Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ludamad
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
7 ms
·
31.
▲
by
ludamad
4y ago
Not trying to make a strong statement but obviously correlation in moves doesn't mean it's been pegged to SPY. If they could have a digital asset pegged to SPY, maybe it'd be attractive, as well
32.
▲
by
ludamad
4y ago
The volatility in price seems okay as far as diversification, there was some low probability of crypto prices being this high and it happened. The private key hygiene is really worrying though. You can have private key multisig but at some
33.
▲
by
ludamad
4y ago
You obviously first need to vectorize the Bible and generate the 4 billion plus word generalization of Biblespace
34.
▲
by
ludamad
4y ago
Welcome to hacker news! I learned a lot from the video (watched it all, not something I expected when I saw the length :))
35.
▲
by
ludamad
5y ago
I like e.g. '2% liquidity depth' measure along with market cap, to show how steep the iceberg is if you start selling
36.
▲
by
ludamad
5y ago
Quoting a 2017 book about how thinly traded bitcoin is? I appreciate the concept but it doesn't always feel like it rounds out the takes. (FTR, I appreciate it's thin enough to make market cap misleading, but up-to-date liquidity
37.
▲
by
ludamad
5y ago
It'd be pretty annoying for a JIT compiler to have to manage too. Java can do a garbage collection with barely any memory left - cool! But, it'd be a shame if similar tricky stuff was written just to e.g. decompile code just to ma
38.
▲
by
ludamad
5y ago
The problem is that the more a company wants to draw a line in the sand against a government, the more it must be a monopoly of some kind otherwise be replaced by someone who doesn't
39.
▲
by
ludamad
5y ago
History does not make much distinction between "language misuse" and "paradigm shifts"
40.
▲
by
ludamad
5y ago
For businesses in times of gold-based commerce, reliable bank notes were greatly welcome. Gold was heavy and hard to validate, whereas crypto is easy to validate but also so light that without proper key management it will float away :). I
41.
▲
by
ludamad
5y ago
I think this is a strong counterpoint to the "don't test your abstract, implementation-detail machines" philosophy. If you have decomposed a system into iterative passes with pre-conditions and post-conditions it can be hard
42.
▲
by
ludamad
5y ago
This author presumably wants a dialogue, though? "database-key-provenance-as-art is always a scam" is a fine opinion, but I would have picked one with more nuance
43.
▲
by
ludamad
5y ago
Among official responses from game platforms - what I meant by 'the field'. Note that I personally like the creator of itch and respect his opinion, I just did not think it was a professional use of platform.
44.
▲
by
ludamad
5y ago
I was curious to this myself (I'm speaking more generally than any project). One reason I found is just to provide margin investing. So some company buys some $FOO and it goes up 10x. It may prefer to borrow stablecoins with it as coll
45.
▲
by
ludamad
5y ago
The itch.io response here is uniquely abrasive in the field (although undeniably a sentiment some hold). I hope we don't think that's a good example of dialogue.
46.
▲
by
ludamad
5y ago
It seems silly to expect liquidity in bitcoin to support huge countries, and a bitcoin fork for each major side would be a mess but there is a question of what neutral credibility a brand new world currency would have
47.
▲
by
ludamad
5y ago
There's a few ways of doing anonymous functions now, and some capture `this` and some don't. This is not a nice part of the language perhaps, but it's not like e.g. Java doesn't have similar complexity with inner classes
48.
▲
by
ludamad
5y ago
If I like to travel I am less likely to focus on "wanting places to be shitholes", that's all I meant in your quoted text
49.
▲
by
ludamad
5y ago
"a lot of people like to feel like the rest of the world is a shithole." I wouldn't say that's my bias - I like to travel so I like to feel like the rest of the world is welcoming. However, clear problems that are diffe
50.
▲
by
ludamad
5y ago
The C++ additions surrounding move constructors, rvalue references increased my practical complexity understanding the language (although mostly grokked now). Everything else seemed the good kinds of additions; just having a new, more palat
51.
▲
by
ludamad
5y ago
You can track, but large amounts merge with small and disperse as they hit brokers etc. If your claim to someone is that their BTC is 0.01% stolen, it's not so strong. Faster you act, more you can do
52.
▲
by
ludamad
5y ago
The most plausible theory is a botnet - but probably looking to just mine and used defaults
53.
▲
by
ludamad
5y ago
Or we can be happy about this crypto spam - openly available code - while decrying most
54.
▲
by
ludamad
5y ago
I'd rather an industry standard payout, but the workers getting pay proportional to revenue is actually not a bad model. I just hate the analysis paralysis
55.
▲
by
ludamad
5y ago
I think they have a problem with hated content that's technically allowed - "buy AnimalCoin" etc. I would hate to just be brigaded against, though, Twitter already feels like a wild west of getting unwelcome replies
56.
▲
by
ludamad
5y ago
But what stops a well-functioning democracy from getting more corporate influence? Or woops-emergency-powers? I've come to realize any government is better than the alternative - until an extreme point. As a result they tend to be stab
57.
▲
by
ludamad
5y ago
As well, even with decentralization, I don't see how the bridge updates wouldn't eventually propagate. The core of this is a smart contract issue
58.
▲
by
ludamad
5y ago
Your spell metaphor doesn't serve you being this handwavey. What are you even saying happened on the ethereum blockchain during this?
59.
▲
by
ludamad
5y ago
You overcomplicate things. There is plenty of liquidity to sell 120k eth; the opportunity cost of doing this is near $300 million
60.
▲
by
ludamad
5y ago
I meant detecting clean water, as opposed to water still washing dishes/dirt
More ›