Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
bloaf
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
46 ms
·
481.
▲
by
bloaf
8y ago
I would guess that it is legitimate. The fake review problem is real. Here is an example of an item with 2k good reviews: https://www.fakespot.com/product/dewalt-dw745-10-inch-compac... Notice that it is not a cheap
482.
▲
by
bloaf
8y ago
An ancient Sumerian ledger: https://cdn-images-1.medium.com/max/1250/0*7LNb7XKpR6vAdsBH.... A ledger from the 1700s: https://recordoffice.files.wordpress.com/2014/12/d2375m-277-... I thi
483.
▲
by
bloaf
8y ago
It could also be the admission that you can't beat pure skepticism. That is, you can't actually beat Descartes' deceiving demon; you need to start any line of reasoning with an assumption that your reasoning is somewhat soun
484.
▲
by
bloaf
8y ago
Not only that, but their home "meals" are a little bit ridiculous. Evidently "less than 250 calories [1] of straight pork with no veggies, carbs, or drink" is what they consider a meal. [1] https://www.wolfram
485.
▲
by
bloaf
8y ago
It'd been going on so long, I would argue that there is an element of "coming to the nuisance" [1] going on. That is, the situation is like people who move next to a farm in order to "get back to nature." Hey! Tur
486.
▲
by
bloaf
8y ago
Which part: the subservience of metaphysics or the the hypocrisy of the metaphycisists who deny science?
487.
▲
by
bloaf
8y ago
>To illustrate: there is common belief that metaphysics became irrelevant -- that it was replaced by science. This is, in itself, a metaphysical position I think that is wrong. What happened is that we realized we knew so little about p
488.
▲
by
bloaf
8y ago
Isn't the real issue that meta-physics is that it always follows actual-physics? That is, if the metaphysicist says something like "guns don't really exist" and the physicist says "but I made a real gun, can I sh
489.
▲
by
bloaf
9y ago
You do realize that much of wikipedia's success came at a time when the moderation standards were much lower than they are today, right? Indeed, I would suggest that the current high moderation standards at both Wikipedia and SO are n
490.
▲
by
bloaf
9y ago
Re: Speed Mathematica is undergoing a compiler re-vamp, which may change that.
491.
▲
by
bloaf
9y ago
People tend to underestimate the extent of Mathematica's libraries. Take for example process control [1]. Not only does Mathematica have a pretty thorough set of functions for solving process control problems, it works with both line
492.
▲
by
bloaf
9y ago
I think these "well-kept gardens" are basically the HOA's of the internet world. The property values may be high, but that doesn't mean they're good or healthy places to live.
493.
▲
by
bloaf
9y ago
I tend to agree about the hyper-strict parts (e.g. less than $500k in savings). The thing to think about is whether or not you'd be better off if you kept a job, but had a lower savings rate. How certain are you that you'd be abl
494.
▲
by
bloaf
9y ago
https://supreme.justia.com/cases/federal/us/326/501/case.htm... >...we have recognized that the preservation of a free society is so far dependent upon the right of each individual citizen to rec
495.
▲
by
bloaf
9y ago
I think you dropped a zero somewhere along the way.
496.
▲
by
bloaf
9y ago
He basically ran a campaign of "People in Washington suck and I'm going to ruin their day" and that is the one campaign promise he's knocked out of the park. Every time the media gets bent out of shape about something h
497.
▲
by
bloaf
9y ago
1. The german language page is actually targeting Liechtenstein, not Germany. 2. Similar projects are not the actual project in question. 3. ICANN is not a governmental agency; filing for a particular TLD is neither a necessary nor sufficie
498.
▲
by
bloaf
9y ago
Does anyone have any F# books they would recommend?
499.
▲
by
bloaf
9y ago
My understanding goes like this: The brain is a super information processing machine. It executes incredibly and unimaginably complex algorithms without our knowing about it. But one of the things it does is to construct simplified models
500.
▲
by
bloaf
9y ago
I am willing to bet that companies have a reasonably good handle on the returns from their advertising spending. I suspect they have no better than guesses as to why those returns change over time, or what might improve the returns. That b
501.
▲
by
bloaf
9y ago
Big companies ask themselves "how can we lower our advertising prices?" and come up with the idea of claiming that their ad-buying decisions hinge on aspects of the advertising platform that will be very expensive to fix. In addi
502.
▲
by
bloaf
9y ago
That section is confuse-making. Presumably their house, car, and boats DO cost more than 10% of their income. Their payments on those items might not exceed 10%.
503.
▲
by
bloaf
9y ago
I've read about this for a few mins now, and want to check if my understanding is correct. There are these classes of algorithms: 1. Cache aware algos are tailored to specific environmental-cache behavior. They may perform poorly in o
504.
▲
by
bloaf
9y ago
“Only those who decline to scramble up the career ladder are interesting as human beings. Nothing is more boring than a man with a career.” --Solzhenitsyn I think there's probably more to your observation than meets the eye. Specifica
505.
▲
by
bloaf
9y ago
>there are other cues you can* pick up on *Assuming that you are neurotypical
506.
▲
by
bloaf
9y ago
Do communities need to justify their existence by reference to some problem that they solve?
507.
▲
by
bloaf
9y ago
See also: Jet.com before it got bought.
508.
▲
by
bloaf
9y ago
Aren't some manufacturers better than others, though? https://arstechnica.com/information-technology/2016/03/once-...
509.
▲
by
bloaf
9y ago
If anyone knows of a board designed to work with power-over-ethernet, let me know.
510.
▲
by
bloaf
9y ago
Wait, it ran out of memory because it couldn't hold the class structure?
More ›