It looks like my draft hasn't been approved by the mailing list so if
anyone would like to read it it's also on Gist:

https://gist.github.com/toby/9e71811d387923a71a53

Luke - As stated in the Github thread, I totally understand where you're
coming from but the fact is people *will* encode data on the blockchain
using worse methods. For all of the reasons that OP_RETURN was a good idea
in the first place, it's a good idea to support it in PaymentRequests.

As for keyless - there's no way (that I know of) to construct a transaction
with a zero value OP_RETURN in an environment without keys since the
Payment Protocol is what defines the method for getting a transaction from
a server to a wallet. You can make a custom transaction and execute it in
the same application but without Payments there's no way to move
transactions between two applications. You need to build the transaction
where you execute it and thus need a key.



On Mon, Jan 25, 2016 at 6:24 PM, Luke Dashjr <luke@dashjr.org> wrote:

> This is a bad idea. OP_RETURN attachments are tolerated (not encouraged!)
> for
> the sake of the network, since the spam cannot be outright stopped. If it
> could be outright stopped, it would not be reasonable to allow OP_RETURN.
> When
> it comes to the payment protocol, however, changing the current behaviour
> has
> literally no benefit to the network at all, and the changes proposed herein
> are clearly detrimental since it would both encourage spam, and potentially
> make users unwilling (maybe even unaware) participants in it. For these
> reasons, *I highly advise against publishing or implementing this BIP,
> even if
> the later mentioned issues are fixed.*
>
> On Tuesday, January 26, 2016 1:02:44 AM Toby Padilla wrote:
> > An example might be a merchant that adds the hash of a plain text invoice
> > to the checkout transaction. The merchant could construct the
> > PaymentRequest with the invoice hash in an OP_RETURN and pass it to the
> > customer's wallet. The wallet could then submit the transaction,
> including
> > the invoice hash from the PaymentRequest. The wallet will have encoded a
> > proof of purchase to the blockchain without the wallet developer having
> to
> > coordinate with the merchant software or add features beyond this BIP.
>
> Such a "proof" is useless without wallet support. Even if you argue it
> could
> be implemented later on, it stands to reason that a scammer will simply
> encode
> garbage if the wallet is not checking the proof-of-purchase upfront. To
> check
> it, you would also need further protocol extensions which are not included
> in
> this draft.
>
> > Merchants and Bitcoin application developers benefit from this BIP
> because
> > they can now construct transactions that include OP_RETURN data in a
> > keyless environment. Again, prior to this BIP, transactions that used
> > OP_RETURN (with zero value) needed to be constructed and executed in the
> > same software. By separating the two concerns, this BIP allows merchant
> > software to create transactions with OP_RETURN metadata on a server
> without
> > storing public or private Bitcoin keys. This greatly enhances security
> > where OP_RETURN applications currently need access to a private key to
> sign
> > transactions.
>
> I don't see how this has any relevance to keys at all...
>
> > ## Specification
> >
> > The specification for this BIP is straightforward. BIP70 should be fully
> > implemented with two changes:
> >
> > 1. Outputs where the script is an OP_RETURN and the value is zero should
> be
> > accepted by the wallet.
> > 2. Outputs where the script is an OP_RETURN and the value is greater than
> > zero should be rejected.
> >
> > This is a change from the BIP70 requirement that all zero value outputs
> be
> > ignored.
>
> This does not appear to be backward nor even forward compatible. Old
> clients
> will continue to use the previous behaviour and transparently omit any
> commitments. New clients on the other hand will fail to include commitments
> produced by old servers. In other words, it is impossible to produce
> software
> compatible with both BIP 70 and this draft, and implementing either would
> result in severe consequences.
>
> > As it exists today, BIP70 allows for OP_RETURN data storage at the
> expense
> > of permanently destroyed Bitcoin.
>
> It is better for the spammers to lose burned bitcoins, than have a way to
> avoid them.
>
> Luke
>