Silk Road forums

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

You may now upload a picture of your items!  Only .jpg is supported at the moment.  Have fun and be creative
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.14 (GNU/Linux)

iQEcBAEBAgAGBQJOOh06AAoJEAIiQjtnt/olf40H/iexlkw2tZFmkgMyF1DcTdbc
NVH5fWDqSIDs1FnOl15xPrfXlQ5dRfY9X+Rqs+tftrgDtYrw1olyQNNM/madzmz6
UJiqt6oCdBwABSzyWAAcFLtZIcAZjQuDYQ1soOG4eUmR4Cv1UfV6EEOppFzRN0+j
J9xunC84pDrJDayJ86CcmBTaP0V0Z/rrDUn/j1Ew/yW8EnjmtLTIVJZCoANvkNSw
m7+SDm25tZ+/dqzUvGoPRSBD4SYrjuoOpY9eqJjNsjxvOW90HQ3XBJxD7Xx1+c8P
P4tK0ArLUOeJ+M+FZkejlvrqApixcDTANhJPpyIxx/WPwMY/6qnceMSCQW/kPwo=
=jG2D
-----END PGP SIGNATURE-----

When payment in escrow is hedged, the exchange rate we use is not the last price, nor is it the 24hr moving average we use when you peg your prices to the dollar.  Instead, it is the price of the best open order or orders for your transaction volume.  When hedging, this is the best asking price, and when un-hedging, this is the best bidding price.  So when hedging and then immediately un-hedging, your always going to lose this small percentage.  For those who know, this is called the bid/ask spread.

So...

Does this mean a small-scale seller with a previously stellar track record can get one bad review from a flakey buyer and have their rating totally fucked because the new feedbacks carry more weight and they don't have as many feedbacks as a large seller?

Or am I misunderstanding something here.  Because I'm not really digging the sound of that.

if you have a bunch of good feedback for a month, no feedback for a couple of months, and then start getting bad feedback, then yes, your avg rating will drop more quickly than if there wasn't more weight given to newer reviews, but this is what we want.  Basically you have to stay consistently good to keep a high score and can't rely on your performance from months and months ago.

Would this mean then that as the volume grows exponentially that there is a good chance that the fees will slide downward; both because of volume and the fact that there will be somewhat less to do in the way of major updates of the board because it will become both highly efficient and more bulletproof over time with each new iteration?  Based on your philosophy of money (the tool part is the same as my own, however I do *love* the creature comforts that come from having it) I would be very interested to hear your answer, even if it is just in a philosophical bent rather than any estimation or indication of future fee structures.  thanks!

Not necessarily.  Basically what I'm saying is, MUCH more is possible than just being able to buy drugs freely.  With the revenues from Silk Road (the more the better), we can invest in even more ambitious and revolutionary projects.  Money is a tool, but so is Silk Road.  Our aim is to one day live in a world where we won't need Silk Road, because we won't need to hide anymore.

(A shameless plug for my self-serving interests:)

I'd like to see a variable number tempered by a standard deviation of total reviews used as the base of the power function determining the aging weight; a general change in quality could increase the base and make past reviews have less weight and minimal change in quality could allow past reviews to hold higher weight. It just seems like a nice stability buffer for generally-consistent vendors (plus, I don't see why past reviews should lose half of their value every month if the vendor's staying consistent).

our thinking here is to keep established vendors from resting on their laurels.  If the weight of ones feedback doesn't fall off over time in some way, then new feedback becomes less and less important, and new vendors will have little chance of catching up with the senior sellers.

we can definitely share the detail, though it gets a bit technical, and we are tuning some the parameters, so the whole thing isn't set in stone.

basically, each feedback score is weighted by three factors: buyer activity, age, and purchase price.

The buyer activity weight is determined by finding the average and standard deviation of the log of the number of vendors reviewed by buyers who've ever left a review.  if the log of the number of vendors reviewed by the buyer who left the feedback is greater than the average, the review has a "buyer activity" weight of the number of standard deviations above the average plus one.  If it is below average, then the weight is the inverse of the number of standard deviations below the average minus one.

The age weigh is found by raising 1.02337 to the power of the review's age in days and then inverting it.  We got 1.02337 by making a review half as valuable after one month as one that's just been posted.

The price weight comes from the following formula: 0.2*log(0.02*($price)+1)+1 where $price is in dollars.

You then multiply the weights together to get the total weight for that review.  Add up all of the weighted reviews and divide by the total weight to get the weighted average, then do 25*(1-avg) to get "percent positive reviews" metric.

To get the seller rank, we find the averages and standard deviations of the "percent positive feedback" and the log of the total weight of the reviews for all active sellers. We then average the number of standard deviations a seller has above or below the average with a 4 to 1 weight on their average feedback over their total weight.  This is a seller's rating, and we rank everyone according to this rating.

There are a few little details left out for brevity, but that's basically it.

Quote from: todamnhellit on July 26, 2011, 04:43 am

Quote from: PotHead on July 25, 2011, 06:32 am

I was wondering, how do you know if an item is hedged or not? Does it tell you when you add it to your cart or something?

While adding an item to your cart, it specifies if it's hedged or not.

I don't get it - why does the buyer care?

If they get refunded fewer/more bitcoins, they'll know why.

Check how many sales you have listentothemusic.  I noticed they have me down for an extra sale that doesnt exist.  I have 19 sales and they have it down as 20 which really fucked up my ratio.  I have only 1 4 rating the rest are 5.  The math is off.  Check yours out and let us know as well as silk road admin.

This is due to transactions that have finalized but the buyer has yet to leave a review.