Re: [bitcoin-dev] BIP Number Request: Open Asset

[Luke Dashjr <luke@dashjr.org>]

On Thursday, May 26, 2016 2:50:26 AM Nicolas Dorier via bitcoin-dev wrote:
>   Author: Flavien Charlon <flavien@charlon.net>

Is he the author of this BIP, or merely the protocol described in it?
Would it perhaps make sense to include yourself in the author list?

> The ID of an asset is the RIPEMD-160 hash of the SHA-256 hash of the
> output script referenced by the first input of the transaction that
> initially issued that asset (<code>script_hash =
> RIPEMD160(SHA256(script))</code>). An issuer can reissue more of an
> already existing asset as long as they retain the private key for that
> asset ID. Assets on two different outputs can only be mixed together
> if they have the same asset ID.

Quite a bit ugly, giving a meaning to an input's pubkey script like that.
But more problematically: how can this work with other pubkey scripts? 
Particularly relevant now that this old script format is being deprecated.

Another possible problem is that I don't see a way to provably guarantee an 
asset issuance is final.

> Transactions that are not recognized as an Open Assets transaction are
> considered as having all their outputs uncolored.

And the assets attached to its inputs are destroyed? Or?

> If multiple parsable PUSHDATA opcodes exist in the same output, the
> first one is used, and the other ones are ignored.
> 
> If multiple valid marker outputs exist in the same transaction, the
> first one is used and the other ones are considered as regular
> outputs.

Is it intentional that the first case is "parsable", and the second "valid"?
I think these need to be better specified; for example, it is not so clear how 
to reach if the OAP version number is something other than 1: is that 
parsable? valid?

> ! Asset quantity count || A
> [https://en.bitcoin.it/wiki/Protocol_specification#Variable_length_integer
> var-integer] representing the number of items in the <code>asset quantity
> list</code> field. || 1-9 bytes
> 
> |-
> 
> ! Asset quantity list  || A list of zero or more
> [http://en.wikipedia.org/wiki/LEB128 LEB128-encoded] unsigned integers
> representing the asset quantity of every output in order (excluding the
> marker output). || Variable

What determines the asset id? How would one issue and/or transfer multiple 
asset ids in the same transaction?

> The marker output is always uncolored.

What if I have a transaction with 5 outputs, the marker output at position 3, 
and all 4 other outputs are to receive assets? Does the marker output get 
skipped in the list (ie, the list is 4 elements long) or must it be set to 
zero quantity (ie, the list is 5 elements long)?

> Inputs are seen as a sequence of asset units, each having an asset ID.
> Similarly, outputs are seen as a sequence of asset units to be
> assigned an asset ID. These two sequences are built by taking each
> input or output in order, each of them adding a number of asset units
> equal to their asset quantity. The process starts with the first input
> of the transaction and the first output after the marker output.
> 
> After the sequences have been built, the asset ID of every asset unit
> in the input sequence is assigned to the asset unit at the same
> position in the output sequence until all the asset units in the
> output sequence have received an asset ID. If there are less asset
> units in the input sequence than in the output sequence, the marker
> output is considered invalid.
> 
> Finally, for each transfer output, if the asset units forming that
> output all have the same asset ID, the output gets assigned that asset
> ID. If any output is mixing units with more than one distinct asset
> ID, the marker output is considered invalid. Outputs with an asset
> quantity of zero are always considered uncolored.

I don't understand this.

> # This approach uses the recommended way to embed data in the Blockchain
> (OP_RETURN), and therefore does not pollute the UTXO.

Embedding data is not recommended at all. It seems a better way to have done 
this would be to put the info in an OP_DROP within a P2SH or witness script.

> # The whole cryptographic infrastructure that Bitcoin provides for
> securing the spending of outputs is reused for securing the ability to
> issue assets. There is a symmetry between ''an address + private key''
> as a way to spend Bitcoins, and ''an address + private key'' as a way
> to issue assets.

Addresses are not used for spending bitcoins, only for receiving them. The way 
this BIP uses inputs' pubkey script is extremely unusual and probably a bad 
idea.

> # Reissuing more of an existing asset is easy and can be done quickly
> and at no cost (except for the transaction fee) as long as the issuer
> retains the private key for the asset ID.

As I understand it, this would require address reuse to setup, which is not 
supported behaviour and insecure.

> For backward compatibility reasons, we consider than an older client
> is allowed to see a colored output as uncolored.

How is this compatible? Won't an older client then accidentally destroy 
assets?

Luke