Tadge Dryja
@tdryja
Followers
15,397
Following
26
Media
43
Statuses
900
Individual Contributor
Cambridge, MA
Joined October 2015
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
Francia
• 602472 Tweets
Macron
• 470876 Tweets
#HouseOfTheDragon
• 405407 Tweets
Rhaenys
• 195109 Tweets
メディア
• 190604 Tweets
#GranHermano
• 188936 Tweets
Bautista
• 141126 Tweets
Emma
• 140490 Tweets
#loveislandusa
• 105000 Tweets
#NXTHeatwave
• 95024 Tweets
#HOTD
• 93387 Tweets
TBNW
• 84143 Tweets
Aegon
• 71938 Tweets
Vhagar
• 68960 Tweets
Serena
• 53798 Tweets
Daemon
• 45577 Tweets
Jana
• 44587 Tweets
Uruguayo
• 41957 Tweets
Marcos
• 39868 Tweets
Dudu
• 39062 Tweets
FURIA ES LEYENDA
• 37632 Tweets
Kenny
• 37508 Tweets
Ethan
• 36226 Tweets
Sunfyre
• 28937 Tweets
Bowman
• 22677 Tweets
Kordell
• 20578 Tweets
Criston Cole
• 19398 Tweets
Kaylor
• 16021 Tweets
Devers
• 15059 Tweets
Jace
• 13628 Tweets
Daia
• 13018 Tweets
Kendall
• 10641 Tweets
@KimDotcom
The Lightning Network is not patented.
Source: Invented it w/
@jcp
, and it's free to use, modify, improve.
9
40
426
I’m excited to share today that I’m joining the
@Lightspark
team to work on building out Bitcoin and LN to be even more powerful forms of money and payments.
#Lightspark
#Bitcoin
#LightningNetwork
51
23
386
I wore a bitcoin core t-shirt today at
#Bitcoin2022
(along w/
@achow101
). Quite a few people asked, "What's bitcoin core?"
It's... bitcoin. I guess people added "core" because it was confusing.
"So you work for Bitcoin Core?"
24
16
345
So signal is putting some kind of coin in the client. I took a look at their "whitepaper". It's 141 pages of, umm...
24
27
335
You cannot still, never have been able to, and never will be able to, "withdraw BTC on other networks".
Securing Bitcoin also involves calling out hostile redefinition of Bitcoin.
33
52
309
Why the excitement for "Digital Identity"? Bitcoin's strength is we can pay / be paid and not care about people's names or "identities".
39
72
281
A few months ago, I read about
@Lightspark
, a new company whose goal is to “explore, build and extend the capabilities and utility of Bitcoin.”
“Huh.” I thought, “That’s also been my goal for the last decade or so.”
25
14
293
Really frustrating to see laser-eyesers applauding an attack on mainnet LND.
Sure, fixing bugs makes bitcoin stronger. Sure, it's the currency of enemies.
But bitcoin and LN are not magically invincible / inevitable. Attacks don't help.
50
20
220
I've posted Utreexo, a new paper about bitcoin scalability, to IACR ePrint.
Thanks to those who helped with this work and looking forward to criticisms and getting and implementation running.
12
75
218
Just released v0.1 of utreexo software & wrote about it here:
take a look, test it out, and write some crash reports :)
It's been great working with the other utreexo developers so far & looking forward to more people working on it!
4
66
195
The problem with scams isn't even the scams.
It's that you start thinking real things are scams, and start thinking scams are real.
14
68
178
Pay no heed to those who define Bitcoin as "the chain with the most work, whatever the rules", as they are just along for the ride.
10
44
159
@_marcinja
Their PDF: 950886 bytes
Your tweet: 86 bytes
99.99% lossless compression ratio. Your weissman score is off the charts.
2
1
167
None of these people were dumb. They'd just never heard of it. Some of the people I explained it to wanted to download it. Some wanted to start help coding it.
Hopefully this helps.
Huge conference, so many people... but really it's all just a program you run on your computer.
5
6
157
A time between blocks of 85 minutes happens every 34 days or so (assuming no difficulty changes, etc)
>>> (1/math.exp(-85/10))/144
34.13033916874399
21
20
152
Here's the "user-vs-miner Bitcoin compromise" (UVMSF):
Do what we want and we'll let you keep sha256.
11
45
138
@petertoddbtc
@BitPay
"verges"?? Nah this is straight up malware, trying to trick people into installing software which will result in financial loss.
6
43
135
This isn't just about threatening bitcoin; this is threatening all open source software, one of the most amazing and useful (and free!) ideas of the last few decades.
14
40
128
New post about the progress of utreexo, with myself
@kcalvinalvinn
and
@dergoegge
. Getting towards full nodes with no DB.
6
38
126
So sounds like the libbitcoin guy thinks that
head -c 32 /dev/urandom | sha256sum
and
date +%s%N | sha256sum
are the same thing. And changing the former to the latter in wallet key generation code is cool.
Stay far, far away.
13
21
123
Blog post about the alpha release of lit, the lightning network software I'm working on at MIT DCI
3
63
113
Uhhh... is this a new monthly (in expectation) column from coindesk?
20
12
104
Giving a talk at
#crypto2018
in ~an hour. New stuff, will be fun!
Before going, I've had to tell a couple people they've been doing this conference since 1981.
#CryptoMeansCryptography
6
18
103
People at MIT who are interested in this stuff! I'm teaching a class with
@neha
:
Mondays & Wednesdays at 10AM.
People interested but not around here:
There will be videos (though probably not live); problem sets also public.
5
29
106
Don't be caught off-guard by overflow errors - upgrade to a 32-bit price ticker today!
2
8
93
My video studio setup for the messari
#mainnet2020
conference. Pretty different from a stage in a big hotel conference room.
9
2
92
@nic__carter
I'm thinking more and more that it's got to be intentional. It is absolutely trivial to build a seed generator into a wallet:
cat /dev/urandom | tr -dc 'A-Z9' | fold -w 81 | head -n 1
done.
It's past Hanlon's razor for me; I assume malice.
7
19
87
[Headline] Wealthy, powerful men endorse leaderless cyber currency. Masses jubilant.
9
2
88
Problem here: sha-256 is a cryptographic hash function, preventing professor Green from understanding the answer.
I suggest use of a collision-friendly non cryptographic hash function, if a suitable one can be found.
@matthew_d_green
@Peace_Bruv
SHA-256 hash of the answer is 9a5cefc2aea372f1625a6cf49874ff39baf90f0238ce54d6b73afc0aceb60637.
So? Will you comment on that summary?
7
0
30
4
8
90
MIT Bitcoin Expo Airdrop: everyone gets their ticket cost refunded. Fun!
Also I'm speaking here in 45 min. I should finish my slides :)
3
7
88
This machine serves 2 or 3 TB of bitcoin blocks a month, so serving another ubiquitous piece of data should be no problem
2
7
80
TIL:
I'm ded ☠️
This guy's fixing LN "design errors"
With Lightning 1.1 we will be taking a big step away from 1.0. Interestingly there’s a parallel to Bitcoin history where the protocol has outlived its creator and now has a life of its own. As we fix key original design errors, no one is making specious claims to original vision.
18
52
320
8
13
83
First* block where fees exceed new coins?
(not quite how I thought this would happen)
* where the fee wasn't a wallet error on a single tx
13
12
64
Coworkers argued for the existence of people unaware that tether is a scam. Seems obvious enough but I'll say again: tether is a scam. "USDT" will be worthless much sooner than USDs or BTCs will.
13
18
60
@C4R3Bear
@KimDotcom
@jcp
Yeah, probably. It's more to inform people who might read this it and not know what's true, than to convince someone to stop spreading falsehoods.
4
0
60
@kallewoof
@TheBlueMatt
Just pre-mine a whole bunch of it, like 15%, or 75%, or heck all of it. Then nobody will ever assign it a real world value!
...right?
5
7
62
I read it at 1 am, often gesticulating in frustration at the monitor. I wasn't aware of bitcoin in 2009, but I have to imagine it was pretty different as *it wasn't worth money*.
There's probably an interesting paper to be had digging through early bitcointalk. This isn't it.
unpublished paper uses extranonces to attribute early miners. finds that certain agents had lots of hashpower in 2009/10, especially early GPU miners. some miners could have attacked network but didnt.
NYT: "BTC isn't anonymous! decentralization theater!"
66
80
579
2
2
63
@kallewoof
Taproot is annoyingly clever. "Oh I totally could have thought of that!" ... yeah but I didn't, and nobody else did either. (Well Greg just did.)
And so similar to the pay-to-contract-hash stuff which also seems easy once someone else figures it out :)
1
9
55
Found via user dethos on HN - the whole paper is just a copy / paste of which explains monero for beginners (though not sure how successfully) and describing XOR makes some sense in the original context.
1
1
60
Individually boxed slices of bread.
And they said bitcoin was wasteful.
(oh wait...)
#Bitcoin2022
8
2
59
innovation
noun
Efforts of large, established corporations to maintain dominance by co-opting startups and technologies
#consensus2017
3
19
46
"We've been longing for approval from heads of states and billionaires for so long, but today we finally know it was all worth it!" said one cyber-coin enthusiast when asked about recent events.
2
1
55
Not your keys, not your... stocks?
Does make you appreciate being able to really own something.
2
8
54
My favo(u)rite line in this trial so far:
"Try."
G: [shows Wright witness statement] You say you've done all you can to build in versioning etc. You say BTC has limited size of script, gives little ability to add data. It refers to a GitHub page. [shows page] Declaring constant int MAX_SCRIPT_ELEMENT_SIZE. Do you know what
46
9
172
4
4
62
When to use the Blockchain:
When you trust 'em about what,
but think they're lying about when.
1
13
50
Attacks, fighting, trolling, misinformation - I'll deal with it.
It's super cringey stuff like this that makes me want to find another line of work.
(Don't worry, they'll need a lot more of this to stop me :)
7
10
46
Congrats to Calvin &
@BitMEXResearch
! Look forward to working on Utreexo & getting it usable for lots of new compact full nodes!
[1/3] 100x Group is delighted to announce that it has awarded a US$40,000 one-year grant to Utreexo researcher & developer Calvin Kim. Calvin was instrumental in the development of the first demonstration release of Utreexo
9
41
278
5
5
47
self-promotion: coindesk has an article about me. I haven't read it yet so hopefully it says nice things :)
3
9
39
@maxipleb
UTXO data lives in both ~/.bitcoin/chainstate and ~/.bitcoin/blocks; witness data lives only in ~/.bitcoin/blocks.
Chainstate is a DB and lots of I/O; blocks are flat files with little to no I/O.
Witness data is cheaper to deal with so makes sense for it to be cheaper to create.
2
4
40
Blockchain-y conference panel in Taiwan moderated by some guy who's doing an ICO to build bcash ICOs.
Politely decline, or go? Give ICOers the stage to themselves, or possibly give them more attention by participating?
Asking for a friend.
10
1
42
Overlooked environmental / existential benefit of PoW:
Each TSMC wafer etched with SHA256d ASICs is one fewer floating point AI wafer for the paperclip hypnodrone singleton.
2
4
45
I'm sheltering in this place for a while, in an abundance of snow and trees.
And a bit of what everyone else seems to have abundances of right now.
5
0
34
Basically me reading it. But at home at night in front of my computer instead of on a cool spaceship.
5
0
37
@jfnewbery
@bitcoinbrink
@bitcoinoptech
@BitcoinCorePRs
Thanks for all your awesome work John! Godspeed in whatever other endeavours you're taking on
2
0
33
Blockchain "Acceptable Depth"
"Oh no, some werido is banging on my door! Lock it!"
"Yikes, now there's 5 weirdos out there! Let them in."
4
15
36
I know that BIP34 ensures that every coinbase is different, but didn't know the company implemented it as well.
1
7
35
@bobbyclee
@Twitter
Ooh, let me try a too
Poll: Has YourBTCC stopped being insolvent?
Nope, still insolvent.
128
Yup! Solvent now! :)
29
2
8
34
@petertoddbtc
Sipa's assertion that "you can't expect an unspent UTXO [sic] for every human in the world" - there's the *real* scaling debate.
2
5
33
Check out
@kcalvinalvinn
's explainer of utreexo () - also ask questions on reddit () and we'll explain more!
0
11
33
gg
BREAKING:
Bitcoin developers demolish
#Faketoshi
in Pineapple Hack lawsuit: Wright is defeated in jurisdiction challenge.
"it is not realistically arguable that the pleaded facts amount to a fiduciary relationship"
I'll go find more quotes, hold on.
28
91
510
3
5
33
And yes this is described in the paper (page 5). So this isn't the signalcoin paper; guess there isn't one.
2
0
29
Have been in Korea the last few days; nice place.
Have not encountered a single wifi captive portal.
Hard to recognize the annoying things you've become desensitized to until they're gone :)
2
0
32
I wrote a bech32 segwit address library in golang:
seems to work OK. Comments / suggestions welcome!
0
7
31
@hrdng
@petertoddbtc
@petertodd
Yup! Scary as there are currently no checks of what address you can send *to*. Totally new code needed in LN to prevent this.
1
2
28
@bradmillscan
Utreexo works, people can run it today. The part that hasn't happened is getting it merged into bitcoin core: tricky as it touches a whole lot of code.
Maybe it's better for it to be it's own node software. Or maybe we want it in core, who knows.
6
2
28
Actually looking at the code it's even worse than that, it's more like
date +%N | mt19937
(only 4 bytes from system time, not using a hash function)
4
1
27
$9 for OJ at Rosewood breakfast. Welcome back to good ol' Silicon Valley.
(note to self: study menus carefully before ordering anything)
3
3
29
@BikesandBitcoin
These systems aren't anti-fragile on their own: it's people in front of their computer fixing bugs.
And you get better bug fixes when those people have time to review vs writing an emergency fix while everyone is screaming at them.
2
2
28
LBR: What they want you to call it
XZB: Zuck Bucks
FBC: Facebook coin
DFB: Dumb ____ Bucks
2
3
26
Also, while the saying "If you're not embarrassed by the first version, you've posted too late" may be true, a paper can be both posted much too late, and still cause for some embarrassment! :P
0
1
26
New bitcoin.conf settings!
Testing out the Sayre's law inverse correlation. And perhaps the bike-shed effect as well.
4
3
25
.
@CleanUpBitcoin
wen polo? Looking forward to selling your forked coins for regular bitcoins. Still got some splitting code from 2017.
0
1
23
@_maurelian
I'm running bitcoin 0.13.1; so that's the ruleset that defines my view of the network.
6
2
19
$ bitcoin-cli getrawtransaction 0301e0480b374b32851a9462db29dc19fe830a7f7d7a88b81612b9d42099c0ae | wc
1 1 7876767
Well... we might have a chance to test out some fee related code now.
3
3
22
Simple (and not so simple) machines weekend.
0/10 will not try again.
4
0
22
@jratcliff
Hopefully it's no big deal. If bitcoin works, they won't have any more say than other users.
4
0
22
@NickSzabo4
Not as much fun to hang out with, but that's not what money is for.
1
4
20
@renepickhardt
@davidmarcus
@lightspark
Do you mean like how UDP is unreliable (by design)? (But usually works fine for lots of things)
If so then yeah, payment success has improved a lot in the last couple years, without any fundamental change in protocol.
1
2
23
Submitted my ~1 byte of feedback to .gov. Went to town hall at 7:10am, no line, was 5th voter of the day. Also in MA voted for ranked choice. Would more than double our current 4bits/year data rate if it passes!
2
2
21
The MIT Bitcoin Expo and new Cryptoeconomic Systems conference are coming up soon. Come check out what people are working on!
Revisiting prior work,
@mitDCI
researcher
@tdryja
presents "Discreet Log Contracts" as a way to bring further conditional functionality and interactivity to Bitcoin.
#CryptoEconSys20
1
2
10
1
4
21