Indeed, the number Joe was counting was the number who filled out a registration form. Counting those who actually paid their registration yields closer numbers.
rbarnes$ for n in $(jot 15 73); do
att=$(curl -s "https://www.ietf.org/registration/ietf${n}/attendance.py" | grep -o ">Yes<" | wc -l);
echo $n $att;
done
73 969
74 1170
75 1102
76 1129
77 1242
78 1159
79 1144
80 1231
81 1127
82 948
83 1395
84 1199
85 1157
86 1115
87 1435
On Tue, Oct 8, 2013 at 8:51 AM, Adrian Farrel <adrian@xxxxxxxxxxxx> wrote:
Curiously these numbers do not match those at
https://www.ietf.org/meeting/past.html
Registration, we may conclude, does not equate to attendance.
Adrian
> -----Original Message-----
> From: ietf-bounces@xxxxxxxx [mailto:ietf-bounces@xxxxxxxx] On Behalf Of Joe
> Abley
> Sent: 08 October 2013 02:38
> To: Ted Lemon
> Cc: diversity@xxxxxxxx; IETF
> Subject: Re: year for highest number of IETF participants
>
> [krill:~]% for n in $(jot 15 73); do
> curl -s "https://www.ietf.org/registration/ietf${n}/attendance.py" | \
> awk -v n=${n} '/ registrations:/ { sub(/ registrations:.*$/, "");
sub(/^.*\>/, "");
> print n, $0; }'
> done
> 73 1111
> 74 1332
> 75 1230
> 76 1249
> 77 1350
> 78 1304
> 79 1337
> 80 1317
> 81 1244
> 82 1051
> 83 1529
> 84 1356
> 85 1351
> 86 1223
> 87 1585
> [krill:~]%