<p>My thought around using introducers was in part to address a use case of Zooko's suggestion...but as always it might just be my intended use case...</p>
<p>If we are using a local copy of the trusted nodes then it's like we're back in the good ol' days of maintaining hosts files.</p>
<p>I like the idea (don't get me wrong) but I can envisage setting up an introducer at location #1 (say, my house) and one at location #2 (different country perhaps) and also joining a pubgrid. The introducers I control can have some setup only allowing trusted nodes to connect.</p>
<p>I can then say I treat *my* introducers differently than the pubgrid without needing to know all nodes connected.<br></p>
<p>I *think* a generic descriptor is what's needed. Some grids might want geocoding, some might want to trust at a introducer level, some might want something else. I guess the issue then becomes how does a user build a ruleset for their use case? Is JSON & some scripting language (GNU Guile?) the answer, or is it over engineering? <br>
</p>
<p>Cheers,</p>
<p>Olaf<br>
</p>
<div class="gmail_quote">On 17/01/2012 7:53 AM, "Zooko Wilcox-O'Hearn" <<a href="mailto:zooko@zooko.com" target="_blank">zooko@zooko.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On Mon, Jan 16, 2012 at 12:47 PM, Greg Troxel <<a href="mailto:gdt@ir.bbn.com" target="_blank">gdt@ir.bbn.com</a>> wrote:<br>
><br>
> I'd say write up a plan and send it to the list.<br>
<br>
Yay!<br>
<br>
> 1) Decide if we are going to trust storage nodes to express the variables that are correlated honestly. I think it's at least near impossible not to trust them and make progress.<br>
<br>
You could have a text file containing each storage server's furl and<br>
its properties in some flat textual format, and put that into your<br>
gateway's node directory (~/.tahoe). Then that gateway would believe<br>
the contents of that file when choosing which shares to upload to<br>
which servers.<br>
<br>
As a bonus, this automatically solves the Sybil Attack, in which<br>
someone sets up a large number of storage servers so that by chance<br>
your gateway uses only all or mostly storage servers controlled by<br>
them. Having a locally-controlled file describing storage servers<br>
means that the gateway could have a requirement like "make sure at<br>
least K servers that I upload to appear on this list".<br>
<br>
Therefore, I don't think there is any need for the gateway to receive<br>
a description about a storage server directly from that storage server<br>
and then rely on that description. At least as a first cut, it would<br>
seem better to rely on the gateway's system administrator for that.<br>
<br>
(I'm sure this will turn out to be inadequate for some uses, but it<br>
seems like a good starting point.)<br>
<br>
Regards,<br>
<br>
Zooko<br>
</blockquote></div>