So when I said three years ago that a Lightning transfer borrowed the funds of the intermediate nodes, I wasn't wrong. They've now made it explicit. Lightning is a credit-based system. And always has been.
This is actually an improvement on the original design, since those who need liquidity can now explicitly borrow it from those willing to offer it. In the original design, channel funds were simply hijacked to make payments.
The devs I spoke to at the time insisted that there was no need to obtain channel owners' permission before hijacking channel funds to transmit a payment for someone else, because in their view it wasn't borrowing.
It was borrowing, of course, but they didn't seem to understand the concept of property rights. To them, it was simply a tech problem - "how do we get a payment from A to C? obvs, we send it via B". The fact that B was someone's property was lost on them.

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with (((Frances "Cassandra" Coppola)))


Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @Frances_Coppola

18 Nov
I have now blocked both the antisemite who couldn't tell the difference between the UK and the State of Israel, and the antisemite who couldn't understand that alleging someone with Jewish connections is working for Israel is antisemitism.
The tweets I have seen tonight have convinced me that the Labour Party really does have a serious problem with antisemitism among ordinary party members and supporters.
And furthermore, that Jeremy Corbyn is the focal point for antisemitism in the Labour Party.
Read 4 tweets
17 Nov
Your regular reminder that I block people who insult me and mute people who are boring. Bitcoiners are not exempt from this rule. Those who called me "inflationist scum", "f*cking evil", "ugly old hag" and other choice epithets have all been blocked.
The behaviour of Bitcoin cultists on this site is disgraceful. This is not your personal echo chamber, it is a forum for open debate. You do not have any right to order people you disagree with to be silent, and nor do you have any right to issue disgusting misogynist insults.
I came back after two hours off Twitter to find a barrage of insults, offensive remarks and childish GIFs from Bitcoiners who seem to think no-one except them has any right to an opinion and no-one except them knows anything about crypto, markets and money. Utterly pathetic.
Read 4 tweets
17 Aug
I have to go and do some work in a minute. But before I go, there is one more little gem that I discovered in this algorithm.
The grade distribution calculation process starts with the school or college's exam performance in previous years. This is then adjusted to take account of the difference between 2020's cohort and previous years. The difference is calculated using GCSE data for 2019 and 2020.
But they can only do this for students that have GCSE data that fulfil the criteria. So the formula that calculates the grade distribution is weighted by the proportion of students that have qualifying GCSE data.
Read 6 tweets
15 Aug
When you are trying to find out what has gone wrong with an algorithm, it's always worth checking the code, or if you can't get hold of that, the tech spec. Here's the data spec for @ofqual's standardisation algo.…
cc @odtorson this is right up your street
The specification details the required calculations at each stage in the process of generating grades, together with the data inputs and outputs.
Step X3 calculates a "national value-added" using (for A levels) historical GCSE results across the entire cohort of historical "Learners".
Read 11 tweets
15 Aug
The DofE's algo does not "moderate down" teacher-assessed grades. It ignores them. This is deliberate, since it replaces the external exams that students have been unable to take. Teacher assessment (now) plays no part in external exams, so it doesn't in the algo either. 1/
However, the algo has serious limitations that make it a very imperfect substitute for exams. And the solutions they have designed to deal with the algo's limitations introduce intrinsic unfairnesses that are absent from the exam system they aimed to replicate.
1) The statistical methods used to infer grades don't work where student numbers are small. So they have used teacher-assessed grades for small numbers. This is equivalent to telling students in small classes that they don't have to take exams.
Read 12 tweets
14 Aug
"The grading system this year puts teachers at the centre of the process – they know their students best so we asked them to set out the grades they think pupils would have got if exams had gone ahead"
This is not true. 1/…
You did not ask teachers to "set out the grades they thought pupils would have got if exams had gone ahead". In fact you completely ignored the grades teachers predicted for their students. What you actually asked teachers to do was rank students in relation to each other. 2/
The ranking order was grade independent. A teacher could predict her 20 students would get 5 A*, 10 A, 3 B and 2 C. But the algorithm could come up with grades of 1 A*, 9 A, 4 B, 2 C, 2 D, 1 E and 1U for her students using HER RANKING of them. 3/
Read 6 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!