[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

[Freedombox-discuss] Store-and-forward is a necessity

On Fri, Mar 04, 2011 at 10:02:11AM +0000, stillyet at googlemail.com spake thusly:
> But if we're to provide a system which works for people living under repressive
> regimes, then we have to allow anonymity. And if we allow anonymity, then we
> allow nym-shifts and sock-puppetry and all the trollish behaviour which has
> eroded Usenet. I don't see a solution to this, beyond using webs of trust and
> reputation to allow our clients to discriminate between 'trusted',
> likely-to-be-valuable messages and untrusted, unlikely-to-be-valuable ones...

Part of Usenet's problem was the entirely broadcast and relatively immediate
nature of it.  You troll now, you get your flame in an hour or less. A system
such as we are discussing could potentially have very significant delays. And a
lot of the communication would be one to one. Sure, back in 1989 (when I first
read Usenet) there may have been long delays due to daily dialups and UUCP
transfers but that pretty much ended up the mid 90's. Trolls want a reaction
and they are impatient.

> If we do that, then some new user to the system reporting some significant
> human rights violation is unlikely to be listened to because (s)he won't have a
> reputation and won't have a well-developed web-of-trust.

Even if there were such a reputation system in use someone would read the
untrusted stuff and pass on up (by signing their pubkey and passing the message
along) the message to others and the person would gain a following. I really
like web of trust and it could certainly find application in this sort of
system but I don't think the troll problem would be that bad. It just doesn't
seem like an attractive environment for them.

> At a technical level Usenet has a lot of merit and a sort of UsenetNG with the
> addition of opportunistic routing, webs-of-trust and automatic digital signing
> of messages to prevent (or at least expose) in-flight modification is a good
> model for at least part of what we want to achieve; Usenet successfully mixes
> real-time and store-and-forward transmission models to achieve rapid
> decentralised worldwide broadcast of messages, with relatively low bandwidth

Indeed. Usenet may be something to look at utilizing and improving on here.
Although I do wonder how one would geographically route usenet messages if the
messages are to be transferred via wifi and hand-carried on portable devices

Tracy Reed
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/freedombox-discuss/attachments/20110304/bce22ae1/attachment.pgp>

Reply to: