sipa changed the topic of #bitcoin-wizards to: This channel is for discussing theoretical ideas with regard to cryptocurrencies, not about short-term Bitcoin development | http://bitcoin.ninja/ | This channel is logged. | For logs and more information, visit http://bitcoin.ninja
kenshi84 has quit [Read error: Connection reset by peer]
Alina-malina has quit [Ping timeout: 276 seconds]
<fluffypony>
andytoshi: if you're still looking for a streaming thing I can check what we used last time
sausage_factory has joined #bitcoin-wizards
blackwraith has quit [Ping timeout: 260 seconds]
Alina-malina has joined #bitcoin-wizards
bildramer1 is now known as bildramer
espes__ has joined #bitcoin-wizards
sausage_factory has quit [Ping timeout: 256 seconds]
kenshi84 has joined #bitcoin-wizards
Alina-malina has quit [Ping timeout: 260 seconds]
rusty2 has quit [Ping timeout: 240 seconds]
face has quit [Ping timeout: 240 seconds]
face has joined #bitcoin-wizards
kenshi84 has quit [Remote host closed the connection]
<yoleaux>
Elliptic curves over Q(i) are potentially automorphic – totallydisconnected
domwoe has quit [Remote host closed the connection]
waxwing has quit [Quit: Leaving]
DTZUZU has joined #bitcoin-wizards
aalex has quit [Quit: Connection reset by beer]
waxwing has joined #bitcoin-wizards
blackwraith has joined #bitcoin-wizards
JayDugger has quit [Ping timeout: 252 seconds]
<bsm117532>
I'm gonna need an interpreter on that one...
arubi has joined #bitcoin-wizards
arubi has quit [Client Quit]
<kanzure>
roasbeef: streaming y/n?
<instagibbs>
kanzure, there will be a video posted according to moli, posted at least
<moli_>
instagibbs, actually i got an email reply from someone at Stanford that said, "We're recording the talks, and they'll be on the conference website next week."
<instagibbs>
err right, what i meant, thanks
<kanzure>
alright.
<moli_>
yea, i'm a bit disappointed but i can wait :)
* bsm117532
realizes CSV and Segwit will activate simultaneously on Litecoin. (if they activate) I guess BIP9 will be stress tested too!
sausage_factory has joined #bitcoin-wizards
<gmaxwell>
litecoin does not use BIP9
<sipa>
how do the trigger segwit?
<sipa>
*they
<gmaxwell>
through frightening cluelessness they thought it wouldn't work for them.
<gmaxwell>
(they managed to conclude it was incompatible with merged mining because -- derp)
<gmaxwell>
sipa: they almost didn't.. cerated an ISM activation but then got it wrong in their original version.
belcher has quit [Ping timeout: 255 seconds]
blackwraith has quit [Ping timeout: 276 seconds]
Noldorin has quit [Ping timeout: 255 seconds]
belcher has joined #bitcoin-wizards
instagibbs has quit [Read error: Connection reset by peer]
instagibbs has joined #bitcoin-wizards
rusty2 has joined #bitcoin-wizards
Giszmo has joined #bitcoin-wizards
instagibbs has quit [Ping timeout: 255 seconds]
Giszmo has quit [Client Quit]
rusty2 has quit [Quit: Leaving.]
rusty2 has joined #bitcoin-wizards
blackwraith has joined #bitcoin-wizards
sausage_factory has quit [Ping timeout: 252 seconds]
<petertodd>
re: Arvinds bitcoin instability due to no block reward paper, did anyone here personally inform him of the prior art on this topic before?
<petertodd>
I'm at his talk right now, and will be mentioning that I personally submitted a pull-req related to this issue on mar 2013; he claimed at the beginning of the talk that the issue wasn't known until his paper
<petertodd>
I think we can both make a case that is completely wrong, and if someone has already informed him of that prior art, he's being dishonest
<adiabat>
Lots of people told him "hey you realize you're assuming unbounded blocksize, right?" (I think)
<petertodd>
ok, that's a start
<petertodd>
did anyone inform him of, for isntance, my pull-req?
<bsm117532>
It looks like they forward ported scrypt mining to bitcoin 0.13 rather than backporting segwit (+ bazillions of other changes)
<brg444>
arvind's paper makes very weak assumption about fee variance under a block size limit, this is pretty much where his research falls apart imho
<fluffypony>
brg444: agreed
arubi has joined #bitcoin-wizards
blackwraith has joined #bitcoin-wizards
<brg444>
my response to his paper was widely circulated so I was a bit puzzled to see that he would be presenting it at the conf. as I wrote the paper is, if anything, only useful at pointing out that a hard block size limit is absolutely essential
<petertodd>
brg444: well, what I'm more concerned about right now is he's being dishonest - he started his talk by claiming it wasn't previously known, then tried to say "oh, no, I meant not widely analysed"
sausage_factory has quit [Ping timeout: 276 seconds]
<gmaxwell>
"he claimed at the beginning of the talk that the issue wasn't known until his paper
<gmaxwell>
"
<gmaxwell>
thats absurd!
<gmaxwell>
I mean, my very first post on the modern block size debate bring up the subject.
<gmaxwell>
I think it's plausable enough that he didn't personally know of it, for sure.
<fluffypony>
if he didn't personally know of it then that should be his statement
<brg444>
from the paper.... "There is some work on understanding the market for transaction fees and its relation to the block size (i.e. what fees will users have to pay in order for transactions to be included in a block?) [14, 11, 21, 18]."
frstTimer has joined #bitcoin-wizards
<brg444>
I'm concerned that he would cite Peter Rizun's work [21] and not bother consult people here when researching previous work
<gmaxwell>
The second is that when subsidy has fallen well below fees, the incentive
jtimon has quit [Ping timeout: 240 seconds]
<gmaxwell>
to move the blockchain forward goes away. An optimal rational miner
<gmaxwell>
would be best off forking off the current best block in order to capture
<gmaxwell>
its fees, rather than moving the blockchain forward, until they hit
<gmaxwell>
the maximum. That's where the "backlog" comment comes from, since when
<gmaxwell>
there is a sufficient backlog it's better to go forward. I'm not aware
<gmaxwell>
of specific research into this subquestion; it's somewhat fuzzy because
<gmaxwell>
of uncertainty about the security model. If we try to say that Bitcoin
<gmaxwell>
should work even in the face of most miners being profit-maximizing
<gmaxwell>
instead of altruistically-honest, we must assume the chain will not
<gmaxwell>
more forward so long as a block isn't full. In reality there is more
<gmaxwell>
altruism than zero; there are public pressures; there is laziness, etc.
<gmaxwell>
I posted this on reddit in reply to their paper.
<gmaxwell>
---
<gmaxwell>
and even that was just summarizing discussions well know going back eons.
<gmaxwell>
I'm happy to see people researching the subject. Though that particular research was not very interesting in my view because it did not respect the state of the art in our understanding.
<gmaxwell>
(e.g. didn't assume anti-sniping and a backlog-- and just consider the impact of fee variance)
<adlai>
it seems to be driven by precisely the same currents that hosted just yesterday what they alleged as the "first academic blockchain conference"
<gmaxwell>
what event is this?
<brg444>
gmaxwell: stanford's I presume
<adlai>
it was at some private university (in israel it's backwards, those are the shitty ones)
<brg444>
oh sorry
* adlai
tries to find a link for this conf that isn't a fb page
<gmaxwell>
petertodd: I think it's a pretty sad state were in when many academics are just unaware of things that us in industry take for granted.
<kanzure>
nobody reads
<kanzure>
nor ask around, evidently
<gmaxwell>
well I think, if I'm not mistaken, his group spends a lot of time face to face with ethereum people, may give a false sense of awareness.
<adlai>
this seems to link talks if anybody's interested. my impression was that this was a "blockchain buzzwords 101" for clueless students </offTopic>
<gmaxwell>
petertodd: I listed the mitigations we've considered, and which are implemented already.
rusty2 has quit [Ping timeout: 260 seconds]
<gmaxwell>
there is a follow up reply from me at the same link where I provide hyperlinks
<gmaxwell>
(including places where you discussed it with Gun previously)
cbits has quit [Ping timeout: 255 seconds]
<gmaxwell>
a bit absurd that we've gone people in the same group like gun antagonizing over the block size stuff, then later someone in the same group "discovers" one of the concerns we've made clear all along, ... and then proposes perpetual inflation to fix it. :-/
<gmaxwell>
petertodd: as far as fee variance goes, here are some actual traces from the network for fees available immedately before a new block and instantly after: https://people.xiph.org/~greg/temp/fee_avail3.png
<gmaxwell>
So at least during periods of higher use, we're already doing a pretty good job of not having a huge incentive to snipe
pero0 has joined #bitcoin-wizards
<gmaxwell>
I wish I had data for 1 minute after a block, which would have looked even better.
<gmaxwell>
actually I guess I do have that data.. will just have to process it.
pero has quit [Ping timeout: 264 seconds]
pero0 has quit [Quit: Leaving]
pero0 has joined #bitcoin-wizards
<Jeremy_Rand[m]>
gmaxwell: AFAIK BIP9 is incompatible with AuxPoW (at least the specific AuxPoW spec that Namecoin and Dogecoin use). AuxPoW abuses the block nVersion field in ways that break with BIP9.
cbits has joined #bitcoin-wizards
<Jeremy_Rand[m]>
Namecoin is planning to do a hardfork to fix this; I'm not sure what the other AuxPoW chains are planning to do.
draynium has quit [Ping timeout: 264 seconds]
pero0 is now known as pero
Sosumi has quit [Quit: Bye]
<gmaxwell>
Jeremy_Rand[m]: no, it's not
<gmaxwell>
it incompatible for use in the child chain (as are any version changing upgrade mechenisms, including ISM) but not in the parent.
<gmaxwell>
From virtually everything's perspective BIP9 is the same as ISM. both require the version field to be set to specific values.
<gmaxwell>
and litecoin is not merged mined with anything
<Jeremy_Rand[m]>
gmaxwell: if the parent chain uses BIP9 bit 8 (I think that's the bit), the child chain interprets that as the parent block having the AuxPoW bit set. And AuxPoW blocks are invalid if their parent block has the AuxPoW bit set.
<Jeremy_Rand[m]>
gmaxwell: so BIP9 usage in the parent chain bricks the child chain once bit 8 is used
<gmaxwell>
Jeremy_Rand[m]: great this has nothing to do with SW or anything else. And it is _still_ not less compatible with ISM, which would also set bit 8 eventually.
<sipa>
yeah, so just never pick bit 8 in VB
<Jeremy_Rand[m]>
gmaxwell: I think we may be talking past each other. I'm not talking about SW specifically, only saying that BIP9 does pose problems for AuxPoW. I didn't realize that you were specifically talking about there being no problems for the parent chain.
<Jeremy_Rand[m]>
Apologies for miscommunication there
<gmaxwell>
and as a result of this confusion litecoin almost had a massive network incident.
<gmaxwell>
Jeremy_Rand[m]: BIP 9 doesn't touch any bits in the version field. It's instructions for creating proposals that touch bits.
<gmaxwell>
(well okay, it sets one of the MSB bits, to be pedantic but otherwise it doesn't)
<gmaxwell>
well no apology needed, it was a nice test for cluelessness of people working on litecoin... which hopefully didn't have any side effects.
<Jeremy_Rand[m]>
gmaxwell: The AuxPoW spec assumes that only the lowest 8 bits of nVersion are used; bit 8 and the top 16 bits are abused by AuxPoW (the top 16 bits are the chain ID)
<Jeremy_Rand[m]>
So, I think it's fair to say that BIP9 exacerbates the stupidity of that particular design choice by AuxPoW
<gmaxwell>
Jeremy_Rand[m]: so does namecoin no longer exist? BIP29 has been set on ~every bitcoin block for most of a year.
<gmaxwell>
er BIT 29
<gmaxwell>
(man, took three retypes to not write BIP)
<Jeremy_Rand[m]>
gmaxwell: Namecoin hasn't activated BIP9 because doing so is a hardfork because of bit 29 (which we're planning to do).
<Jeremy_Rand[m]>
The only way that the top 16 bits in the parent chain would break Namecoin is if those 16 bits are equal to 0x0001 (the Namecoin chain ID).
<gmaxwell>
...
<gmaxwell>
I want to drill my head.
* gmaxwell
looks for a power drill
<Jeremy_Rand[m]>
yeah, the AuxPoW spec is a pile of fail. I wish Luke-Jr's spec were deployable so we could switch to that.
<Jeremy_Rand[m]>
What we're going to end up doing in our hardfork is make AuxPoW mandatory (so the AuxPoW bit isn't used for anything), and move the Chain ID into the nonce field (which is unused right now for AuxPoW blocks)
<gmaxwell>
Jeremy_Rand[m]: BIP 9 doesn't "activate", it isn't in and of itself a consensus change. Bitcoin has used BIP9 multiple times now. Namecoin still exists. QED Use of BIP9 does not necessarily cause any problems for mergedmined things.
<gmaxwell>
QED litecoin did not need to crap up their clone of segwit and cause another perpetual divergence from the Bitcoin chain, and the almost take a consensus disaster for it.
<gmaxwell>
yes, making auxpower mandatory is obvious. it's not like you need to actually merge mine anything to use it. and also a softfork to make it mandatoy. your hardfork would just be not being tripped up by things in the parent chain.
<Jeremy_Rand[m]>
gmaxwell: I don't disagree with you that it's stupid for a parent chain to make any changes due to BIP9 and AuxPoW. I already reached out to the Dogecoin guys many months ago to tell them exactly what the issues were
<Jeremy_Rand[m]>
Dogecoin was planning to hardfork to fix it on their end; I have no clue why Litecoin would have needed to change anything at all
<gmaxwell>
whew.
<Jeremy_Rand[m]>
gmaxwell: and yes, I agree with your assessment that mandatory auxpow is a nobrainer. And yes, that part is a softfork.
cbits has quit [Ping timeout: 255 seconds]
<Jeremy_Rand[m]>
gmaxwell: it's not just the parent chain tripping things up that requires a hardfork for Namecoin. In Namecoin, the top 16 bits of nVersion must be exactly 0x0001 (because those bits are the chain ID)
<Jeremy_Rand[m]>
(or rather, they must be 0x0001 if it's an AuxPoW block)
<Jeremy_Rand[m]>
So removing that constraint is a hardfork
<Jeremy_Rand[m]>
I mean, we hypothetically could get away with changing the bits used by BIP9 so that we reserve all those bits. But that's a big enough patchset against upstream Bitcoin Core that we prefer to just hardfork
chjj has quit [Ping timeout: 255 seconds]
ipwn has joined #bitcoin-wizards
sausage_factory has joined #bitcoin-wizards
blackwraith has quit [Ping timeout: 240 seconds]
draynium has joined #bitcoin-wizards
waxwing has quit [Ping timeout: 248 seconds]
saintromuald has quit [Ping timeout: 264 seconds]
blackwraith has joined #bitcoin-wizards
draynium has quit [Ping timeout: 248 seconds]
sausage_factory has quit [Ping timeout: 260 seconds]
frstTimer has quit []
cbits has joined #bitcoin-wizards
blackwraith has quit [Ping timeout: 264 seconds]
q4 has quit [Ping timeout: 245 seconds]
blackwraith has joined #bitcoin-wizards
cbits_ has joined #bitcoin-wizards
saintromuald has joined #bitcoin-wizards
ipwn has quit [Ping timeout: 240 seconds]
blackwraith has quit [Ping timeout: 240 seconds]
cbits has quit [Ping timeout: 240 seconds]
<petertodd>
gmaxwell: thanks - I think you can also make an argument that fees will be relatively flat so long as people can tolerate some delays in their transactions; be good to better formalize that argument
chjj has joined #bitcoin-wizards
<petertodd>
gmaxwell: I'd definitely be interested in evidence that arvind was actually aware of the existing mitigations; he seemed to imply that in his response to me, but that's not concrete
<roasbeef>
kanzure: no official streaming afaik, charging phone rn, then might start streaming in like 30 mins or so
aguycalled has joined #bitcoin-wizards
cbits has joined #bitcoin-wizards
jtimon has joined #bitcoin-wizards
cbits_ has quit [Ping timeout: 256 seconds]
CheckDavid has quit [Quit: Connection closed for inactivity]
shesek has quit [Ping timeout: 260 seconds]
cbits_ has joined #bitcoin-wizards
cbits has quit [Ping timeout: 260 seconds]
cbits has joined #bitcoin-wizards
shesek has joined #bitcoin-wizards
cbits_ has quit [Ping timeout: 256 seconds]
<gmaxwell>
petertodd: if he says he wasn't, I'm sure he wasn't. Still a problem that people think they're researching bitcoin when they're not. :)
cbits_ has joined #bitcoin-wizards
<gmaxwell>
petertodd: I think there are a lot of interesting and open questions in that space. E.g. I've suggested a fee forwarding scheme where miners should, using their mempools, estimate the income for the next miner, and forward part of the excess of an even split... but how does that game theory work out?
cbits has quit [Ping timeout: 240 seconds]
draynium has joined #bitcoin-wizards
<bsm117532>
petertodd, gmaxwell I often see anger from you guys about academics and their citations. Just realize that academics read journals. There you find carefully laid out, refereed ideas. Citing reddit or some random mailing list would be like citing graffiti on a bathroom wall.
<bsm117532>
Academics simply don't look for these sources because if you do, you end up wading through miles of shit.
<bsm117532>
Just politely point out to them your prior work. They'll catch on. But I wouldn't expect them to start reading 4-year old mailing lists or reddit posts anytime soon.
<bsm117532>
And I don't think it's helpful to call anyone "dishonest". They just live in a different echo chamber...
<gmaxwell>
bsm117532: stop.
Davasny has quit [Read error: Connection reset by peer]
<gmaxwell>
We're not talking about reddit or the bitcoin development mailing list (which is hardly random).
<gmaxwell>
We're talking about the bitcoin software running on the network and on people's hands all over the world.
<gmaxwell>
Being unaware of varrious proposals is one thing, -- but if you want to be unaware of the Bitcoin system you shouldn't write papers that claim that they're about Bitcoin (which is a _perfectly_ fine thing to do!)
<gmaxwell>
bsm117532: as far as dishonest goes, petertodd wrote "if someone has already informed him". Clearly not. There isn't any dishonesty but just a lack of communication.
<bsm117532>
^^^ yep
<gmaxwell>
As far as journals go, they reject our work because it's "boring" industrial stuff. And a lot of what they print is miles of shit too. There is no solving that, I can't blame someone for not knowing about a proposal anywhere... but I can blame people for writing about "Bitcoin" but not actually describing the existing system.
<bsm117532>
People are generally lazy about referencing. Happens in all fields.
<bsm117532>
And most *papers* are shit. Academics are incentivized to write as many papers as possible. You get what you incentivize.
<gmaxwell>
right.
<gmaxwell>
in our case, there is very little incentive to publish in academic venues; even ignoring they often prefer spherical cows that no one will ever use over industrially important advances.
<gmaxwell>
ignoring that they*
<gmaxwell>
though we try. Even also ignoring that its a lot of extra work and conforming to those norms often makes the material more opaque (unless written with a kind of special genius that few have access to)
<kanzure>
let's just pipe irc logs into a pdf if that's what they want
<bsm117532>
There are a lot of problems with the academic pipeline. But one thing they do better than IRC/mailing lists is peer review.
<gmaxwell>
bsm117532: not so sure about that.
<gmaxwell>
sometimes, but its spotty.
<kanzure>
bsm117532: that's objectively false; see pubpeer for evidence.
<bsm117532>
This is an append-only medium, and the burden of finding things in it grows with time.
<gmaxwell>
Leger happy published the monero ringct scheme where you can steal everything. People in this channel spotted the weakness, not reviewers.
<gmaxwell>
Ledger*
<kanzure>
pubpeer has a habit of showing how all the "peer preview" propaganda is a hue farce
<kanzure>
*hue
<kanzure>
.. *huge
<gmaxwell>
I think the error is confusiong _any external review_, which is an amazing benefit, with the academic process. But the work we do in the Bitcoin industry _is_ reviewed.
<bsm117532>
kanzure: that one is new to me...what's the point of pubpeer?
<kanzure>
pubpeer shows all the drama going on under the hood in peer review; science is not some sort of magic consensus land, it's often politics and should be destroyed for it
<gmaxwell>
bsm117532: think of it this way, how good could peer review be when the paper in discussion here made it to print without the authors knowing about the mitigations _already deployed_ in Bitcoin. Your argument above that papers can't find citations to random mailing list posts is also a statement that peer review doesn't work.
<bsm117532>
gmaxwell: The academic system is deeply flawed. The second half of the failure point out is finding reviewer time. The pressure to publish is also a pressure to rubber-stamp.
<kanzure>
(oops i meant that page-- not that comment)
<sipa>
I think the academic peer review process is better at resulting in material that has a procedural way of learning about.
<gmaxwell>
yes, sure better in established fields.
<gmaxwell>
But the incentives are all still screwy.
<sipa>
The actual quality of the material that gets published of course depends on the quality of the review.
<bsm117532>
sipa: that's my point about an "append-only log". The academic system does do some collation now and again. (Especially, "review articles")
<sipa>
But by design, I think that the academic mechanism of publishes has merits - even though it may fail.
<gmaxwell>
Everything has merits, at least when you ignore the costs. :)
<sipa>
(and there is a ton of legacy in that world that needs improving, no doubt)
<gmaxwell>
bsm117532: as an aside, in many things I've published the peer review primarily provided more (excellent) citations.
<kanzure>
i think the more frightening reality is that nobody has a good proposal for how to organize science in a way that doesn't suck as much as journals and papers
<bsm117532>
kanzure: let me tell you about my *other* project...
<sipa>
gmaxwell: ok, i guess i'm saying there is a huge merit in "us" (widely speaking) getting things published in academic venues.
<bsm117532>
sipa: Yes!!!!
<sipa>
and that won't be easy
<bsm117532>
Why?
<gmaxwell>
net of the costs, far from clear to me.
<sipa>
bsm117532: i'm familiar with what it takes
<gmaxwell>
bsm117532: because we've been trying (and many of us have published before in other fields)
<bsm117532>
This is why I joined Ledger...I see that need.
rusty2 has joined #bitcoin-wizards
* bsm117532
has 2 interns this summer and is trying to push them towards a publication in collaboration with one of their CS professors...
waxwing has joined #bitcoin-wizards
<gmaxwell>
Ledger makes the situation worse. Run by highly politiced parties with a serious axe to grind... and also just putting an academic polish on another echo chamber isn't a clear value. As mentioned, ledger peer review didn't find the gaping hole in monero's cryptosystem. (and sadly, no other venue would likely have taken that paper no matter how well it was written, because it doesn't fit with the
<gmaxwell>
common hot technology that no one will ever use. :P )
cbits_ has quit [Ping timeout: 264 seconds]
<gmaxwell>
We could certantly stand up a journal of bitcoin wizards and publish all our own work. But there is no need for the pretext, we can already publish things without permission. :)
<bsm117532>
There's a definite value in massaging an idea into it's most presentable form though.
cbits has joined #bitcoin-wizards
<sipa>
absolutely agree with that, but i'm not sure that ledger will make it reach the most useful reviewers
<bsm117532>
Heh you guys are probably some of the most useful reviewers. IIRC the most useful people here have declined for political reasons. :-/
<sipa>
understandably
<bsm117532>
So...
<bsm117532>
Self fulfilling prophecy...
<sipa>
my personal inclination to publish things is to get review not from the people here
<kanzure>
saying they "declined" is bullshit-- just read the logs, wtf
<kanzure>
oh, logs not in pd format. so... :).
<sipa>
bsm117532: self-fulfilling prophecy... i don't think Ledger is very useful. I do think that academic publishing in general is useful.
<bsm117532>
Let's face it, peer review can be replaced with a very simple website. But some organization is still necessary: (1) polish your presentation (2) collect relevant targeted responses -- instead of a streaming chatlog.
<kanzure>
streaming chatlog is highly relevant, why are you here if this is irrelevant
<gmaxwell>
bsm117532: the reviews had already been published and were ignored, as kanzure says.
<kanzure>
*not in pdf format
cbits has quit [Ping timeout: 240 seconds]
<gmaxwell>
lol
<kanzure>
it would be an interesting study on how you don't actually get "sniped" for chatting about your work
<bsm117532>
I find reading chatlogs to be an exercise in frustration. I like *participating* though :-P
<kanzure>
lots of folks in science are super worried about that. but it turns out that nobody reads.
<uiuc-slack1>
<amiller> Let's make a better open peer review system then, what would be the minimum to do
<kanzure>
all content must be submitted in prolog
<uiuc-slack1>
<amiller> The ledger process involves an editor picking reviewers, in CS conferences the PC chairs assign reviewers but the pool of reviewers is chosen ahead of time and public before ppl submit
<uiuc-slack1>
<amiller> Maybe just a forum where people submit prepritns, and then reviewers get assigned some how?
<kanzure>
you can't force people to do good review, it's not going to work.
cbits has joined #bitcoin-wizards
<uiuc-slack1>
<amiller> If people want to submit reviews unprompted, those should all be included, and for papers no one wants to review maybe some can volunteer to be assigned based on expertise or randomly
<kanzure>
measurements of expertise is going to fail because it will end up politicized