Topic: My Ideal In-Store Bitcoin Shopping Experience (Read 2476 times)

That's why they send the coins from home. By the time they get to the store with a car, they'll probably already have a confirmation or two (or few). It's possible their order won't be confirmed in time, though, and for regular customers (this could either be done via ID or less intrusive with a store card), the store could accept theirs on a kind of "trust credit" without making them wait. Users could adjust their behaviors a bit, here. If they used Mycelium or another Android wallet with needed capabilities, they could do something like place the order on break at work and then pick up the order on their way back with no risk of having to wait.

Ahhhh I see.  "...They sign the public key they used to pay" is slightly arbitrary, but yeah, I'm tired

Here's my small concern:

As a consumer, my ideal scenario is to pay at the store, but extremely quickly.  This means not having to wait for a confirmation which would likely be required, and it also means not paying at home with a non-reversible payment method before I've physically seen whatever I just bought.

Call me picky.  Every time, I see confirmation times being a legitimate issue for many types of brick-and-mortar businesses (though certainly not all).

You send the coins at home (you "sign the transaction" at home). You're only signing a message with the same address you sent from at the service window so groceries go to you instead of some other random guy.

I am sure that it would be possible for coinbase (or an exchange) to sign a message for you on your behalf, although any message they sign would also likely include the TXID in question - this may or may not be an issue for merchants verifying a signature.

I wouldn't be surprised if we see apps in the future that would allow a smartphone to sign a message, it just isn't available now.

I also don't think that most merchants would really be taking on that much risk to accepting a 0/unconfirmed TX as it is fairly easy to detect a double spend attack, especially not so much risk that they would want to incur the extra expense of having to pay someone to gather the items purchased for you

Amazing.

Hijack away!  I really like the model you propose

One question, though:  Maybe I'm just really tired, but how does placing the order at home allow time for confirmations if you're signing with your key at the warehouse?

Edit:  It's the bolded selection I'm confused about.

Yessss.  I love it.  This definitely works in favor of a card pre-paid with BTC.

We already have technology and code to handle signing. Mycelium can already do this on Android, but most don't (and none are "automated" -- there isn't a URI protocol for sending pre-grenerated messages for the phone to sign, AFAIK) - but this could change if there's investment in an automated store handling BTC. It's not terribly complicated stuff. Coinbase might be able to come up with a solution for this (BCI can handle this since users control keys), but if not, there are alternative web & Android-native clients which can.

(sorry to hijack, OP)

ETA: Of course, for store-owners who aren't risk-averse, they could still accept credit/debit at the service window, or even cash (cash'd be super-risky, though, because you have to transport it from a man-less store). This could happen without bitcoin... the real challenge is in writing a script to handle things like bagging (a store COULD have users bag themselves like "no frills" grocery stores, though), figuring out how to shelve and move all these items, and determining when the store needs to sell on clearance or dispose of inventory (including factors like humidity and temperature)... but if the community's underserved or unserved, it's not like you need to spend any thought on things like marketing. In my town and in a good few other rural areas, Dollar General's stepped in to provide a very limited number of extremely over-priced groceries, but it's really a piss-poor solution as far as the community members are concerned.

This is an interesting idea, it is somewhat similar to what Ikea does (but they charge for this service).

The main issue is that I don't think there are any applications available for smartphones ATM that allow for people to sign a message plus some people who use 'wallets' like coinbase or send directly from an exchange would not be able to sign a message regardless

1) Honestly, it's a little shocking that we don't already see this means of shopping with any other payment system (unless you count something like a Mobil Speedpass at gas stations...that's about the closest I've seen).

2) Word

3) I'm talking about a change indicator (qr-code/LED/whatever) and getting it to respond to a BTC transaction without in any way jeopardizing the security of that transaction.  This would be good enough as a proof of concept.  The payment device is not what I'm focused on.

I'd have a different idea for underserved or completely unserved rural and rural-suburban areas (towns of <~1k people which can't sustain a manned store). It could work in urban and urban-suburban areas, but it'd necessarily need to be a new structure built, so retrofitting a store and giving scanners would be better there. An old one can't really be retro-fitted cost-effectively. A grocery store could be fully automated, possibly even including stocking if a way could be figured out to have that outsourced to the trucking/logistics company at the time they unload.

A user would place their order online, go to the service window where they sign the public key they used to pay (with some automated solution), and then their order is then bagged inside (would probably need to be paper), coming straight out to the user, almost like how an ATM works, but with conveyor belts and many "non-standard" products. This design may also even allow some type of third-party utility for vehicles (either each user buys their own, or the store could provide a chained or maybe even just keyfobbed version), which'd get around issues where an incapable person requires help to grocery-shop (excluding getting groceries from the car to the house).

The store would look like a warehouse with a little conveyor belt coming out through one or a few services windows with a small BT/RFID communicator for phones or dedicated HW wallets (since payment would be made at home, a phone is probably fine for signing the address). Other payment sources would also be valid. No scanners or individually scanning products - it'd work just like Amazon but without the shipping costs, pantry schemes, drones, or waiting. It also allows stores to get a couple confirmations in to prevent double-spending since it'll take a short while for the user to drive to the service window.

Thank you for your comment, but it misses the point of the OP.  The OP describes my ideal in-store shopping experience. I don't want a QR-code at the register (though it's a great solution right now for brick-and-mortars).

The green LED was an example of one potential change indicator, not necessarily the exact one I need.  I'm simply proposing a general model and trying to flesh out the specific details a bit more to something that's viable.  I fully believe that you would be able to outfit every product with some type of change indicator for less than 15 cents (ideally, less than 5).  How much do those little magnetic strips cost that cashiers take off and toss away so you don't set the alarm off when you walk out the door?  It can't be much, and it works just fine.  Each of these strips is also "programmed."  It's perfectly reasonable to imagine creating a semi-automated process for quickly tagging products with a change indicator that corresponds to a unique public address.

The TX fee issue is a legitimate one; good point.  I suppose this could be solved by having a device that can hold your selected items in a queue such that you can pay for all of them at once when you are done shopping.  This is already currently doable, though the issue here becomes the number of public addresses used and where/how you fund them.

Yes, something like this would be fine as long as you don't have to spend any longer than a few seconds in the scanning area.  I'm only interested in an in-store model that bypasses anything similar to waiting in line.  I would even be turned off if I had to walk out of my way whatsoever to get to the scanning area; if it's on the way out the door, that's fine.

I would think that scanners would be pretty expensive, especially considering that some stores will sometimes have hundreds of customers in their store. I don't see why you couldn't simply have customers put their items in their shopping cart and then have them use a self checkout as many grocery stores have now.

Want to get rich?   Come up with a way to direct customers straight to a product on your shelf.

This is the elephant in the room.

How many times have you wandered aisles looking for something?   Every single time you go to the store?

Last week I wasted a half hour walking through Walmart trying to find an item.

Ultimately I had to go to the front, find the store manager, and only because she had memorized every single thing in that store, was she able to take me right to the product.

The entire time I was repeating to myself "Never again.  Just shop online.  There's a Search Box".

-----------

This is something that can be done today.  And not surprisingly, its the first step towards automated checkouts as well.  Once you've got all your products indexed and locatable, they're already in a "system" and can easily be scanned and paid for.

-B-

The system I posted is even better than that because it can eliminate shoplifting too. Any payment method can be incorporated into it easily.

This is a good example as to why this specific setup would not work. The setup as described in the OP would really not save anything on labor costs as stores would still need to employ people to make sure everything has been paid for and that there is nothing that would be a risk to the company (for example a double spend attack).

One risk that has not been mentioned is that someone could potentially send a TX that is not likely to be confirmed by the miners (for example a very large transaction with no fee) and then use inputs from that TX to fund all the TXs that pay for each of the goods. Since the store would normally be relying on a number of 0/unconfirmed TXs this original TX would likely not be noticed/unacceptable. On the other hand if someone were to pay for everything at once then the store would only need to look at one TX to judge the risk as to if it is likely the TX is safe to accept with 0/unconfirmed status.

Probably, but thats a pretty long shot since adoption takes time and we don't know what kind of technology is going to compete with bitcoin in the near future.


Indeed, and thats the beauty of bitcoin. It wouldn't be that hard getting it to work


Are you simply talking about the attached QR codes on the items in shops or about a whole payment/checkout system including the POS terminal and the payment devices?

When my wife had a clothing store I tried to set something like this up. I created QR codes on stickers that I put on the price tags of the clothes. Each QR code took you to a BitPay link with the price and a place to pay with BTC.

After that experiment I realized that it would be too much hassle to set up a distinct sticker and QR code for each item and women's clothing was not the best place for early adoption of such a thing, especially in 2011 when hardly anyone had heard of Bitcoin.

What OP describe can be done with a smart phone. I think we need something better.