Cryptography-Digest Digest #348, Volume #9 Tue, 6 Apr 99 20:13:04 EDT
Contents:
Histogram of my RC4 cipher stream (Anonymous)
Perl calls PGP ([EMAIL PROTECTED])
Geometric Identification ([EMAIL PROTECTED])
SFS, recovering a backup
Re: Live from the Second AES Conference (Ian Goldberg)
Calculation of CVV's ("Dave Bergert")
Re: True Randomness & The Law Of Large Numbers (R. Knauer)
Re: True Randomness & The Law Of Large Numbers (R. Knauer)
Re: True Randomness & The Law Of Large Numbers (R. Knauer)
Re: True Randomness & The Law Of Large Numbers (R. Knauer)
----------------------------------------------------------------------------
Date: Tue, 6 Apr 1999 23:34:13 +0200 (CEST)
From: Anonymous <[EMAIL PROTECTED]>
Subject: Histogram of my RC4 cipher stream
Statistical analysis such as this proves little as to a cipher's strength,
but in response to the other fellow's postings, this is the histograms
I've found from several test runs of my RC4 cipher code. I don't see any
immediately-apparent bias towards generating zeros.
K
Histograms of RC4 cipher streams. No source plaintext was included in
this; it is purely the output of the RC4 "engine."
Keys generated randomly, from 30 to 50 chars long, each character of key
could be ANY of 256 byte values. (Note, this CAN create weak keys!!!)
First 256 bytes of cipher stream discarded.
Next 1024*1024 (1mb) of cipher stream created and tallied below.
Test run #1 histogram.
0 - 4047
1 - 4088
2 - 4058
3 - 4167
4 - 4048
5 - 4196
6 - 4209
7 - 4043
8 - 4053
9 - 4082
10 - 4091
11 - 3973
12 - 4060
13 - 4150
14 - 4170
15 - 4148
16 - 4075
17 - 4152
18 - 4222
19 - 4047
20 - 4075
21 - 4084
22 - 4156
23 - 4058
24 - 4129
25 - 4134
26 - 4140
27 - 4071
28 - 4160
29 - 4048
30 - 4076
31 - 4153
32 - 4094
33 - 4139
34 - 4034
35 - 4166
36 - 3907
37 - 4138
38 - 4126
39 - 4257
40 - 4159
41 - 4163
42 - 4062
43 - 4069
44 - 4163
45 - 4072
46 - 4059
47 - 4306
48 - 4171
49 - 4108
50 - 4170
51 - 4032
52 - 3989
53 - 4010
54 - 4031
55 - 4109
56 - 4073
57 - 4080
58 - 4078
59 - 4149
60 - 4055
61 - 3982
62 - 4130
63 - 4032
64 - 4139
65 - 4109
66 - 4064
67 - 4206
68 - 4030
69 - 4153
70 - 4190
71 - 4167
72 - 4065
73 - 4114
74 - 4213
75 - 4056
76 - 4138
77 - 4236
78 - 4115
79 - 4108
80 - 4216
81 - 4108
82 - 4061
83 - 4102
84 - 4128
85 - 4042
86 - 4098
87 - 4063
88 - 4069
89 - 4091
90 - 4076
91 - 4047
92 - 4142
93 - 4102
94 - 4028
95 - 4161
96 - 4138
97 - 4123
98 - 4022
99 - 4037
100 - 4053
101 - 4138
102 - 3995
103 - 4011
104 - 4105
105 - 4154
106 - 4140
107 - 4039
108 - 4195
109 - 4127
110 - 4109
111 - 4047
112 - 4193
113 - 4038
114 - 4224
115 - 4124
116 - 4223
117 - 4063
118 - 4089
119 - 4168
120 - 4027
121 - 4031
122 - 4096
123 - 4001
124 - 4172
125 - 4018
126 - 3960
127 - 4078
128 - 4120
129 - 4161
130 - 4145
131 - 4077
132 - 4101
133 - 4175
134 - 4142
135 - 4106
136 - 4085
137 - 4197
138 - 4053
139 - 4150
140 - 4062
141 - 4092
142 - 4121
143 - 4076
144 - 4142
145 - 4100
146 - 4116
147 - 4083
148 - 4194
149 - 4018
150 - 4038
151 - 4048
152 - 4082
153 - 4159
154 - 4070
155 - 4023
156 - 4055
157 - 4001
158 - 4037
159 - 4022
160 - 4025
161 - 4124
162 - 3963
163 - 4087
164 - 4109
165 - 4039
166 - 4047
167 - 4127
168 - 4082
169 - 4077
170 - 4124
171 - 4050
172 - 4313
173 - 3989
174 - 3978
175 - 4025
176 - 4143
177 - 4115
178 - 3965
179 - 4106
180 - 4146
181 - 4057
182 - 4123
183 - 4220
184 - 4014
185 - 4011
186 - 4037
187 - 4088
188 - 4141
189 - 4046
190 - 4093
191 - 4147
192 - 4136
193 - 4099
194 - 4063
195 - 4120
196 - 4048
197 - 4072
198 - 4080
199 - 3968
200 - 4098
201 - 4152
202 - 4043
203 - 3995
204 - 4218
205 - 4010
206 - 4009
207 - 4129
208 - 4121
209 - 4023
210 - 4163
211 - 4065
212 - 4161
213 - 4069
214 - 4064
215 - 4175
216 - 4157
217 - 4026
218 - 4196
219 - 4097
220 - 4141
221 - 4089
222 - 4035
223 - 4147
224 - 4152
225 - 4081
226 - 4115
227 - 4013
228 - 4063
229 - 4174
230 - 4101
231 - 4108
232 - 4091
233 - 4171
234 - 4021
235 - 4010
236 - 4238
237 - 4142
238 - 4040
239 - 4036
240 - 4086
241 - 4123
242 - 4007
243 - 3943
244 - 4157
245 - 4067
246 - 4147
247 - 4125
248 - 3970
249 - 4115
250 - 4144
251 - 4089
252 - 4080
253 - 4064
254 - 4113
255 - 4108
Test run #2 histogram
0 - 3968
1 - 4156
2 - 4113
3 - 4024
4 - 4025
5 - 3964
6 - 3973
7 - 4108
8 - 4190
9 - 4058
10 - 4015
11 - 4188
12 - 4139
13 - 4209
14 - 4096
15 - 3985
16 - 4194
17 - 4149
18 - 4202
19 - 4091
20 - 4132
21 - 4103
22 - 4047
23 - 4132
24 - 4114
25 - 4076
26 - 4055
27 - 4126
28 - 4278
29 - 4066
30 - 4006
31 - 4144
32 - 4219
33 - 4061
34 - 4089
35 - 4002
36 - 4058
37 - 4175
38 - 4121
39 - 4103
40 - 4155
41 - 4027
42 - 3971
43 - 4102
44 - 4153
45 - 3981
46 - 4056
47 - 4108
48 - 4068
49 - 4123
50 - 3947
51 - 4106
52 - 4144
53 - 4102
54 - 3971
55 - 4003
56 - 3959
57 - 3991
58 - 4121
59 - 4068
60 - 4137
61 - 4195
62 - 4117
63 - 4015
64 - 4049
65 - 4064
66 - 4090
67 - 4020
68 - 4080
69 - 4032
70 - 4014
71 - 4105
72 - 4189
73 - 4025
74 - 4123
75 - 4154
76 - 4077
77 - 4162
78 - 4139
79 - 4165
80 - 4058
81 - 4280
82 - 4238
83 - 4084
84 - 4045
85 - 4079
86 - 4205
87 - 4090
88 - 4119
89 - 4142
90 - 4055
91 - 4101
92 - 4010
93 - 4077
94 - 4139
95 - 4175
96 - 4160
97 - 4068
98 - 4067
99 - 4084
100 - 4107
101 - 4097
102 - 4001
103 - 3973
104 - 4002
105 - 4111
106 - 4130
107 - 4063
108 - 4128
109 - 4092
110 - 4146
111 - 4076
112 - 4128
113 - 4115
114 - 4048
115 - 4156
116 - 4078
117 - 4121
118 - 4081
119 - 4137
120 - 4108
121 - 4062
122 - 4154
123 - 4130
124 - 4083
125 - 4077
126 - 4054
127 - 4020
128 - 4105
129 - 4023
130 - 4134
131 - 3977
132 - 4189
133 - 4008
134 - 4113
135 - 4035
136 - 4132
137 - 4182
138 - 4075
139 - 4179
140 - 4158
141 - 4050
142 - 4044
143 - 4072
144 - 4100
145 - 4112
146 - 4050
147 - 4133
148 - 4036
149 - 4170
150 - 4111
151 - 4023
152 - 4078
153 - 4147
154 - 4044
155 - 4190
156 - 4055
157 - 4161
158 - 4149
159 - 4085
160 - 4067
161 - 4113
162 - 4176
163 - 3994
164 - 4114
165 - 4071
166 - 4152
167 - 3949
168 - 3999
169 - 4010
170 - 4027
171 - 4098
172 - 4090
173 - 4133
174 - 4060
175 - 3929
176 - 4115
177 - 4032
178 - 4126
179 - 4095
180 - 4157
181 - 4145
182 - 4034
183 - 4174
184 - 4028
185 - 4022
186 - 4140
187 - 4254
188 - 4132
189 - 3934
190 - 4119
191 - 4118
192 - 4054
193 - 4200
194 - 4108
195 - 4135
196 - 4100
197 - 4166
198 - 4171
199 - 4119
200 - 4186
201 - 4091
202 - 4199
203 - 4117
204 - 4066
205 - 4084
206 - 4098
207 - 4083
208 - 4134
209 - 4184
210 - 4081
211 - 3972
212 - 4058
213 - 4264
214 - 4123
215 - 4070
216 - 4033
217 - 4095
218 - 4083
219 - 4157
220 - 4171
221 - 4197
222 - 4055
223 - 4026
224 - 4191
225 - 4054
226 - 4159
227 - 4082
228 - 4105
229 - 4048
230 - 4185
231 - 4107
232 - 4086
233 - 4081
234 - 4137
235 - 4025
236 - 4003
237 - 4079
238 - 4133
239 - 4100
240 - 4146
241 - 4084
242 - 4198
243 - 4067
244 - 4101
245 - 4209
246 - 4008
247 - 4114
248 - 4079
249 - 4176
250 - 4123
251 - 4007
252 - 4147
253 - 4122
254 - 4135
255 - 4062
Test run #3 histogram
0 - 4052
1 - 4171
2 - 4059
3 - 4126
4 - 4088
5 - 4176
6 - 3973
7 - 4054
8 - 4157
9 - 4051
10 - 4141
11 - 4088
12 - 4110
13 - 4075
14 - 4114
15 - 4169
16 - 4144
17 - 4038
18 - 4129
19 - 4139
20 - 4057
21 - 4147
22 - 4122
23 - 4125
24 - 4035
25 - 4234
26 - 4095
27 - 4105
28 - 4054
29 - 4175
30 - 4107
31 - 4104
32 - 3944
33 - 4138
34 - 4042
35 - 4142
36 - 4060
37 - 4195
38 - 4037
39 - 4050
40 - 4020
41 - 4224
42 - 4004
43 - 4096
44 - 4033
45 - 4037
46 - 4056
47 - 4048
48 - 4190
49 - 4041
50 - 4182
51 - 4147
52 - 4142
53 - 4048
54 - 4117
55 - 4062
56 - 4158
57 - 4034
58 - 4082
59 - 4035
60 - 4037
61 - 4016
62 - 4171
63 - 4154
64 - 4141
65 - 3996
66 - 4003
67 - 4059
68 - 4114
69 - 4098
70 - 4149
71 - 4148
72 - 4143
73 - 4027
74 - 4110
75 - 4004
76 - 4192
77 - 4028
78 - 4112
79 - 4168
80 - 4071
81 - 4151
82 - 4134
83 - 4086
84 - 4071
85 - 4075
86 - 4057
87 - 4092
88 - 4116
89 - 4145
90 - 4011
91 - 4024
92 - 4003
93 - 3990
94 - 4113
95 - 4169
96 - 4023
97 - 4019
98 - 4045
99 - 4074
100 - 4193
101 - 4092
102 - 3997
103 - 4141
104 - 4023
105 - 4085
106 - 4089
107 - 4184
108 - 4141
109 - 4051
110 - 4149
111 - 4036
112 - 4030
113 - 4025
114 - 4071
115 - 4182
116 - 3982
117 - 4109
118 - 4190
119 - 4199
120 - 4122
121 - 4001
122 - 4063
123 - 4155
124 - 4041
125 - 4034
126 - 4045
127 - 4094
128 - 4127
129 - 4086
130 - 4045
131 - 4099
132 - 4025
133 - 4148
134 - 4103
135 - 4144
136 - 4025
137 - 4126
138 - 4039
139 - 4118
140 - 4128
141 - 4083
142 - 4096
143 - 4166
144 - 4129
145 - 4175
146 - 4218
147 - 4080
148 - 4148
149 - 4229
150 - 4184
151 - 4078
152 - 4103
153 - 4070
154 - 4042
155 - 4160
156 - 4089
157 - 4119
158 - 4095
159 - 4074
160 - 4113
161 - 4140
162 - 4098
163 - 3963
164 - 4108
165 - 4075
166 - 4023
167 - 4037
168 - 4154
169 - 4147
170 - 3994
171 - 4080
172 - 4081
173 - 4198
174 - 4162
175 - 4084
176 - 4142
177 - 3978
178 - 4074
179 - 4122
180 - 4072
181 - 4109
182 - 4127
183 - 4244
184 - 4034
185 - 4064
186 - 4027
187 - 4184
188 - 4173
189 - 4085
190 - 4171
191 - 4130
192 - 4115
193 - 4095
194 - 4051
195 - 4171
196 - 4182
197 - 4050
198 - 4048
199 - 4073
200 - 4198
201 - 4126
202 - 4004
203 - 4119
204 - 4144
205 - 4145
206 - 4184
207 - 4077
208 - 4218
209 - 4101
210 - 4126
211 - 4120
212 - 4028
213 - 4092
214 - 4113
215 - 4107
216 - 4059
217 - 4179
218 - 4092
219 - 4128
220 - 4083
221 - 4112
222 - 4147
223 - 4134
224 - 4058
225 - 4066
226 - 4082
227 - 4082
228 - 4204
229 - 4126
230 - 4069
231 - 4207
232 - 3993
233 - 4086
234 - 4107
235 - 4108
236 - 4116
237 - 3978
238 - 4063
239 - 4013
240 - 4092
241 - 4102
242 - 4025
243 - 4073
244 - 4167
245 - 4178
246 - 4087
247 - 4100
248 - 4035
249 - 4070
250 - 3930
251 - 4034
252 - 4143
253 - 4146
254 - 4060
255 - 3970
Test run #4 histogram
0 - 4191
1 - 4126
2 - 4053
3 - 4151
4 - 4091
5 - 4119
6 - 4134
7 - 4117
8 - 4099
9 - 4117
10 - 4079
11 - 4098
12 - 4097
13 - 4107
14 - 4067
15 - 4044
16 - 4067
17 - 4095
18 - 4195
19 - 4051
20 - 4075
21 - 4068
22 - 4047
23 - 4241
24 - 4151
25 - 4105
26 - 3990
27 - 4014
28 - 4096
29 - 4113
30 - 4084
31 - 4048
32 - 4095
33 - 4158
34 - 3942
35 - 4042
36 - 4054
37 - 4141
38 - 4112
39 - 4065
40 - 4219
41 - 4180
42 - 4098
43 - 4052
44 - 4199
45 - 4048
46 - 4131
47 - 4031
48 - 4140
49 - 3959
50 - 4018
51 - 4071
52 - 4141
53 - 4026
54 - 4012
55 - 4109
56 - 4201
57 - 4077
58 - 4103
59 - 4126
60 - 4103
61 - 4066
62 - 4164
63 - 4068
64 - 4001
65 - 4118
66 - 4052
67 - 4061
68 - 4198
69 - 4194
70 - 4134
71 - 4017
72 - 4151
73 - 4107
74 - 4056
75 - 3915
76 - 4094
77 - 4101
78 - 3969
79 - 4063
80 - 4063
81 - 4080
82 - 3995
83 - 4190
84 - 4092
85 - 4021
86 - 3985
87 - 4071
88 - 4026
89 - 4144
90 - 4179
91 - 4117
92 - 4070
93 - 4130
94 - 4155
95 - 4054
96 - 4211
97 - 4105
98 - 4045
99 - 4129
100 - 4036
101 - 4131
102 - 4079
103 - 4310
104 - 4072
105 - 4095
106 - 4137
107 - 4110
108 - 4041
109 - 4116
110 - 4041
111 - 4062
112 - 4021
113 - 4073
114 - 4175
115 - 4201
116 - 4125
117 - 4027
118 - 3994
119 - 4144
120 - 4012
121 - 4225
122 - 4178
123 - 4160
124 - 4077
125 - 4161
126 - 4171
127 - 4098
128 - 4154
129 - 4118
130 - 4100
131 - 4164
132 - 4085
133 - 4057
134 - 4102
135 - 4054
136 - 4131
137 - 4016
138 - 4171
139 - 4071
140 - 4061
141 - 4090
142 - 4147
143 - 4111
144 - 3999
145 - 4012
146 - 4027
147 - 4083
148 - 4020
149 - 4107
150 - 4103
151 - 4113
152 - 4130
153 - 4045
154 - 4196
155 - 4143
156 - 4061
157 - 4150
158 - 4062
159 - 4160
160 - 4073
161 - 4110
162 - 4187
163 - 4041
164 - 4098
165 - 4119
166 - 4112
167 - 4166
168 - 4210
169 - 4010
170 - 4134
171 - 4026
172 - 4162
173 - 4025
174 - 4166
175 - 4034
176 - 4032
177 - 4081
178 - 4206
179 - 4058
180 - 4112
181 - 3983
182 - 4018
183 - 4042
184 - 3972
185 - 4144
186 - 4129
187 - 4038
188 - 3949
189 - 4018
190 - 4142
191 - 4250
192 - 4160
193 - 4134
194 - 4058
195 - 4147
196 - 4092
197 - 4073
198 - 4115
199 - 4134
200 - 4058
201 - 4153
202 - 4006
203 - 4127
204 - 4080
205 - 4127
206 - 4180
207 - 4108
208 - 4186
209 - 4014
210 - 4059
211 - 4080
212 - 4066
213 - 4161
214 - 4135
215 - 4082
216 - 4022
217 - 4064
218 - 4113
219 - 4032
220 - 4180
221 - 4168
222 - 4072
223 - 3989
224 - 4065
225 - 4074
226 - 4075
227 - 4180
228 - 4136
229 - 4108
230 - 4095
231 - 4154
232 - 4133
233 - 4023
234 - 4154
235 - 4102
236 - 4147
237 - 3996
238 - 4103
239 - 4076
240 - 4129
241 - 4086
242 - 4117
243 - 4104
244 - 4159
245 - 4116
246 - 3957
247 - 4052
248 - 4193
249 - 4128
250 - 4044
251 - 3995
252 - 4116
253 - 4222
254 - 4093
255 - 4133
------------------------------
From: [EMAIL PROTECTED]
Subject: Perl calls PGP
Date: Tue, 06 Apr 1999 18:57:14 GMT
Hi,
I write Perl's script to call PGPV by doing:
open (WRITEME, "|pgpv message.sig -o message");
print WRITEME "\n";
close (WRITEME);
It should enter a Carriage Return when PGP displays "File to check signature
against [message]: " but it doesn't work.
Could u tell me what is the problem exactly because I made the same test for
"rm -i test.txt" and it works.
Thanks!
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED]
Subject: Geometric Identification
Date: Tue, 06 Apr 1999 18:55:13 GMT
Abstract. This paper presents a simple, yet effective method of
user identification using geometric objects. The process is quite simple, and
uses little memory and can be perform relatively quickly.
--- snip ---
This is a small paper I wrote on Geometric Identification. It is available in
RTF format (.TXT upon demand), it currently is at:
http://members.tripod.com/~tomstdenis/gi.rtf
The paper is still in draft stages (I wrote it today). I hope you enjoy.
Sincerely,
Tom St Denis
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: <[EMAIL PROTECTED]>
Subject: SFS, recovering a backup
Date: 6 Apr 1999 19:04:52 -0500
I've got an image backup of a SFS partition, I know the key, but not
what the geometry of the disk it's on was.
Anyone have any hints?
------------------------------
From: [EMAIL PROTECTED] (Ian Goldberg)
Subject: Re: Live from the Second AES Conference
Date: 6 Apr 1999 23:05:57 GMT
In article <[EMAIL PROTECTED]>,
Matthias Bruestle <[EMAIL PROTECTED]> wrote:
>Ian Goldberg ([EMAIL PROTECTED]) wrote:
>> >With PalmPilots this scheme is no problem. There is enough space to
>> >store the coins. But with smart cards, where you have only up to 16kB
>> >the coins will feel like sardines in a tin.
>
>> Why? Each coin is only about 100-132 bytes, so you can have well over
>> 100 coins, which should be more than enough (assuming your ecash system
>> has the ability to make change).
>
>With change 16kB should be enough. Are there Ecoin systems where
>the change can be again spent?
Yes. HINDE, for example.
- Ian "HINDE Is Not Digicash's Ecash"
------------------------------
From: "Dave Bergert" <[EMAIL PROTECTED]>
Subject: Calculation of CVV's
Date: Wed, 7 Apr 1999 06:34:08 -0500
Anyone here familar with CVV calculations?
After encrypting block1 with Key A using DES
I recieve a result that is not in Hexadecimal format...
Any ideas of what I am overlooking to convert the DES output to a
Hex string ???
Thanks Much
Dave Bergert
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (R. Knauer)
Subject: Re: True Randomness & The Law Of Large Numbers
Date: Tue, 06 Apr 1999 23:41:23 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 06 Apr 1999 03:10:10 GMT, "Douglas A. Gwyn" <[EMAIL PROTECTED]>
wrote:
>And any *thoughtful* person who has been involved with the network
>newsgroups understands that an attribution to someone whose words
>*are not cited at all* is inappropriate and should have been omitted.
I have no quarrel with that, but it still is not an excuse for
misattribution.
Bob Knauer
Signature files traditionally present comments from a population with an IQ over 100.
I thought it would be appropriate to present comments from the other half of the
population,
namely the one with an IQ under 100. After all, they do make up 50% of the human
condition.
Never mind that they also tend to become politicians. Here is a representative example:
"People blame me because these water mains break, but I ask you, if the water mains
didn't break, would it be my responsibility to fix them then? WOULD IT!?!"
- Marion Barry, Mayor of Washington DC
------------------------------
From: [EMAIL PROTECTED] (R. Knauer)
Subject: Re: True Randomness & The Law Of Large Numbers
Date: Tue, 06 Apr 1999 23:41:25 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 6 Apr 1999 00:51:02 -0500, "Franzen" <[EMAIL PROTECTED]> wrote:
>>I would ask what are you using for the TRNG sequence in that test.
>>IOW, you must be comparing a purported TRNG to a known TRNG. What is
>>that known TRNG?
>The fair flipping of a two-sided coin is one.
Unfortunately it is a classical process, and therefore not TRULY
random. For all practical purposes it is as nearly random as one can
expect for a classical process, but it is still not a true random
process.
>It matches most closely
>the kind of TRNG process I think underlys most of your recent postings
>in this newsgroup.
That would be the Uniform Bernoulli Process, which results in the
uniform random walk in one dimension. That process results in abnormal
properties for the bias, which is measured by the sum of the
individual steps.
>>You evidently were not here when we discussed that issue.
>I am here now.
Welcome to Bedlam.
>I am asking you about the discrepancy in your two
>statements. The "we" must be you and some other person(s); my not being
>"here" excludes me from this particular "we."
>The rest of your words answering my original posting do not pertain to
>my above extracted discrepancy statement to you. I would appreciate a
>responsive answer from you, or none at all.
I frankly forgot what the issue was.
>>We all realize that it is impossible to build a classical TRNG that is 100%
>>random.
>What the devil is a "classical" TRNG?
Any process that can be decribed by classical physics. A fair coin
toss, for example. A quantum TRNG would be a process that depends on
quantum physics.
>100% random? That implies 99, 98, ... . Are you one of those who think there
>exists some sort of uniform randomness acme.
No, actually I think there are only one of two possibilities: certain
true randomness, and all the rest.
The standard for true randomness is a quantum computer that is
programmed to calculate true random numbers. The algorithm for that
already exists but the quantum computer does not. But it will be
built, so the standard will exist some day.
>I think of uniform randomness as concurrently "perfectly imperfect"and
>"imperfectly perfect."
Berry's Paradox in disguise.
>>There will always be flaws which disturb the process and give
>>it some small amount of non-randomness, such as slight 1-bit bias.
>
>So your hypothetical TRNG will "always" be flawed. I am not very encouraged
>by this part of your vision.
I was referring to existing TRNGs like a radioactive TRNG (Cf.
"HotBits"). But a quantum computer will presumably not suffer from any
such flaws.
>Just what is 1-bit bias in your view? Most of the concepts you present as
>parts of your current position I can visualize; your bias concept I cannot.
1-bit bias is measured by the distance from the origin in the random
walk. It is the excess of one bit over the other. There are other
kinds of bias, such as 2-bit bias, etc. Borel normality applies to all
possible bit-group biases.
>I think taking the mystery out of uniform randomness can change the current
>view of "truth" in many unimaginable ways. That may be a QM layman (non-TM)
>equivalent to what you say; or maybe not. I will let you decide.
If we could define true randomness algorithmically, we could use that
to fabricate a hidden variable theory for quantum mechanics.
Bob Knauer
Signature files traditionally present comments from a population with an IQ over 100.
I thought it would be appropriate to present comments from the other half of the
population,
namely the one with an IQ under 100. After all, they do make up 50% of the human
condition.
Never mind that they also tend to become politicians. Here is a representative example:
"People blame me because these water mains break, but I ask you, if the water mains
didn't break, would it be my responsibility to fix them then? WOULD IT!?!"
- Marion Barry, Mayor of Washington DC
------------------------------
From: [EMAIL PROTECTED] (R. Knauer)
Subject: Re: True Randomness & The Law Of Large Numbers
Date: Tue, 06 Apr 1999 23:41:26 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 06 Apr 1999 06:10:31 GMT, [EMAIL PROTECTED] wrote:
>> Set #1: Reasonable certainty that the process is not random;
>> Set #2: Processed which do not exist in set #1.
>> There is no middle set of RNGs that are maybe random, maybe not random
>> on the basis of reasonable certainty.
>There is set #2. Where does FIPS-140 say that sources in set
>#2 are truly random?
It doesn't. I never claimed it did. I was pointing out that FIPS-140
claims that #1 exists, and from that I concluded that #2 also exists.
>I'm saying that membership in set #2 does not imply a generator
>is truly random.
Membership in set #2 is based on non-membership in set #1. There is no
other sets. A RBG is either truly random or not. There is no set of
"half random" processes.
>Does FIPS-140 say otherwise?
All I said was that FIPS-140 claims that set #1 exists, the set
containing all non-random processes. It follows that set #2 exists as
a disjunction. Therefore the tests can be used to test set membership.
If it is not in #1 then the process must be in #2.
If that weren't the case then the tests would be unreliable.
>We can have it the way it is. If the facts are inconsistent with
>the theory, then the theory is wrong. If the facts are consistent
>with the theory, then the theory remains plausible.
>I challenged you to quote FIPS-140 where it says what you claimed
>it says, as re-included at the top. A reasonable response would
>include text from FIPS-140, or a retraction. I am not asking you
>to quote yourself.
There is a third possibility, namely that you misread my statements.
I did not say that FIPS-140 claimed the existence of set #2. I said
that FIPS-140 claims the existence of set #1. I am the one who
concluded that set #2 exists, since I believe there is such a thing as
a true random number generator.
A RNG is either a true random number generator, like a quantum
computer programmed to calculate true random numbers, or it is not a
true random number generator. True random number generation is a
nominal distinction, not a quantitative distinction.
If tests claim reasonable certainty that a RNG is not random, then
when an RNG does pass the tests, that constitutes reasonable certainty
that the RNG is truly random. But we know that cannot be, so the
conclusion is that there can be no tests that claim reasonable
certainty that a RNG is not truly random.
BTW, I will tentatively accept a very large significance value as a
measure of resonable certainty for the sake of discussion. In reality
even that is flawed in terms of an ideal TRNG. How about a Z-score
that is outrageously large, which necessitates a large sample.
I will then submit my trial TRNG to the tests and if it passes that
means it is reasonably certain to be a properly functioning TRNG.
Never mind that I plan to submit the digit expansion for pi, which is
not random at all.
Thus far in my exposure to statistical methods I have come across an
incredible amount of circular reasoning. Test are presented which can
give you certainty to an arbitrarily high degree of certainty and yet
are so unreliable that if a RNG passes them you still cannot say the
process is random. That makes the tests practically worthless.
The objective is not to produce bad RNGs, the objective is to produce
good RNGs. If you cannot reject a bad RNG with a very high of
certainty, enough to be able to consider those that do pass the tests
are in the opposite category, you are just guzzling snale oil.
Bob Knauer
Signature files traditionally present comments from a population with an IQ over 100.
I thought it would be appropriate to present comments from the other half of the
population,
namely the one with an IQ under 100. After all, they do make up 50% of the human
condition.
Never mind that they also tend to become politicians. Here is a representative example:
"People blame me because these water mains break, but I ask you, if the water mains
didn't break, would it be my responsibility to fix them then? WOULD IT!?!"
- Marion Barry, Mayor of Washington DC
------------------------------
From: [EMAIL PROTECTED] (R. Knauer)
Subject: Re: True Randomness & The Law Of Large Numbers
Date: Tue, 06 Apr 1999 23:41:24 GMT
Reply-To: [EMAIL PROTECTED]
On Mon, 5 Apr 1999 23:16:47 -0600, [EMAIL PROTECTED] (Jerry Coffin)
wrote:
>A test of a random number generator can never be an absolute thing.
>You can NEVER say that failing a particular test means a generator is
>not random. A generator that could NOT generate a failing sequence of
>numbers would, by that very fact, be provably non-random.
>To give an example: if I feed a string of 100 1's to a test of a
>random number generator, I think it's safe to say that NO test would
>pass this as a random sequence. Despite this, a truly random
>generator must be EXACTLY as likely to produce this particular
>sequence of numbers as ANY other sequence of the same length. If it's
>not, we can predict that it's less likely to do so, and the output is
>now predictable rather than entirely random.
BINGO!
That's the same argument that is used to prove the non-existence of an
algorithm for deciding the halting problem.
Are we starting to build a prevailing consensus behind my position?
I hope it doesn't happen too fast - I am having too much fun right
showing that the so-called "experts" are wrong.
Bob Knauer
Signature files traditionally present comments from a population with an IQ over 100.
I thought it would be appropriate to present comments from the other half of the
population,
namely the one with an IQ under 100. After all, they do make up 50% of the human
condition.
Never mind that they also tend to become politicians. Here is a representative example:
"People blame me because these water mains break, but I ask you, if the water mains
didn't break, would it be my responsibility to fix them then? WOULD IT!?!"
- Marion Barry, Mayor of Washington DC
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and sci.crypt) via:
Internet: [EMAIL PROTECTED]
End of Cryptography-Digest Digest
******************************
