1,432 karma · joined May 20, 2011
The main thing I learnt from ESR was that you can, actually, do worse than Crazy Uncle Stallman.
But we got "Everyone Loves Eric Raymond", which made me feel less like the odd one out!
Disputed, especially how significant it is...
The original futex code used physical addresses, and pinned the page, which is simple: if you can share memory somehow, you can use a futex.
Of course, most people use pthreads and increasingly the idea of a Linux userspace locking primitive gave way to a glibc-pthread accelerator...
Not because these functions couldn't handle it, but because this assertion simplifies optimizations elsewhere.
This has required adding extra checks in my code, found mainly by trial and error, and has made it less readable and less optimal.
Finally, the checked arithmetic operations returning false on success is a horror show. Fortunately it will be found on the first time the code is run, but that's a damnably low bar :(
Binance came out of nowhere with massive fake volume, as a China-based exchange pretending to be based in Hong Kong. They've proceeded to thrive by having a reputation for giving accounts to anyone, globally (which, TBH, is not a bad thing as they operate in countries where it's better than trying to save in the local currency). And their home country is "somewhere you are not".
There is no basis for believing that they have maintained all the funds: why would they?
This really does mirror the Library model, but on the internet.
They felt fine changing the argument order, why stick with the reverse polarity?
Throughout the talk they showed Linux vs Solaris performance, and talked about their optimizations. It was lmbench (Larry McVoy the author was ex-Sun, which adds something) that, and by the end they beat Solaris on every result.
Obviously this was impressive, but it's hard for me to clearly express how much this shook my assumptions about how to build good software. A bunch of students shouldn't have been able to beat Sun on their own hardware in anything!
There are trustworthy businesses (e.g keys.casa) which help with custody. But it's a different business these days.
This is probably a good thing, as it's the only real way to ensure they actually have your money. And it's a story if you're worried about being last one out in that case! (I think it unlikely for Coinbase TBH).
Definitely some people (myself included!) hope this becomes a cultural norm for Bitcoin. But that might be a bit optimistic!
But seriously, write down the 12 seed words, keep it somewhere safe. Make another copy, tear it in half (6 words each) and keep one with a relative and one at work. Check them at 1, 2, 4, then every 6 months.
But for most people this is all overkill: just use a mobile wallet.
Most people seem to be just using "prove you hold some amount" which is a very poor cousin. However, it is pretty clear that FTX would have failed even that, so maybe we should lower our expectations?
Two important points: 1. Generally Proof of Reserves is a term where all holders can validate cryptographically that their funds are accounted for. This is harder because it's natural and efficient for an exchange to co-mingle multiple customers. 2. Doesn't mean they're solvent! They might owe customers $100bn for all anyone knows.
It's still far better than nothing, but FWIW their history of lying about their home jurisdiction, and massive "out of nowhere" growth is a huge red flag that you should be aware of (FTX did the latter even faster, of course, being founded in 2019!).
Remember, this was a casino, not really an exchange.
Would you like to expand your point?
I guess Bitcoin gets the clicks more than Ethereum?
To be fair, the slow polishing of a thousand hands continues. libsecp256k1 is tested to a level few if any FOSS libraries achieve, for example.
Though I responded in 2012, so it may be older (the log on the site says 2013 though).
FWIW, my response, as FHS editor:
To quote Kernighan and Plauger: "Don't patch bad code: rewrite it".