Paul Puey [ARCHIVE] / is on Nostr

⚠️ This account is not associated with a real person. It's an archival account for mailing list messages.
🗓️ Active from 2015-02-05 to 2015-02-06
💬 Posted a total of 10 messages
📚 Participated in the following categories: Bitcoin-development
🔇 Please do not DM this user as it's used for archive purposes only.

Last Notes

2023-06-07 17:29:53

- reply

📅 Original date posted:2015-02-06
📝 Original message:Thanks for all the feedback Eric. You know we value all that you have to say. That's what this forum is for. We're looking for great ideas to harden this protocol and we're not closed to better ideas and we'll improve it as suggestions come up.
Paul Puey CEO / Co-Founder, Airbitz Inc
619.850.8624 | http://airbitz.co | San Diego
On Feb 5, 2015, at 5:05 PM, Eric Voskuil <eric at voskuil.org> wrote:
> On 02/05/2015 04:49 PM, Paul Puey wrote:
> The trust can be considered bootstrapped by visual verification of the
> address prefix.
Another (unspendable) address can trivially match the prefix. Imagine
someone walking around in a mall with a phone in the pocket with a
malicious app, just disrupting business by causing money to be burned.
Manual verification doesn't fix this attack.
> If we are really concerned about someone jamming a Bluetooth signal
> in a coffeeshop then the UI can encourage verification of the prefix.
I don't think it would be great to constrain a standard implementation
to low cost purchases or the need for manual verification, but again
manual prefix verification isn't actually a solution.
> Much like how regular Bluetooth requires 'pairing' via entering a 4-6
> digit code.
An appeal to the security of BT bootstrapping isn't exactly flattering.
You know I love Airbitz, and I know you guys are extremely privacy
conscious. I personally would have no problem using this feature under
certain circumstances. My question is only whether it would be wise to
standardize on the proposal as-is.
e
> On Feb 5, 2015, at 3:46 PM, Eric Voskuil <eric at voskuil.org
> <mailto:eric at voskuil.org>> wrote:
>
> On 02/05/2015 03:36 PM, MⒶrtin HⒶboⓋštiak wrote:
>>> A BIP-70 signed payment request in the initial broadcast can resolve the
>>> integrity issues, but because of the public nature of the broadcast
>>> coupled with strong public identity, the privacy compromise is much
>>> worse. Now transactions are cryptographically tainted.
>>>
>>> This is also the problem with BIP-70 over the web. TLS and other
>>> security precautions aside, an interloper on the communication, desktop,
>>> datacenter, etc., can capture payment requests and strongly correlate
>>> transactions to identities in an automated manner. The payment request
>>> must be kept private between the parties, and that's hard to do.
>>
>> What about using encryption with forward secrecy? Merchant would
>> generate signed request containing public ECDH part, buyer would send
>> back transaction encrypted with ECDH and his public ECDH part. If
>> receiving address/amount is meant to be private, use commit protocol
>> (see ZRTP/RedPhone) and short authentication phrase (which is hard to
>> spoof thanks to commit protocol - see RedPhone)?
>
> Hi Martin,
>
> The problem is that you need to verify the ownership of the public key.
> A MITM can substitute the key. If you don't have verifiable identity
> associated with the public key (PKI/WoT), you need a shared secret (such
> as a secret phrase). But the problem is then establishing that secret
> over a public channel.
>
> You can bootstrap a private session over the untrusted network using a
> trusted public key (PKI/WoT). But the presumption is that you are
> already doing this over the web (using TLS). That process is subject to
> attack at the CA. WoT is not subject to a CA attack, because it's
> decentralized. But it's also not sufficiently deployed for some scenarios.
>
> e
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/c64e1322/attachment.html>

2023-06-07 17:29:52

- reply

📅 Original date posted:2015-02-05
📝 Original message:Although not perfect, and it may require visual/verbal verification, I
don't see what the trust issue is.
[image: logo]
*Paul Puey* CEO / Co-Founder, Airbitz Inc
+1-619-850-8624 | http://airbitz.co | San Diego
<http://facebook.com/airbitz> <http://twitter.com/airbitz>
<https://plus.google.com/118173667510609425617>
<https://go.airbitz.co/comments/feed/> <http://linkedin.com/in/paulpuey>
<https://angel.co/paul-puey>
*DOWNLOAD THE AIRBITZ WALLET:*
<https://play.google.com/store/apps/details?id=com.airbitz>
<https://itunes.apple.com/us/app/airbitz/id843536046>
On Thu, Feb 5, 2015 at 2:05 PM, Eric Voskuil <eric at voskuil.org> wrote:
> Hi Paul,
>
> The issue is in the establishment of trust. Anyone can broadcast the
> initial information.
>
> e
>
> On Feb 5, 2015, at 2:01 PM, Paul Puey <paul at airbitz.co> wrote:
>
> The broadcast is ONLY done when the wallet is in Receive mode. Same as
> when the QR code is visible. The use of the *Name* section is specifically
> so that a recipient can broadcast their name/handle. Not so the recipient
> would broadcast the name of the Sender.
>
> On Thu, Feb 5, 2015 at 12:50 PM, Mike Hearn <mike at plan99.net> wrote:
>
>> I'm imagining myself walking around broadcasting my photo and MAC
>>> address while hucksters push payment requests to me for approval
>>
>>
>> I hate to break it to you, but you broadcast a photo of your face every
>> time you walk outside ;)
>>
>> Bluetooth MAC addresses are random, they aren't useful identifiers. If
>> someone can see you, a face is a far more uniquely identifying thing than a
>> MAC.
>>
>> "Payment spam" might be a problem. I can imagine a wallet requiring that
>> such requests are signed and then spammers can be blacklisted in the usual
>> fashion so they can't push things to your phone anymore. Anyway, a hurdle
>> that can be jumped if/when it becomes an issue.
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/1d6f9332/attachment.html>

2023-06-07 17:29:52

- reply

📅 Original date posted:2015-02-05
📝 Original message:The broadcast is ONLY done when the wallet is in Receive mode. Same as when
the QR code is visible. The use of the *Name* section is specifically so
that a recipient can broadcast their name/handle. Not so the recipient
would broadcast the name of the Sender.
[image: logo]
*Paul Puey* CEO / Co-Founder, Airbitz Inc
+1-619-850-8624 | http://airbitz.co | San Diego
<http://facebook.com/airbitz> <http://twitter.com/airbitz>
<https://plus.google.com/118173667510609425617>
<https://go.airbitz.co/comments/feed/> <http://linkedin.com/in/paulpuey>
<https://angel.co/paul-puey>
*DOWNLOAD THE AIRBITZ WALLET:*
<https://play.google.com/store/apps/details?id=com.airbitz>
<https://itunes.apple.com/us/app/airbitz/id843536046>
On Thu, Feb 5, 2015 at 12:50 PM, Mike Hearn <mike at plan99.net> wrote:
> I'm imagining myself walking around broadcasting my photo and MAC
>> address while hucksters push payment requests to me for approval
>
>
> I hate to break it to you, but you broadcast a photo of your face every
> time you walk outside ;)
>
> Bluetooth MAC addresses are random, they aren't useful identifiers. If
> someone can see you, a face is a far more uniquely identifying thing than a
> MAC.
>
> "Payment spam" might be a problem. I can imagine a wallet requiring that
> such requests are signed and then spammers can be blacklisted in the usual
> fashion so they can't push things to your phone anymore. Anyway, a hurdle
> that can be jumped if/when it becomes an issue.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/8f1e83dc/attachment.html>

2023-06-07 17:29:51

- reply

📅 Original date posted:2015-02-05
📝 Original message:The trust can be considered bootstrapped by visual verification of the address prefix. If we are really concerned about someone jamming a Bluetooth signal in a coffeeshop then the UI can encourage verification of the prefix. Much like how regular Bluetooth requires 'pairing' via entering a 4-6 digit code.
Paul Puey CEO / Co-Founder, Airbitz Inc
619.850.8624 | http://airbitz.co | San Diego
On Feb 5, 2015, at 3:46 PM, Eric Voskuil <eric at voskuil.org> wrote:
On 02/05/2015 03:36 PM, MⒶrtin HⒶboⓋštiak wrote:
>> A BIP-70 signed payment request in the initial broadcast can resolve the
>> integrity issues, but because of the public nature of the broadcast
>> coupled with strong public identity, the privacy compromise is much
>> worse. Now transactions are cryptographically tainted.
>>
>> This is also the problem with BIP-70 over the web. TLS and other
>> security precautions aside, an interloper on the communication, desktop,
>> datacenter, etc., can capture payment requests and strongly correlate
>> transactions to identities in an automated manner. The payment request
>> must be kept private between the parties, and that's hard to do.
>
> What about using encryption with forward secrecy? Merchant would
> generate signed request containing public ECDH part, buyer would send
> back transaction encrypted with ECDH and his public ECDH part. If
> receiving address/amount is meant to be private, use commit protocol
> (see ZRTP/RedPhone) and short authentication phrase (which is hard to
> spoof thanks to commit protocol - see RedPhone)?
Hi Martin,
The problem is that you need to verify the ownership of the public key.
A MITM can substitute the key. If you don't have verifiable identity
associated with the public key (PKI/WoT), you need a shared secret (such
as a secret phrase). But the problem is then establishing that secret
over a public channel.
You can bootstrap a private session over the untrusted network using a
trusted public key (PKI/WoT). But the presumption is that you are
already doing this over the web (using TLS). That process is subject to
attack at the CA. WoT is not subject to a CA attack, because it's
decentralized. But it's also not sufficiently deployed for some scenarios.
e
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/6bf6a7d3/attachment.html>

2023-06-07 17:29:51

- reply

📅 Original date posted:2015-02-05
📝 Original message:The implementation on Airbitz does not encourage or even let a user
broadcast a photo. Just an address prefix and "name/handle". And it's only
broadcast during the Receive request. Not generally while the app is
running although that's up to the implementation.
[image: logo]
*Paul Puey* CEO / Co-Founder, Airbitz Inc
+1-619-850-8624 | http://airbitz.co | San Diego
<http://facebook.com/airbitz> <http://twitter.com/airbitz>
<https://plus.google.com/118173667510609425617>
<https://go.airbitz.co/comments/feed/> <http://linkedin.com/in/paulpuey>
<https://angel.co/paul-puey>
*DOWNLOAD THE AIRBITZ WALLET:*
<https://play.google.com/store/apps/details?id=com.airbitz>
<https://itunes.apple.com/us/app/airbitz/id843536046>
On Thu, Feb 5, 2015 at 1:19 PM, Brian Hoffman <brianchoffman at gmail.com>
wrote:
> This sounds horrible. You could basically monitor anyone with a wallet in
> a highly populated area and track them super easily by doing facial
> recognition. Yes you could photograph people but it's way more burdensome.
> Sorry to go off topic a little.
>
>
> On Feb 5, 2015, at 3:50 PM, Mike Hearn <mike at plan99.net> wrote:
>
> I'm imagining myself walking around broadcasting my photo and MAC
>> address while hucksters push payment requests to me for approval
>
>
> I hate to break it to you, but you broadcast a photo of your face every
> time you walk outside ;)
>
> Bluetooth MAC addresses are random, they aren't useful identifiers. If
> someone can see you, a face is a far more uniquely identifying thing than a
> MAC.
>
> "Payment spam" might be a problem. I can imagine a wallet requiring that
> such requests are signed and then spammers can be blacklisted in the usual
> fashion so they can't push things to your phone anymore. Anyway, a hurdle
> that can be jumped if/when it becomes an issue.
>
>
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming. The Go Parallel Website,
> sponsored by Intel and developed in partnership with Slashdot Media, is
> your
> hub for all things parallel software development, from weekly thought
> leadership blogs to news, videos, case studies, tutorials and more. Take a
> look and join the conversation now. http://goparallel.sourceforge.net/
>
> _______________________________________________
> Bitcoin-development mailing list
> Bitcoin-development at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bitcoin-development
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/0a99ec39/attachment.html>

2023-06-07 17:29:45

- reply

📅 Original date posted:2015-02-05
📝 Original message:Although consumer to merchant is a use case for BLE I would argue that NFC has a higher chance of providing a better user experience in most cases since, at least on Android, a user can tap their phone without even having a wallet running. The URI handler will launch the wallet for them. However a dedicated, user facing, screen can give certainty that the user is connecting to the correct recipient.
1. Because it can show an address prefix
2. It can display the users nickname/handle upon connecting which is only sent to the merchant upon a point to point connection. Not a broadcast.
The Airbitz wallet already does this on the recipient side. A popup shows the most recent person connecting to the recipient.
Paul Puey CEO / Co-Founder, Airbitz Inc
619.850.8624 | http://airbitz.co | San Diego
On Feb 5, 2015, at 3:34 PM, Roy Badami <roy at gnomon.org.uk> wrote:
For peer-to-peer payments, how common do we think that the payment is
of an ad hoc nature rather than to a known contact?
If I want to pay my friends/colleagues/etc over a restaurant table
there's no reason why I couldn't already have their public keys in my
contact list - then it would be pretty straightforward to have a
watertight mechanism where I would know who I was paying. You could
probably even relatively securely bootstrap a key exchange over SMS,
relying only on the contacts already having each other in their
phonebooks.
As for comsumer-to-merchant transactions where the merchant is a
bricks and mortar merchant, IMHO it absolutely has to be "pay that
terminal over there". It's the trust model we all currently use -
whether paying cash or card - and it's the only trust model that works
IMHO (and customers and businesses alike are well aware of the risks
of a fraudster standing behind the counter pretending to be an
employee accepting payment - and by and large are pretty good at
mitigating it). OTOH as we've discussed here before there are many
use cases where the custoemr doesn't actually know or care about the
name of the shop or bar they walked into but is pretty damn sure that
they need to make payment to the person over there behind the counter.
Granted, there are cases taht dont' fall into either of the above -
but they're the cases that are (a) harder to figure out how to
authenticate and consequently (b) the use cases that are going to be
most subject to attempted fraud.
roy
> On Thu, Feb 05, 2015 at 03:02:56PM -0800, William Swanson wrote:
>> On Thu, Feb 5, 2015 at 2:10 PM, Eric Voskuil <eric at voskuil.org> wrote:
>> A MITM can receive the initial broadcast and then spoof it by jamming the
>> original. You then only see one.
>
> You are right, of course. There is no way to make Bluetooth 100%
> secure, since it is an over-the-air technology. You could try securing
> it using a CA or other identity server, but now you've excluded ad-hoc
> person-to-person payments. Plus, you need an active internet
> connection to reach the CA.
>
> You can try using proximity as a substitute for identity, like
> requiring NFC to kick-start the connection, but at that point you
> might as well use QR codes.
>
> This BIP is not trying to provide absolute bullet-proof security,
> since that's impossible given the physical limitations of the
> Bluetooth technology. Instead, it's trying to provide the
> best-possible security given those constraints. In exchange for this,
> we get greatly enhanced usability in common scenarios.
>
> There are plenty of usable, real-world technologies with big security
> holes. Anybody with lock-picking experience will tell you this, but
> nobody is welding their front door shut. The ability to go in and out
> is worth the security risk.
>
> Bluetooth payments add a whole new dimension to real-world Bitcoin
> usability. Do we shut that down because it can't be made perfect, or
> do we do the best we can and move forward?
>
> -William
>
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming. The Go Parallel Website,
> sponsored by Intel and developed in partnership with Slashdot Media, is your
> hub for all things parallel software development, from weekly thought
> leadership blogs to news, videos, case studies, tutorials and more. Take a
> look and join the conversation now. http://goparallel.sourceforge.net/
> _______________________________________________
> Bitcoin-development mailing list
> Bitcoin-development at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bitcoin-development
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/dece93e9/attachment.html>

2023-06-07 17:29:42

- reply

📅 Original date posted:2015-02-05
📝 Original message:So if you picked up the BLE broadcast request. All you know is that
*someone* within 100m is requesting bitcoin at a certain address. Not
necessarily who. The *name* is both optional, and possibly just a *handle*
of the user. If I'm sitting 5 ft away from someone at dinner and wanted to
pay them via BLE, I might see "Monkey Dude" on my list and simply ask him
"is that you?" If so, I send it. If there are two "Monkey Dude's" Then I
have to bother with the address prefix, but not otherwise.
[image: logo]
*Paul Puey* CEO / Co-Founder, Airbitz Inc
+1-619-850-8624 | http://airbitz.co | San Diego
<http://facebook.com/airbitz> <http://twitter.com/airbitz>
<https://plus.google.com/118173667510609425617>
<https://go.airbitz.co/comments/feed/> <http://linkedin.com/in/paulpuey>
<https://angel.co/paul-puey>
*DOWNLOAD THE AIRBITZ WALLET:*
<https://play.google.com/store/apps/details?id=com.airbitz>
<https://itunes.apple.com/us/app/airbitz/id843536046>
On Thu, Feb 5, 2015 at 1:46 PM, Eric Voskuil <eric at voskuil.org> wrote:
> BLE has an advertised range of over 100m.
>
> http://www.bluetooth.com/Pages/low-energy-tech-info.aspx
>
> In the case of mass surveillance that range could most likely be extended
> dramatically by the reviewer. I've seen WiFi ranges of over a mile with a
> strong (not FCC approved) receiver.
>
> WiFi hotspots don't have strong identity or a guaranteed position, so they
> can't be trusted for location.
>
> e
>
> On Feb 5, 2015, at 1:36 PM, Mike Hearn <mike at plan99.net> wrote:
>
> This sounds horrible. You could basically monitor anyone with a wallet in
>> a highly populated area and track them super easily by doing facial
>> recognition.
>>
>
> We're talking about BLE, still? The radio tech that runs in the so called
> "junk bands" because propagation is so poor?
>
> My watch loses its connection to my phone if I just put it down and walk
> around my apartment. I'm all for reasonable paranoia, but Bluetooth isn't
> going to be enabling mass surveillance any time soon. It barely goes
> through air, let alone walls.
>
> Anyway, whatever. I'm just bouncing around ideas for faster user
> interfaces. You could always switch it off or set it to be triggered by the
> presence of particular wifi hotspots, if you don't mind an initial bit of
> setup.
>
> Back on topic - the debate is interesting, but I think to get this to the
> stage of being a BIP we'd need at least another wallet to implement it?
> Then I guess a BIP would be useful regardless of the design issues. The
> prefix matching still feels flaky to me but it's hard to know if you could
> really swipe payments out of the air in practice, without actually trying
> it.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/b562bf05/attachment.html>

2023-06-07 17:29:39

- reply

📅 Original date posted:2015-02-05
📝 Original message:Thanks for CC'ing me Mike. Having trouble receiving maillist list posts.
Even if a user could get the BIP70 URL in the URI, they would still need
internet to access the URL. This BLE spec doesn't preclude BIP70, but can
work with it while still allowing individuals without a certificate to
broadcast a request.
The issue of confused payments becomes less so if the Recipient broadcasts
a name along with the 10 digit public addr prefix. Only if there is a name
conflict will the user have to be concerned with the prefix. The name can
be something like
Mikes Coffee #1 and it can have a "Register #1" at the counter. A customer
facing screen can also show the 10 digit prefix.
[image: logo]
*Paul Puey* CEO / Co-Founder, Airbitz Inc
+1-619-850-8624 | http://airbitz.co | San Diego
<http://facebook.com/airbitz> <http://twitter.com/airbitz>
<https://plus.google.com/118173667510609425617>
<https://go.airbitz.co/comments/feed/> <http://linkedin.com/in/paulpuey>
<https://angel.co/paul-puey>
*DOWNLOAD THE AIRBITZ WALLET:*
<https://play.google.com/store/apps/details?id=com.airbitz>
<https://itunes.apple.com/us/app/airbitz/id843536046>
On Thu, Feb 5, 2015 at 12:28 PM, Mike Hearn <mike at plan99.net> wrote:
> BIP70 requests can be sent over bluetooth as well, as can transactions.
> Bitcoin Wallet can already send money even when offline by doing this. It's
> transparent to the user. I mean original Bluetooth in this context - BLE
> has incredibly tight data constraints and isn't really meant for data
> transfer.
>
> Yes Android Beam has a pretty stupid UI. You can actually tap the devices,
> take them away and then press, but that's not obvious at all. There have
> been new APIs added in recent releases that give more control over this, so
> it's possible we can revisit things and make the UI better these days.
>
> The donation to live performer example is good - there's no issue of
> accidentally paying for someone else in this context as there's only one
> recipient, but many senders.
>
> The issue of confused payments remains in other situations though.
>
> For the coffee shop use case, it'd be nicer (I think) if we aim for a
> Square-style UI where the device broadcasts a (link to) a photo of the user
> combined with a bluetooth MAC. Then the merchant tablet can show faces of
> people in the shop, and can push a payment request to the users device.
> That device can then buzz the user, show a confirmation screen, put
> something on their smart watch etc or just auto-authorise the payment
> because the BIP70 signature is from a trusted merchant. User never even
> needs to touch their phone at all.
>
> On Thu, Feb 5, 2015 at 9:06 PM, Paul Puey <paul at airbitz.co> wrote:
>
>> The BIP70 protocol would preclude individuals from utilizing the P2P
>> transfer spec. It would also require that a Sender have internet
>> connectivity to get the payment protocol info. BLE could enable payment w/o
>> internet by first transferring the URI to from Recipient to Sender. Then in
>> the future, we could sign a Tx and send it over BLE back to the recipient
>> (who would still need internet to verify the Tx). This is an important use
>> case for areas with poor 3G/4G connectivity as I've experience myself.
>>
>> Also, due to Android issues, NFC is incredibly clunky. The URI Sender is
>> required to tap the screen *while* the two phones are in contact. We
>> support NFC the same way Bitcoin Wallet does, but unless the payment
>> recipient has a custom Android device (which a merchant might) then the
>> usage model is worse than scanning a QR code. BLE also allows people to pay
>> at a distance such as for a donation to a live performer. We'll look at
>> adding this to the Motivation section.
>>
>> [image: logo]
>> *Paul Puey* CEO / Co-Founder, Airbitz Inc
>> +1-619-850-8624 | http://airbitz.co | San Diego
>> <http://facebook.com/airbitz> <http://twitter.com/airbitz>
>> <https://plus.google.com/118173667510609425617>
>> <https://go.airbitz.co/comments/feed/> <http://linkedin.com/in/paulpuey>
>> <https://angel.co/paul-puey>
>> *DOWNLOAD THE AIRBITZ WALLET:*
>> <https://play.google.com/store/apps/details?id=com.airbitz>
>> <https://itunes.apple.com/us/app/airbitz/id843536046>
>>
>>
>> From: Andreas Schildbach <andreas at sc...> - 2015-02-05 13:47:04
>>
>> Thanks Paul, for writing up your protocol!
>>
>> First thoughts:
>>
>> For a BIP standard, I think we should skip "bitcoin:" URIs entirely and
>> publish BIP70 payment requests instead. URIs mainly stick around because
>> of QR codes limited capacity. BIP70 would partly address the "copycat"
>> problem by signing payment requests.
>>
>> In your Motivation section, I miss some words about NFC. NFC already
>> addresses all of the usability issues mentioned and is supported by
>> mobile wallets since 2011. That doesn't mean your method doesn't make
>> sense in some situations, but I think it should be explained why to
>> prefer broadcasting payment requests over picking them up via near field
>> radio.
>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Dive into the World of Parallel Programming. The Go Parallel Website,
>> sponsored by Intel and developed in partnership with Slashdot Media, is
>> your
>> hub for all things parallel software development, from weekly thought
>> leadership blogs to news, videos, case studies, tutorials and more. Take a
>> look and join the conversation now. http://goparallel.sourceforge.net/
>> _______________________________________________
>> Bitcoin-development mailing list
>> Bitcoin-development at lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/bitcoin-development
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/52c1b824/attachment.html>

2023-06-07 17:29:38

- reply

📅 Original date posted:2015-02-05
📝 Original message:The BIP70 protocol would preclude individuals from utilizing the P2P
transfer spec. It would also require that a Sender have internet
connectivity to get the payment protocol info. BLE could enable payment w/o
internet by first transferring the URI to from Recipient to Sender. Then in
the future, we could sign a Tx and send it over BLE back to the recipient
(who would still need internet to verify the Tx). This is an important use
case for areas with poor 3G/4G connectivity as I've experience myself.
Also, due to Android issues, NFC is incredibly clunky. The URI Sender is
required to tap the screen *while* the two phones are in contact. We
support NFC the same way Bitcoin Wallet does, but unless the payment
recipient has a custom Android device (which a merchant might) then the
usage model is worse than scanning a QR code. BLE also allows people to pay
at a distance such as for a donation to a live performer. We'll look at
adding this to the Motivation section.
[image: logo]
*Paul Puey* CEO / Co-Founder, Airbitz Inc
+1-619-850-8624 | http://airbitz.co | San Diego
<http://facebook.com/airbitz> <http://twitter.com/airbitz>
<https://plus.google.com/118173667510609425617>
<https://go.airbitz.co/comments/feed/> <http://linkedin.com/in/paulpuey>
<https://angel.co/paul-puey>
*DOWNLOAD THE AIRBITZ WALLET:*
<https://play.google.com/store/apps/details?id=com.airbitz>
<https://itunes.apple.com/us/app/airbitz/id843536046>
From: Andreas Schildbach <andreas at sc...> - 2015-02-05 13:47:04
Thanks Paul, for writing up your protocol!
First thoughts:
For a BIP standard, I think we should skip "bitcoin:" URIs entirely and
publish BIP70 payment requests instead. URIs mainly stick around because
of QR codes limited capacity. BIP70 would partly address the "copycat"
problem by signing payment requests.
In your Motivation section, I miss some words about NFC. NFC already
addresses all of the usability issues mentioned and is supported by
mobile wallets since 2011. That doesn't mean your method doesn't make
sense in some situations, but I think it should be explained why to
prefer broadcasting payment requests over picking them up via near field
radio.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/77152aa7/attachment.html>

2023-06-07 17:29:37

- reply

📅 Original date posted:2015-02-05
📝 Original message:Airbitz has developed and implemented a method for communicating a bitcoin
URI across Bluetooth (BLE) or any other P2P, mid range, wireless, broadcast
medium. The currently documented implementation is available in our iOS and
Android mobile wallet (updated Android version with BLE coming in about 1
week). We would like to have the BIP pulled into Github for review and
discussion. Here is the current BIP:
BIP: TBD
Title: P2P Wireless URI transfer
Authors: Thomas Baker <tom’at’airbitz.co>, Paul Puey <paul’at’airbitz.co>
Contributors: Joey Krug <joeykrug’at’gmail.com>
Status: proposal
Type: Standards Track
Created: 2015-01-12
Table of Contents
-
Abstract
-
Motivation
-
Specification
-
Compatibility
-
Examples
-
References
Abstract
This is a protocol for peer-to-peer wireless transfer of a URI request
using an open broadcast or advertisement channel such as Bluetooth,
Bluetooth Low Energy, or WiFi Direct.
Motivation
There are disadvantages for a merchant (requester) and customer (sender) to
exchange a URI request using QR codes that can be eliminated by using
wireless broadcast or advertisements.
Current QR code scan method to transfer a request URI from merchant
(Requester) to customer (Sender) is cumbersome. A usual scenario is a
merchant with a POS terminal for order entry and a separate tablet for
transacting payments with bitcoin, and a customer with a smartphone. After
the order is entered, the merchant enters payment request information into
the tablet, generates the QR code representing the URI, and presents this
to the customer. The customer prepares to scan the QR code with their
smartphone by maneuvering the camera to the tablet. The tablet screen must
be relatively clean, point at the customer, and held steady. The smartphone
camera lens must be clean, point at the tablet screen, come into range, and
held steady to focus and wait for a QR scan. Environmental conditions such
as bright outdoor sunlight, indoor spot lights, or significant distance
between QR code and camera can create difficult and cumbersome experiences
for users.
Using a wireless local broadcast allows the merchant to just enter the
payment and wait. The tablet and smartphone are not maneuvered to align in
any way. The customer observes broadcast listings, selects the appropriate
one from possible simultaneous broadcasts from other POS stations nearby,
examines the URI request details such as amount, and decides whether to
send funds, initiating a bitcoin network transfer. The merchant and
customer then receive the transaction confirmations and are done with the
sale. Merchant and customer devices are kept private and secured in their
own possession.
The URI and other broadcast identification (Joe’s Grill #1) only contain
public information. However, a copycat broadcaster acting as MITM might
duplicate the broadcast simultaneously as the merchant, attempting to lure
the customer to send funds to the copycat. That attack is mitigated with
this broadcast method because of the partial address in the broadcast.
Specification
Requester generates a bitcoin URI request of variable length, and a limited
descriptive identifier string. Requester then broadcasts the URI’s partial
public address (<paddress>) plus identifier (<id>) over a publicly visible
wireless channel.
Sender scans for broadcasts on their device, examines and selects the
desired request by the identifier and partial address. This connects a data
channel to Requester.
Requester sends full URI back over the data channel.
Sender device ensures <paddress> is part of the full URI public address and
checks the full address integrity. Checking the broadcast and full URI
integrity prevents a copycat device within range from copying the partial
address and fooling the customer into sending funds to the copycat instead.
Below is a description of the protocol through Bluetooth Smart (Low Energy).
Requestor Sender - Bitcoin transaction roles
Peripheral Central - Bluetooth GAP definitions
Mode Mode
1 |------------->| - Requestor Advertises partial bitcoin: URI +
Name
| ... |
2 |<-------------| - Subscribe then send sender's Name, requesting
a response
3 |------------->| - ACK
4 |<-------------| - request Read Characteristic from peripheral
5 |------------->| - Sender receives full bitcoin: URI
1.
Peripheral advertises over a service UUID a BLE extended advertisement
with a Scan Response containing the partial address of a bitcoin URI and a
Name, any plain text. The entire response is limited to 26 characters. The
first 10 make up the first 10 characters of the bitcoin URI public address
where to send bitcoin, and must be present. The remaining characters are
any plain text such as “The Habit 1” or “Starbucks-Reg 1”, more human
readable information. The partial address serves as a check against a
nearby attacker who may try to lure a Sender into sending payment to a
separate wallet by advertising a similar Scan Response but cannot replicate
a public address with the same leading 10 characters and different trailing
characters.
2.
When the Central scans the advertisement, it may display the Scan
Response in a human readable listing using the two pieces of information.
If Central chooses this advertisement to receive the full request, it then
subscribes to the service and writes the characteristic (a second UUID)
with it’s own name, or a blank if not sending a name, to the Peripheral.
3.
Peripheral gets a characteristic write request of the Central’s name,
and acknowledges the receipt by sending a server response.
4.
Central receives a characteristic write (from the response) and
immediately requests the entire bitcoin URI by issuing a read request on
that characteristic.
5.
Peripheral receives the read request and sends the entire bitcoin URI
over that characteristic up to 512 bytes.
This ends the proposed specification as the bitcoin URI transfer is
complete. The Sender would then normally confirm the request and decide
whether to initiate the fund transfer.
Compatibility
There are no prior BIPs covering this.
Examples
Airbitz iOS Bluetooth Low Energy to Bluetooth Low Energy request transfer.
References
[image: logo]
*Paul Puey* CEO / Co-Founder, Airbitz Inc
+1-619-850-8624 | http://airbitz.co | San Diego
<http://facebook.com/airbitz> <http://twitter.com/airbitz>
<https://plus.google.com/118173667510609425617>
<https://go.airbitz.co/comments/feed/> <http://linkedin.com/in/paulpuey>
<https://angel.co/paul-puey>
*DOWNLOAD THE AIRBITZ WALLET:*
<https://play.google.com/store/apps/details?id=com.airbitz>
<https://itunes.apple.com/us/app/airbitz/id843536046>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150205/a00a4c8a/attachment.html>