Re: [bitcoin-dev] Packaged Transaction Relay

[Anthony Towns <aj@erisian.com.au>]

On Sat, Oct 08, 2022 at 12:58:35PM -0700, Eric Voskuil via bitcoin-dev wrote:
> > > Protocol cannot be defined on an ad-hoc basis as a "courtesy"
> > BIPs are a courtesy in the first place.
> I suppose if you felt that you were the authority then this would be your
> perspective. 

You seem to think that I'm arguing courtesy is not a good thing, or that
we couldn't use more of it?

If it helps: courtesy is a good thing, and we could use more of it.

> The BIP process was created by Amir specifically because Bitcoin standards
> were being discussed and developed behind closed doors.

It definitely bothers me that Bitcoin development is not being discussed
out in the open as much as I would like, and to counter that, I try to
encourage people to post their ideas to this list, and write them up as
a BIP; and likewise try to do both myself as well.

But how much value do you think anyone's actually getting from posting
their development ideas to this list these days? Do you really think
people reading your mail will be more inspired to discuss their ideas
in the open, or that they'll prefer to get in a room with their friends
and allies, and close the doors so they can work in peace?

> > There's no central authority to enforce some particular way of doing
> > things.
> As if reaching consensus with other people implies a singular authority.

Reaching consensus with other people doesn't require putting a document
in some particular github repo, either. Which is a good thing, or the
people in control of that repo would become that singular authority.

> > If you think that the version restriction should be part of the BIP,
> > why not do a pull request? The BIP is still marked as "Draft".
> I did not implement and ship a deviation from the posted proposal.

You think BIP 155 is suboptimal, and would rather see it changed, no?

But if you won't put any effort into changing it (and how much effort do
you think a PR to change it document it as being gated by version 70016
would be?), why do you imagine the people who are happy with the BIP as
it is would put any effort in?

> > > I doubt that anyone who's worked with it is terribly fond of Bitcoin's
> > > P2P  protocol versioning. I've spent some time on a proposal to
> > > update it, though it hasn't been a priority. If anyone is 
> > > interested in collaborating on it please contact me directly.

"contact me directly" and wanting something other than standards "being
discussed and developed behind closed doors" seems quite contradictory
to me.

(In my opinion, a big practical advantage of doing things in public is
that it's easy for people to contribute, even if it's not a particular
priority, and that it's also easy for someone new to take over, if the
people previously working on it decide they no longer have time for that
particular project)

> > Bottlenecking a proposal on someone who doesn't see it as a priority
> > doesn't seem smart?
> I didn't realize I was holding you up. As far as I've been able to gather,
> it hasn't been a priority for anyone. Yet somehow, on the same day that I
> posted the fact that I was working on it, it became your top priority.

It's not my top priority; it's just that writing a BIP and posting
it publicly is fundamentally no harder than writing an email to
bitcoin-dev. So since I'm willing to do one, why waste anyone's time by
not also doing the other? Would've been even easier if I'd remembered
Suhas had already written up a draft BIP two years ago...

And if I'm going to suggest you should post a patch to a BIP you think
is flawed, then not drafting a BIP to improve on a practice I think is
flawed would be pretty hypocritical, no?

(I didn't read what you said to imply that you were working on it,
just that you'd spent time thinking about it, were interested, and
might do more if people contacted you. If you have been working on
it, why not do so in public? You already have a public bips fork at
https://github.com/evoskuil/bips/branches -- how about just pushing your
work-in-progress there?)

(Ah, I also see now that I did contact you in Dec 2020/Jan 2021 on this
topic, but never received a response. Apologies; the above was meant as
a general statement in favour of just collaborating in public from the
start for the practical advantages I outline above, not a personal dig)

> > Here's what I think makes sense:
> > https://github.com/ajtowns/bips/blob/202210-p2pfeatures/bip-
> > p2pfeatures.mediawiki
> Looks like you put about 10 minutes of thought into it. In your words, BIPs
> are a courtesy - feel free to do what you want.

So, you wrote a lot of stuff after this, but unless I missed it, it
didn't include any substantive criticism of the proposal, or specific
suggestions for changing it, or even any indication why you would have
any difficulty supporting/implementing it in the software you care about.

> Your contributions notwithstanding, you are in no place to exhibit such
> arrogance.

I don't understand what you think is arrogant about posting a public
proposal about how I think things should work, even if I had only put
10 minutes thought into it. If that *is* arrogance, I guess I think we
could use more of it, as well as more courtesy...

(I mean, if I *had* only spent 10 minutes on it then posting it to the
list might be a waste of everyone else's time; but I could still put it in
my bips fork, and blog/tweet about it, or mention it on irc or similar,
which is at least still a public proposal. And in this case I've been
thinking about it on and of since 2020, and do know that other people
share similar views, so I don't think it's wasting people's time)

Cheers,
aj