dlubarov

63 ms·

751.
by dlubarov 8y ago
Why is that Oracle instance using a max thread pool size of 2048 on a 12-core host? That seems like the real problem. Why not configure a smaller thread pool like cores * 2, and do more queuing on the server? It seems wrong IMO to expect c
752.
by dlubarov 8y ago
I think that Ouroboros paper [1] is a bit overhyped. A common criticism of proof of stake is that an attacker could acquire private keys to old accounts whose coins have already been spent, then use those to build a long fork. It has been s
753.
by dlubarov 8y ago
All of Qatar's traffic used to be routed through 82.148.97.69, though that was back in 2006-2007. At one point it was banned from Wikipedia, which unintentionally affected the whole country. https://simple.wikipedia.org&#x2F
754.
by dlubarov 8y ago
The thing is, this assumes a shortage of capital for productive enterprises. Historically, that's been normal, but not so much since 2008. Banks have been hoarding cash well in excess of reserve requirements [1], despite ~zero interest
755.
by dlubarov 8y ago
> any forum discussing any cryptocurrency A counterexample would be http://ethresear.ch/ , though it's the exception that proves the rule.
756.
by dlubarov 8y ago
There's Enigma "v1" (SGX only) and "v2" (SGX + MPC). My understanding is that in both schemes, all code will be public, although inputs and outputs can be private. So e.g. if a company wrote a multisig contract, the
757.
by dlubarov 8y ago
When space-intensive computations are used for consensus, it's usually called proof of space. There are a few projects doing it already. https://en.wikipedia.org/wiki/Proof-of-space
758.
by dlubarov 8y ago
That must be for SSDs, which are ~$150/tb vs ~$20/tb for HDDs. I expect IA must use HDDs for the bulk of their data.
759.
by dlubarov 8y ago
But the company could get approxately the same amount of cash by selling the new issuance on a public exchange.
760.
by dlubarov 8y ago
Possibly -- if the attacker accessed session IDs, they could potentially hijack the sessions of logged-in users. If you log out, most servers will destroy the session data on their backend, so there's no session that can be hijacked.
761.
by dlubarov 8y ago
It could just be a 3xx redirect over clear http, right? The http site can redirect to a https site with a similar name.
762.
by dlubarov 8y ago
A CT-style system could work for supply chains, but let's say for the sake of convenience, we wanted to use some shared storage system rather than building a new system for supply chains only. We could have one big CT-style log system
763.
by dlubarov 8y ago
In practice Bitcoin treats SHA-256 outputs as integers, and requires them to be under a certain limit, based on the current difficulty. That way we can make small adjustments to the difficulty. If we actually used a rule like "this man
764.
by dlubarov 8y ago
> I am sure somebody else can provide more color but it is my understanding that there are 6 or 7 intermediaries that each credit card transaction travels through - each taking their fraction of a penny and delivering questionable value
765.
by dlubarov 8y ago
I haven't used Chrome Remote Desktop recently. What exactly were you doing when sync turned on? I've been using Gmail without signing into Chrome. Post-update, I am signed in, but sync is still off.
766.
by dlubarov 8y ago
> Money supply is not the same thing as inflation! Understood; I wasn't trying to say anything about spending power. I like to talk about inflation of money supplies since it sheds light on raw monetary policy. CPI metrics are more
767.
by dlubarov 8y ago
> there is nothing stopping the Bitcoin network, for instance, from deciding it wants to inflate after it has mined all the coins But at least it would be up to a community of users, rather than government decree. I doubt any of the cryp
768.
by dlubarov 8y ago
> people do not want to be engaged in the process of running a bank or a government But casual users can choose to store their crypto assets with a bank/custodian/exchange. Granted, it defeats some of the benefits, but not all.
769.
by dlubarov 8y ago
> We can do consensus in many (much cheaper) ways. Agreed, but that's why we're working on proof of stake systems. With BFT algorithms, consensus can also be much faster, on the order of 1 second. > Most people don't wa
770.
by dlubarov 8y ago
I think "centralization" is pretty ambiguous. Is Bitcoin centralized? Full nodes can be run on home PCs, so verification is decentralized, but only ASIC mining is profitable, so mining has become highly centralized. The Lightning
771.
by dlubarov 8y ago
> If power goes up in value, it becomes more expensive to produce new Bitcoins as well as process transactions. Difficulty is adjusted every 2016 blocks based on the timing of previous blocks. So in the long term, the rate of block creat
772.
by dlubarov 8y ago
The volatility isn't a problem if crypto is a small part of a diversified portfolio, though. In theory, if a portfolio has many small investments in volatile but uncorrelated assets, the overall portfolio won't be volatile. I have
773.
by dlubarov 8y ago
> Having 1/0 = +inf is a perfect solution. It makes sense if we have signed zeros; then +0 represents an infinitesimal positive value. IEEE 754 has signed zeros, though the signs may sometimes not match programmers' expectation
774.
by dlubarov 8y ago
Unfortunately most BIOS services are only available in real mode, so it's often necessary to switch back and forth.
775.
by dlubarov 8y ago
HSP/HFP for calls is very standard; A2DP isn't suitable for calls since it's unidirectional. HFP 1.6 supports wide band (16khz) speech with the mSBC codec, which is a big improvement, but adoption has been very slow. E.g. in
776.
by dlubarov 8y ago
It's flexible; presubmit tests can be configured per-directory. There's also an option to run all tests of packages that could be affected by a change based on the Blaze dependency graph. If you're making changes to a package
777.
by dlubarov 8y ago
They must have meant that 95% of Google engineers use the monorepo in some capacity, even if the majority of their work is done in a different repo.
778.
by dlubarov 8y ago
Not sure what you mean? Correct evaluation (modulo some negligible error) is part of the usual FHE definition.
779.
by dlubarov 8y ago
There was a good discussion about this on law.SE: https://law.stackexchange.com/questions/596/what-legal-liabi...
780.
by dlubarov 8y ago
Even if some shared code is currently branch-free though, it may be unlikely to remain that way if the abstraction is fragile. A red flag is vague function names like "processThing" or "afterThingHappens". If a function

More ›