addaon

58 ms·

1051.
by addaon 4y ago
What's the advantage of taking risk in this scenario? It seems that the obvious goal is to minimize chance of running out of dollars before days, which sounds like risk reduction to me.
1052.
by addaon 4y ago
The white plastic iBooks, especially the G3s, were incredibly prone to damage this way. I think I went through three G3s (on warranty) before they replaced it with a G4, and then swapped out a busted G4 still on the original warranty. Never
1053.
by addaon 4y ago
This doesn't pass the smell test, and doesn't seem to be true. See e.g. [1]. [1] https://www.gov.uk/jobseekers-allowance
1054.
by addaon 4y ago
> Calculate how much time you can afford to be unemployed, and how to extend that time if necessary. I’d be interested in thoughts from others for whom this time comes out to decades, but less than actuarial time remaining. What’s the t
1055.
by addaon 4y ago
If the company went through the approval chain and got everyone to sign off on you being worth $comp + $relo, and you're value isn't effected by not choosing to $relo (because, in this case, you're already there), it should b
1056.
by addaon 4y ago
If an employer offers a bonus buyout during negotiations, that's a straight-up offer of more compensation, in the form of a signing bonus. It's hard to come up with scenarios where an employer would offer you a certain total compe
1057.
by addaon 4y ago
Sure, getting your bonus bought out provides some superficial relief for this problem. But if your next employer thinks that your value is high enough to buy out your bonus, they likely think that independent of the current date (modulo a s
1058.
by addaon 4y ago
Using zero as a default (the main topic of this article) is one thing... I'm pointing out that floating point is able to represent points with greater density near Null Island than it is in, say, Alaska, just due to the spacing between
1059.
by addaon 4y ago
Disagree with this on several levels. One specific challenge comes as tech companies push more and more compensation to bonuses and other must-be-present-to-win approaches (RSUs, etc). If you leave shortly before a trigger date, you've
1060.
by addaon 4y ago
Null Island is also the place where so many self-driving cars, UAVs, and navigation systems achieve their best possible performance. I've run across multiple products that represent latitude and longitude as floats internally. At doubl
1061.
by addaon 4y ago
Yeah, it's pretty rare to have a program that's correct regardless of the behavior (and defects) of the layers below it -- supporting multiple platforms of course involves performing your verification activities on all of them. As
1062.
by addaon 4y ago
If you're doing large stack allocations like this (or even if you're not, but care about correctness), you should be doing static analysis to prove absence of stack overflows. One of the main advantages of heap allocation over sta
1063.
by addaon 4y ago
LG (at least for my OLED model) supports firmware updates over USB, and posts firmware updates to their website. It's a very smooth flow -- my years-old TV has never had an internet connection and is up-to-date.
1064.
by addaon 4y ago
I can't speak for < 2010 or so, but even then 500 nA was common, and half that wasn't rare. I'd expect that we crossed 1 µA around 2003 - 2005, as a very rough guess. But that doesn't mean that every implementation is
1065.
by addaon 4y ago
A modern RTC should draw such low power from the battery (tens of nanoamps to maybe a microamp) that the life of a primary battery will basically match the shelf life. Indeed, many coin cells actually have a longer life with a small amount
1066.
by addaon 4y ago
The problem with replacing the Messier catalog naming is just how neat and clean the single namespace ends up. There have been alternate proposals, but they all ended up messier.
1067.
by addaon 4y ago
I /mostly/ agree -- it's hard for me to think of a modern or future platform that would use ones complement integers... with one possible exception. Integers represented on top of IEEE floating point are inherently ones-compl
1068.
by addaon 4y ago
It's a philosophy thing. There are plenty of languages that just solve for flat Von Neumann memory models. In the past, that did not describe all machines -- e.g. the pre-standard Borland C/C++ compilers for x86 real mode that use
1069.
by addaon 4y ago
Sure. The C standard has always erred on the side of supporting odd platforms -- see ones complement, decimal floating point, sizeof function pointer != sizeof data pointer, etc. Most platforms today have converged around an approach that d
1070.
by addaon 4y ago
I mean, a C spec compatible compiler for the MSP430 would be an example (see paragraph 7.20.3), TI has just decided to deviate from the language standard here. It's an example of a platform that would meet the requirements, and chooses
1071.
by addaon 4y ago
I'm sure they're out there. The closest that comes to mind is the MSP430, but not quite -- although it has 20 bit pointers (with sizeof ptr being... 4, since they're padded to 4 bytes for storage) and has a 16 bit size_t, my
1072.
by addaon 4y ago
Yes -- anything that uses fat pointers and defines ptrdiff_t to be a fat pointer type that is larger than size_t/ssize_t.
1073.
by addaon 4y ago
But both C and C++ are comfortable with *(foo() + bar()) having undefined evaluation order, so foo()[bar()] potentially having order-dependent results is clearly not sufficient reason for this decision...
1074.
by addaon 4y ago
This was previously an undefined order. It is now defined. Any code that you would have reviewed previously requires no additional knowledge to review now. Any code that is depending on this new C++17 defined behavior is either pushing the
1075.
by addaon 4y ago
The key point of the article here is that this is no longer strict syntactic sugar in C++ as of C++17. The addition operator, like many operators and all function calls, does not specify the order that its arguments are evaluated. In C, and
1076.
by addaon 4y ago
> maybe it's the drooping line towards the back? Yeah, this particular line, and the large C pillar mask in the rear windows combined with super tiny quarter glass, just do not at all work for me. But I can see how they might for ot
1077.
by addaon 4y ago
Yep. And it's worth saying that, contrary to my previous over-simplification, the trade-offs change once you give up on single-cycle instruction decoding. With a longer pipeline, you can do a split into 16 simple, small pre-decoders, t
1078.
by addaon 4y ago
> The hard part about a variable-length ISA, as I understand it (hardware is not my expertise), is that you have to build essentially a mux that can steer from 16 source places to 5 outputs. It's not just the mux, it's the cont
1079.
by addaon 4y ago
Keep in mind that the photon is the fourth electroweak boson, giving a nice power of two. W+, W-, Z0, and the photon are all mixed from W1, W2, W3, and B, after symmetry breaking. The photon doesn't couple to Higgs, but still, at highe
1080.
by addaon 4y ago
But fundamentally you haven't shown that there's any method of taxation that eliminates the revocability of property. What's your proposed response to non-payment if not compelled payment / forfeiture of assets?

More ›