i wrote-
Delay--time--isn't what matters. It's confusion about which message is which that matters. So if I get 10 messages in one minute, I can scramble the order and send them out the next minute, and I've done my job--at least the order-scrambling part. (You also need to pad or packetize messages.)
So use serial numbers, not times! Send a message for every one you get, keep a fixed number of messages queued, and add dummies if necessary to keep things moving.
Jim replies.
Wrongo...the random time stamp does randomly re-order then. As to bogus messages, not on my system you won't....
Okay, first I'll go over the case where delay without dummies does NOT reorder. Then I'll go over the case where delay simply adds needless, well, delay. Then I'll talk about the cost of dummy messages. Assumption: Your remailer assigns each message a number from 0 to 59 and remails it at that minute of the hour. Whether it's hours in the day, minutes in the hour or seconds in the minute only changes which of the following two cases is more likely: Case 1: The remailer receives no messages for 61 minutes, then one message, then no messages for 61 minutes. Result: In the hour following receipt of that one message, only one message is sent. Guess which message it was. Case 2: 60 messages arrive in one minute. Result: The last one(s) go out about an hour later. They could have all been sent in the next minute with equivalent reordering. P.s., if 60 messages arrive *every* minute, under the assumption above, you have to save an average of 3600 messages. So, with this method, you can adjust the delay time down to guarantee delivery time, or up to make reordering *more likely*, but you can't guarantee reordering. If you want 1/N reordering to be likely, you need to set the response time to N times the inter-arrival time for the *quietest* traffic periods. To guarantee reordering you have to either wait indefinitely for enough messages, or after a while insert some of your own. To get over the problem of needless delay, you either need to invent some kind of tricky variable-delay scheme, or turn your attention away from clock time and focus on ordering.
I have a system which runs of a SLIP feed and bandwidth is sacrosanct. If you would like to pay for an additional line to handle the added load then fine but my pocket book won't support it.
Say the worst turnaround you want is 24 hours, and you want to get 1-out- of-10 reordering. Then on a day where you receive only one message (for this you got a SLIP connection?) you would need to generate 9 dummies. Assuming 10Kbyte messages, the bandwidth required is... 10.4 baud. For a turnaround of 2.4 hours ... 104 baud. 15 minutes ... 1040 baud. And remember, you generate only enough dummies to keep up the minimum *total* traffic, so in reasonable-traffic periods, you generate no dummies and the amount of real traffic you can handle isn't affected.
the consideration of the future where there will be many small systems with minimal bandwidth and monetary resources then I realy doubt they will be interested in any system which slows down or otherwise wastes a precious and critical resource.
I guess this sentence, which you quote, wasn't clear:
Plus, if the remailers only generate dummies when necessary, the total dummy traffic is self-regulating, since multi-hop dummies are x-lax for every remailer they pass through.
In other words, in a world of communicating forwarders, a dummy message from one just looks like regular traffic to any others it goes through, and serves to keep their traffic levels up--the more remailers the fewer dummy messages each remailer has to generate.
I also oppose the implied synchronicity of your methods as well. I am looking at a resonably secure asynchronouse method of making the traffic analysis difficult (the real reason for all this mumbo jumbo in the first place).
Traffic analysis is difficult when the order of messages is sufficiently scrambled. If you don't reorder, then delay doesn't help. If you do reorder, then added delay doesn't help. Whether one message is "synchronized" or not with a random other message isn't useful information to an outsider.
Why should I provide a potential monitor with the information that a certain amount of information going out will be bogus?
How might this information help analyze traffic? Also, as I mentioned, if you send your dummies to yourself indirectly, then pretty soon the level of input will match the level of output, and the ratio of bogus to real messages *won't* be visible. As far as I can see, dummy messages are simply necessary if you want to guarantee both reordering and response time. Please explain if you believe differently.
You obviously don't pay all the costs for your feed or else you are very rich...
Are you charged per byte or just a flat rate? -fnerd - - - - - - - - - - - - - - - the snack that eats like a food -----BEGIN PGP SIGNATURE----- Version: 2.3a aKxB8nktcBAeQHabQP/d7yhWgpGZBIoIqII8cY9nG55HYHgvt3niQCVAgUBLMs3K ui6XaCZmKH68fOWYYySKAzPkXyfYKnOlzsIjp2tPEot1Q5A3/n54PBKrUDN9tHVz 3Ch466q9EKUuDulTU6OLsilzmRvQJn0EJhzd4pht6hSnC1R3seYNhUYhoJViCcCG sRjLQs4iVVM= =9wqs -----END PGP SIGNATURE-----