Lloyd Fournier [ARCHIVE] on Nostr: 📅 Original date posted:2021-04-04 📝 Original message:On Wed, 10 Mar 2021 at ...
📅 Original date posted:2021-04-04
📝 Original message:On Wed, 10 Mar 2021 at 11:20, Lloyd Fournier <lloyd.fourn at gmail.com> wrote:
> Hi Andrew & all,
>
> I've been working with PSBTs for a little while now. FWIW I agree with the
> change of removing the global tx and having the input/output data stored
> together in the new unified structures.
>
> One thing I've been wondering about is how output descriptors could fit
> into PSBTs. They are useful since they allow you to determine the maximum
> satisfaction weight for inputs so you can properly align fees as things get
> added. I haven't seen any discussion about including them in this revision.
> Is it simply a matter of time before they make it into a subsequent PSBT
> spec or is there something I'm missing conceptually?
>
Sipa replied to me off list some time ago and explained what I was missing.
PSBTs have all the information you could want from a descriptor already.
For example the maximum satisfaction weight can be determined from the
witness/redeem script (I had forgot these fields existed). Therefore
descriptors are more useful in higher level applications while PSBTs are
useful for communicating with signing devices. Therefore there is no reason
for PSBTs to support descriptors.
LL
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20210405/911a07fe/attachment-0001.html>
📝 Original message:On Wed, 10 Mar 2021 at 11:20, Lloyd Fournier <lloyd.fourn at gmail.com> wrote:
> Hi Andrew & all,
>
> I've been working with PSBTs for a little while now. FWIW I agree with the
> change of removing the global tx and having the input/output data stored
> together in the new unified structures.
>
> One thing I've been wondering about is how output descriptors could fit
> into PSBTs. They are useful since they allow you to determine the maximum
> satisfaction weight for inputs so you can properly align fees as things get
> added. I haven't seen any discussion about including them in this revision.
> Is it simply a matter of time before they make it into a subsequent PSBT
> spec or is there something I'm missing conceptually?
>
Sipa replied to me off list some time ago and explained what I was missing.
PSBTs have all the information you could want from a descriptor already.
For example the maximum satisfaction weight can be determined from the
witness/redeem script (I had forgot these fields existed). Therefore
descriptors are more useful in higher level applications while PSBTs are
useful for communicating with signing devices. Therefore there is no reason
for PSBTs to support descriptors.
LL
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20210405/911a07fe/attachment-0001.html>