.8
08 jan Monday noon 42.4
09 jan Tuesday noon 55.8
10 jan Wednesday noon 64.4

ten-day running total 368.2 and ten-day avg. 36.8

11 jan Thursday noon 68.6
12 jan Friday noon 53.0
13 jan Saturday noon 44.2
14 jan Sunday noon 36.8
15 jan Monday noon 28.4
16 jan Tuesday noon 47.6
17 jan Wednesday noon 61.4
18 jan Thursday noon 63.6
19 jan Friday noon 52.2
20 jan Saturday noon 49.4

ten-day total 505.2 and average 50.5

ten-day total 505.2 and average 50.5

21 jan Sunday noon 37.6
22 jan Monday noon 44.2
23 jan Tuesday noon 54.0
24 jan Wednesday noon 58.6
25 jan Thursday noon 42.6
26 jan Friday noon 59.0
27 jan Saturday noon 41.8
28 jan Sunday noon 34.4
29 jan Monday noon 30.0
30 jan Tuesday noon 39.6
31 jan Wednesday noon 31.0

eleven-day total 472.8 average 43.0

total for month 1346.2 average 43.4
*noon mountain time.

Summary from mid-October to end January.
The "Stringy Dow Jones"---the average standings of the 5 most popular string books on the Amazon physics bestseller list
https://www.amazon.com/gp/bestsellers/books/14545/ref=pd_ts_b_ldr/102-1454980-4598542&tag=pfamazon01-20

Judging from sporadic samples this was typically in the TEENS during September and early October 2006. In mid October I began recording the index at noon each day, and averaging over longer intervals:

17-31 oct average 20.5
01-13 nov average 20.4
14-25 nov average 23.8
26 nov-07 dec average 25.4
08-19 dec average 15.6
20-31 dec average 20.3

01-10 jan average 36.8
11-20 jan average 50.5
21-31 jan average 43.0

The average for the entire period (mid October thru December) was 21.0, while for the month of January the average was 43.4.

And the winner is Chronos by a mile.

String pop books did indeed take a nosedive in sales during January, gauging from their standing on the bestseller list. The average stringy book was about HALF AS POPULAR an item in January as it was, say, in October. I would not have predicted such a sharp change. In fact I didn't, I predicted some more moderate decline in popularity, if I remember right. Congratulations Chronos. :-) Let's see whether the change is permanent or whether it bounces back in February.

 

[Chronos]

A lucky guess based on intuition. Only 4 people even bothered to vote, so that puts me in the top 25% of the class - barely qualified to get admitted to a red state university. ST has failed to deliver on its glittery promises for over twenty years. It is a failed policy. No weapons of mass creation have been discovered.