roca

23 ms·

331.
by roca 5y ago
I wonder how his supporters deal with the fact that, if his test wasn't falsified, then Djokovic's plan for getting into Australia was apparently to catch COVID almost exactly when he claims he did. I.e., at some point Tennis Au
332.
by roca 5y ago
Yeah. Automatically optimizing for space gives good cache footprint if you don't know anything about access patterns so it might as well be the default. Also, with generics, there are cases where the optimal field ordering depends on t
333.
by roca 5y ago
This has nothing to do with what I said.
334.
by roca 5y ago
Taking everything written at face value, a big problem with the author's thinking is that they ask "Did we have a complete picture of what the vaccines did, or were we standing on top of an iceberg with lots of information still l
335.
by roca 5y ago
I haven't used a Web3 site. How is it more seamless than Webauthn's "go to the site and have it verify my identity without me doing anything (using my plugged-in Yubikey plus the browser-stored user credentials)"? Person
336.
by roca 5y ago
I'm open to the idea that there are real problems that are best solved by PoW/PoS blockchains and smart contracts, so I was hoping this article would reveal one. It doesn't. As mentioned elsewhere, Persona was already a perfe
337.
by roca 5y ago
MAGMA is my favourite string currently.
338.
by roca 5y ago
The approach I suggested would do away not only with the intermediate object files but also the "do all the relocations" work.
339.
by roca 5y ago
> Except, the size of the object file will inevitably change. So you are stuck reimplementing a mini-filesystem with logic for allocating/freeing blocks within your final executable, It would be more like malloc/free than a min
340.
by roca 5y ago
That's a question worth considering. But fundamentally, it should be faster to write compiler output directly to the final executable than to write it to an object file that is then copied into the final executable.
341.
by roca 5y ago
Not at all. If they're healthy and well and a pandemic isn't on then moving away is very low-risk.
342.
by roca 5y ago
We would never have been able to construct enough MIQ infrastructure to let people come and go as they pleased, which is what would be necessary to satisfy the "right to return is absolute and cannot be constrained under any circumstan
343.
by roca 5y ago
I don't understand your point here. Are you claiming that we could have dispensed with MIQ altogether and kept COVID out some other way? Or are you claiming that we could have constructed enough MIQ infrastructure to let people come an
344.
by roca 5y ago
I can't answer that in full because I don't know all the details, but one issues is that Taiwan captures phone location data in a way that has no precedent here.
345.
by roca 5y ago
MIQ was an essential component of the elimination strategy. Elimination was a fundamentally egalitarian strategy; it disproptionately benefited the less-well-off, who inevitably suffer more when the virus is allowed to spread, and it was di
346.
by roca 5y ago
"Bounced out"? No citizens are being forced to leave the country. These new visa holders just make it a little bit harder for citizens and residents to enter the country, that's all. Over 200,000 people have gone through MIQ
347.
by roca 5y ago
In practice MIQ capacity has been limited by the number of suitable hotels available, and the difficulty of staffing them. These resources are not very elastic.
348.
by roca 5y ago
The right to return to NZ does not automatically trump all other considerations. Without MIQ we couldn't have executed our elimination strategy. Probably thousands of people would have died of COVID, or we would have had stronger and l
349.
by roca 5y ago
Counterpoint: Larry Page entered NZ in early 2021 and was placed in a middling MIQ hotel. His people tried to get him moved to one of the best hotels but were unable to do so. Page is the world's fifth richest person. I think this coun
350.
by roca 5y ago
Exactly.
351.
by roca 5y ago
It is necessary to train oneself and others to keep on doing the right thing even when no-one else is doing it.
352.
by roca 5y ago
No need to drag Unicode normalization into it; don't require strings to be normalized. Normalization is only relevant in very specific contexts and you don't want to pay for it elsewhere.
353.
by roca 5y ago
I have personally spoken with mainland Chinese Christians who report increasing persecution there.
354.
by roca 5y ago
Regular UTF8, not WTF-8 or any of those other variants (which are for encoding data that is not necessarily Unicode).
355.
by roca 5y ago
Evolving how? I'm not aware of any reason to move beyond UTF8 for encoding Unicode.
356.
by roca 5y ago
Thanks, that's illuminating.
357.
by roca 5y ago
Did the Romans really know that using certain volcanic rocks would make their concrete structures last longer? Or did they mostly just use whatever aggregate was convenient and the structures that have survived are the ones that used the ri
358.
by roca 5y ago
The thing is, code tends to be reused across projects. If you write a library or a utility program and it's full of holes, that's only OK if you're sure it will always be used in a "safe" context with no untrusted
359.
by roca 5y ago
I wonder how finely tuned their implementation is to the 12 performance bugs they reported results for. The paper doesn't seem to say anything about that.
360.
by roca 5y ago
It's interesting to see the number of bugs related to race conditions here, and it would be very interesting to know if Rust's protection against race conditions would have been relevant ... there are plenty of good modern memory-

More ›