[bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Emil Engler]

[-- Attachment #1: Type: text/plain, Size: 1641 bytes --]

Hello dear mailing list subscribers.
Before I'll explain my idea here, I need to define a term first

'address':
When I use the terms address, pubkey, etc., I mean the same: The Base58
string

Ok now let's get into it:
As you should know, sending bitcoins to an address more than once is a
very bad approach.
In my opinion the problem why so many people are still doing this is
because of the term 'address' which is used in lots of wallets,
implementations, BIP 21 and so on. It is a design issue.
With the term 'address' most people identify things that are fixed and
don't change really often (e.g postal address, IP address [depends on
provider], Domain, E-Mail address, ...).
Because of this most people compare bitcoin addresses with e-mail
addresses and use this address to send the recipient money multiple times.

My suggestion would be to change the term address in wallets, the URI
scheme and so on to something of the following options by a
Informational/Process BIP:

* Payment Password
* Transaction Password
* ...

The guideline for the term should indicate that it is:
* temporary
* Something that identifies the recipient

I've chosen 'password' because they can be used as a pseudonym to
identify a person.
This is already used in stuff like bank transfers where something like
the transaction id should be used as the purpose or at universities
there are student numbers.
The first is probably a better example because student numbers aren't
temporary.

What do you think? Should I write a BIP for this or use another term?
Feedback is most welcome :)

Greetings,
Emil Engler


[-- Attachment #2: pEpkey.asc --]
[-- Type: application/pgp-keys, Size: 3199 bytes --]

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Luke Dashjr]

On Saturday 05 October 2019 21:57:48 Emil Engler via bitcoin-dev wrote:
> Hello dear mailing list subscribers.
> Before I'll explain my idea here, I need to define a term first
>
> 'address':
> When I use the terms address, pubkey, etc., I mean the same: The Base58
> string

But a pubkey is not a Base58 string, and fundamentally different from an 
address. An address identifies the recipient and the purpose of the payment; 
a pubkey does not. The pubkey remains with the UTXO; an address does not.

> Ok now let's get into it:
> As you should know, sending bitcoins to an address more than once is a
> very bad approach.
> In my opinion the problem why so many people are still doing this is
> because of the term 'address' which is used in lots of wallets,
> implementations, BIP 21 and so on. It is a design issue.
> With the term 'address' most people identify things that are fixed and
> don't change really often (e.g postal address, IP address [depends on
> provider], Domain, E-Mail address, ...).
> Because of this most people compare bitcoin addresses with e-mail
> addresses and use this address to send the recipient money multiple times.

That problem would require using a different term than "address" to address.
A BIP is unlikely to do the job (though it may help).

> My suggestion would be to change the term address in wallets, the URI
> scheme and so on to something of the following options by a
> Informational/Process BIP:
>
> * Payment Password
> * Transaction Password
> * ...

Neither the address nor pubkey are a password...

Some possible alternative terms would be "invoice id", "payment token", etc.

> The guideline for the term should indicate that it is:
> * temporary
> * Something that identifies the recipient
>
> I've chosen 'password' because they can be used as a pseudonym to
> identify a person.
> This is already used in stuff like bank transfers where something like
> the transaction id should be used as the purpose or at universities
> there are student numbers.
> The first is probably a better example because student numbers aren't
> temporary.
>
> What do you think? Should I write a BIP for this or use another term?
> Feedback is most welcome :)
>
> Greetings,
> Emil Engler

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Emil Engler]

[-- Attachment #1: Type: text/plain, Size: 1133 bytes --]

> But a pubkey is not a Base58 string, and fundamentally different from an 
> address. An address identifies the recipient and the purpose of the payment; 
> a pubkey does not. The pubkey remains with the UTXO; an address does not.

I don't know much about this topic in Bitcoin so far, but I meant the
address :)

> That problem would require using a different term than "address" to address.

I meant this the whole time, sorry for misunderstandings.

> A BIP is unlikely to do the job (though it may help).

Why? They are specifications and implementations need them (even if some
do not implement everything).
Also BIP 21 would need an alternative/separate parameter with the new term.

> Neither the address nor pubkey are a password...

In Germany at least (where I live), you need to provide a 'Password'
(the german translation of thin word of course) in the purpose of the
transaction if you pay with a bank transfer. They are used as IDs but I
agree, ID is definitely a better suffix.

> Some possible alternative terms would be "invoice id", "payment token", etc.

ACK

Greetings, Emil Engler

[-- Attachment #2: pEpkey.asc --]
[-- Type: application/pgp-keys, Size: 3199 bytes --]

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Chris Belcher]

This is an excellent idea and I hope something like this happens.

I've had the idea of using an intermediate name to make the transition
easier, for example "Bitcoin address" becomes "Bitcoin invoice address"
which after 10 years becomes "Bitcoin invoice" (or "Bitcoin invoice").

"Invoice" would also work well because Lightning uses the name invoice
for the object which is used to receive payments. So it's easy to
imagine that the GUI text input presented to users can be called
"invoice" and users can put both kinds of invoices there leaving the
wallet to easily figure out whether it makes an on-chain transaction or
a Lightning Network transaction.

Changing a commonly-used name like this could be very hard, but the
gains in terms of privacy are immense.

On 05/10/2019 22:57, Emil Engler via bitcoin-dev wrote:
> Hello dear mailing list subscribers.
> Before I'll explain my idea here, I need to define a term first
> 
> 'address':
> When I use the terms address, pubkey, etc., I mean the same: The Base58
> string
> 
> Ok now let's get into it:
> As you should know, sending bitcoins to an address more than once is a
> very bad approach.
> In my opinion the problem why so many people are still doing this is
> because of the term 'address' which is used in lots of wallets,
> implementations, BIP 21 and so on. It is a design issue.
> With the term 'address' most people identify things that are fixed and
> don't change really often (e.g postal address, IP address [depends on
> provider], Domain, E-Mail address, ...).
> Because of this most people compare bitcoin addresses with e-mail
> addresses and use this address to send the recipient money multiple times.
> 
> My suggestion would be to change the term address in wallets, the URI
> scheme and so on to something of the following options by a
> Informational/Process BIP:
> 
> * Payment Password
> * Transaction Password
> * ...
> 
> The guideline for the term should indicate that it is:
> * temporary
> * Something that identifies the recipient
> 
> I've chosen 'password' because they can be used as a pseudonym to
> identify a person.
> This is already used in stuff like bank transfers where something like
> the transaction id should be used as the purpose or at universities
> there are student numbers.
> The first is probably a better example because student numbers aren't
> temporary.
> 
> What do you think? Should I write a BIP for this or use another term?
> Feedback is most welcome :)
> 
> Greetings,
> Emil Engler
> 
> 
> _______________________________________________
> bitcoin-dev mailing list
> bitcoin-dev@lists.linuxfoundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
> 

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Emil Engler]

[-- Attachment #1: Type: text/plain, Size: 700 bytes --]

* Sorry if this mail was sent multiple times, my E-Mail client went crazy *

Thanks for all your feedback.
I came to the decision to write a BIP for this, even if it might not be
implemented by many wallets, a standardization is never wrong and this
would be the first step in the correct direction for better on-chain
privacy.

However currently we still need a good term for the 'address' replacement.

The current suggestions are:
* Invoice ID
* Payment Token
* Bitcoin invoice (address)
* Bitcoin invoice (path)

Because of the LN term invoice I really like the term 'Bitcoin Invoice'
by Chris Belcher.

So how do find a consensus about these terms?

Greetings
Emil Engler

[-- Attachment #2: pEpkey.asc --]
[-- Type: application/pgp-keys, Size: 3199 bytes --]

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Lloyd Fournier]

[-- Attachment #1: Type: text/plain, Size: 2030 bytes --]

Hi Thread,

This may not be the most practical information, but there actually did
exist an almost perfect analogy for Bitcoin addresses from the ancient
world: From wikipedia https://en.wikipedia.org/wiki/Bulla_(seal)

"Transactions for trading needed to be accounted for efficiently, so the
clay tokens were placed in a clay ball (bulla), which helped with
dishonesty and kept all the tokens together. In order to account for the
tokens, the bulla would have to be crushed to reveal their content. This
introduced the idea of impressing the token onto the wet bulla before it
dried, to insure trust that the tokens hadn't been tampered with and for
anyone to know what exactly was in the bulla without having to break it."

You could only use the bulla once because it had to be destroyed in order
to get the tokens out! I think there are even examples of bulla with a kind
of "signature" on them (an imprint with the seal of a noble family etc).

"send me a Bitcoin bulla" has a nice ring to it!

Sincerely,

LL





On Fri, Oct 11, 2019 at 2:44 AM Emil Engler via bitcoin-dev <
bitcoin-dev@lists.linuxfoundation.org> wrote:

> * Sorry if this mail was sent multiple times, my E-Mail client went crazy *
>
> Thanks for all your feedback.
> I came to the decision to write a BIP for this, even if it might not be
> implemented by many wallets, a standardization is never wrong and this
> would be the first step in the correct direction for better on-chain
> privacy.
>
> However currently we still need a good term for the 'address' replacement.
>
> The current suggestions are:
> * Invoice ID
> * Payment Token
> * Bitcoin invoice (address)
> * Bitcoin invoice (path)
>
> Because of the LN term invoice I really like the term 'Bitcoin Invoice'
> by Chris Belcher.
>
> So how do find a consensus about these terms?
>
> Greetings
> Emil Engler
> _______________________________________________
> bitcoin-dev mailing list
> bitcoin-dev@lists.linuxfoundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
>

[-- Attachment #2: Type: text/html, Size: 3081 bytes --]

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Karl-Johan Alm]

I've proposed bitcoin invoice for awhile now. See
https://twitter.com/kallewoof/status/1165841566105079808

I like bitcoin invoice address into bitcoin address as proposed by Chris.


On Fri, Oct 11, 2019 at 12:45 AM Emil Engler via bitcoin-dev
<bitcoin-dev@lists.linuxfoundation.org> wrote:
>
> * Sorry if this mail was sent multiple times, my E-Mail client went crazy *
>
> Thanks for all your feedback.
> I came to the decision to write a BIP for this, even if it might not be
> implemented by many wallets, a standardization is never wrong and this
> would be the first step in the correct direction for better on-chain
> privacy.
>
> However currently we still need a good term for the 'address' replacement.
>
> The current suggestions are:
> * Invoice ID
> * Payment Token
> * Bitcoin invoice (address)
> * Bitcoin invoice (path)
>
> Because of the LN term invoice I really like the term 'Bitcoin Invoice'
> by Chris Belcher.
>
> So how do find a consensus about these terms?
>
> Greetings
> Emil Engler
> _______________________________________________
> bitcoin-dev mailing list
> bitcoin-dev@lists.linuxfoundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Ariel Lorenzo-Luaces]

[-- Attachment #1: Type: text/plain, Size: 5986 bytes --]

I would propose a term different than all the others mentioned so far.

I propose Funding Codes.

The word funding can be used as a verb or a noun and this works for the nature of Bitcoin transactions. Funding can be seen as what someone needs to do with the code as well as referring to it as the code that has funding once bitcoins have been transfered to it "give me a funding code".

The word code is fitting since codes are what addresses ultimately describe, the signature script, a piece of code. When speaking about the lightning transaction graph it's easy to talk about transactions making bitcoins flow from code to code, each code different for a different purpose "funding is sent from this code to that code" and "funding is kept in this code for 144 blocks".

- Payment tokens feel like they themselves hold the value and can be transfered but anyone can generate as many payment tokens as they desire. This name conflicts with other cryptocurrencies that focus their blockchain on payments and refer to their currency as tokens.

- Invoices are problematic because they imply that they hold bitcoins and they specify an amount. However addresses don't specify any amounts while they themselves can be included inside a real invoice. I think it is wrong to imply that the "thing" we are sending money to is temporary, because it isn't. Lightning channels can stay open forever until closed but money doesn't stay in an invoice for any amount of time.

- I actually prefer Addresses over both Payment Tokens and Invoices. It's actually very natural to send something to an address and an address can hold something for a long time. However addresses fall short due to people only having one. This makes them think that they have to memorize a bitcoin address. There are also all the other reasons people have mentioned.

The word code fits well in the divide between technical and non-technical people. To the layman a code is just a string of characters and that is what we can visually see and check and compare when one of these funding codes are transfered between two parties "does your finding code end with xyz?". To programmers code is something that can be executed which is exactly what addresses are. Especially ones with complicated logic and lots of conditions "this funding code can only be unlocked by Alice or Bob plus Charlie or Dave after 1000 blocks".

Funding codes work even better when thinking about self executing smart contracts "they are funded and running code with their funds". Contracts don't make sense at all when it's an autonomous thing that didn't strike any contract with anyone. Contracts should only be referred to the transactions people send to funding codes or "smart" codes.

Addresses also fail when transferring between two of your own wallets because it doesn't make sense for someone to have so many addresses but it does make sense for someone to have many codes.

Lightning already has "funding addresses" and "funding transactions". With schnorr and taproot coming it will be possible to create more complex situations and they all need funding codes so that funds can be sent to it and be locked up in the code (sigscript).

One criticism might be that funding codes sound like they are funding something which doesn't appear to be true. But indeed they are! Funding codes fund a situation, a setup. The common setup is that you can unlock them at any time. Other setups demand multi-party coordination. The funding code is what funds all these setups.

Funding codes are also two words and three syllables which is great because using only one word is not descriptive enough and using four or more syllables is way too long. Having the second word "code" makes for easy abbreviation when the conversation is already about Bitcoin "which code did you send them to?"

People will naturally use funding code and bitcoin code interchangeably. This is a good thing because bitcoin is a type of fund, so there is no contradiction. The more common term should still be funding code because there is more than one type of "code" in Bitcoin

Most importantly funding codes sound good when spoken. This goes for both singular and plural.

"First a receiver must generate a funding code to have a sender fund it with their  from their own funding code which had been previously funded"

Cheers
Ariel Lorenzo-Luaces



On Oct 10, 2019, 7:20 PM, at 7:20 PM, Karl-Johan Alm via bitcoin-dev <bitcoin-dev@lists.linuxfoundation.org> wrote:
>I've proposed bitcoin invoice for awhile now. See
>https://twitter.com/kallewoof/status/1165841566105079808
>
>I like bitcoin invoice address into bitcoin address as proposed by
>Chris.
>
>
>On Fri, Oct 11, 2019 at 12:45 AM Emil Engler via bitcoin-dev
><bitcoin-dev@lists.linuxfoundation.org> wrote:
>>
>> * Sorry if this mail was sent multiple times, my E-Mail client went
>crazy *
>>
>> Thanks for all your feedback.
>> I came to the decision to write a BIP for this, even if it might not
>be
>> implemented by many wallets, a standardization is never wrong and
>this
>> would be the first step in the correct direction for better on-chain
>> privacy.
>>
>> However currently we still need a good term for the 'address'
>replacement.
>>
>> The current suggestions are:
>> * Invoice ID
>> * Payment Token
>> * Bitcoin invoice (address)
>> * Bitcoin invoice (path)
>>
>> Because of the LN term invoice I really like the term 'Bitcoin
>Invoice'
>> by Chris Belcher.
>>
>> So how do find a consensus about these terms?
>>
>> Greetings
>> Emil Engler
>> _______________________________________________
>> bitcoin-dev mailing list
>> bitcoin-dev@lists.linuxfoundation.org
>> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
>_______________________________________________
>bitcoin-dev mailing list
>bitcoin-dev@lists.linuxfoundation.org
>https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev

[-- Attachment #2: Type: text/html, Size: 7069 bytes --]

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Emil Engler]

[-- Attachment #1: Type: text/plain, Size: 1745 bytes --]

> This may not be the most practical information, but there actually did exist an almost perfect analogy for Bitcoin addresses from the ancient world: From wikipedia https://en.wikipedia.org/wiki/Bulla_(seal)

I personally do not like the term bulla, it might be a perfect analogy
but I personally don't believe that this term is well known. Tbh I
didn't knew what a 'bulla' was before.

> I propose Funding Codes.

I'm neutral on this. With code I associate something that is slightly
more permanent like a code to unlock your mobile phone or a code to
unlock your bike if you know what I mean. These kind of codes change
sometimes but not as often as a bitcoin address (should).
I also agree that Payment Tokens might confuse with other currencies and
block chain.

> 
> - Invoices are problematic because they imply that they hold bitcoins and they specify an amount. However addresses don't specify any amounts while they themselves can be included inside a real invoice. I think it is wrong to imply that the "thing" we are sending money to is temporary, because it isn't. Lightning channels can stay open forever until closed but money doesn't stay in an invoice for any amount of time.

What is with 'Bitcoin Invoice Address'?
This is the best of both worlds because it implies the temporary factor
with 'invoice' and the way that you send something to something.

Also, this is more a personal opinion but I thunk that 'funding' implies
more to donate to something. I think this could lead to misunderstandings.

To summarize it up, here are the following suggested terms:
* Invoice ID
* Payment Token
* Bitcoin invoice (address)
* Bitcoin invoice (path)
* Bulla
* Funding code

Greetings,
Emil Engler

[-- Attachment #2: pEpkey.asc --]
[-- Type: application/pgp-keys, Size: 3199 bytes --]

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Marco Falke]

I also like the "bitcoin invoice address" term by Chris. Invoice is a
common term and easily translatable into other languages.

Marco

Re: [bitcoin-dev] BIPable-idea: Consistent and better definition of the term 'address'

[Emil Engler]

[-- Attachment #1: Type: text/plain, Size: 528 bytes --]

As the idea got positive feedback, I've written the BIP.
The draft is available here:
https://github.com/bitcoin/bips/pull/856

Am 17.10.19 um 15:23 schrieb Marco Falke via bitcoin-dev:
> I also like the "bitcoin invoice address" term by Chris. Invoice is a
> common term and easily translatable into other languages.
> 
> Marco
> _______________________________________________
> bitcoin-dev mailing list
> bitcoin-dev@lists.linuxfoundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
> 

[-- Attachment #2: pEpkey.asc --]
[-- Type: application/pgp-keys, Size: 3199 bytes --]