Well, it took me about five seconds to find the cache link in its new location. I don't think it's a bad change overall, since the cached page and the preview are (presumably) the same thing. It also unclutters the main results - the only per-link boilerplate I get now is the inane '+1' button.
If I had scripts blocked for google domains (or in general), I might be a little annoyed. But I also appreciate that if I disable scripts, I shouldn't expect everything to work. In Google's case, I don't block their scripts because I use their site heavily and trust it. It'd be a bit of a double standard to distrust Google's JavaScript but still trust their cache.
The 'farm people for power' storyline did seem pretty weak compared to the 'farm people for processing' alternative - after all, they could've just farmed sheep or hemp without having to worry about an uprising.
But here's a few plausible explanations:
The robots had to eliminate the human threat anyway; farming them turned out to be more sustainable than killing them and burning their bodies
It's the future - there were a lot of people and not a lot of anything else still alive
The robots had plans to 'reprogram' the people as robots, but were still developing the technology
Agreed. Have you ever seen anyone actually hold an iPhone up to their head?
And what's with the lack of custom SMS alert sounds? That's a bog standard phone feature completely absent on the iPhone. I need a loud, long noise to wake me up if I get paged in the middle of the night, not a tootle on a bicycle horn.
The magnetic connector thingy is a nice idea, but the power cord is too short and the join between the connector and the cable is too flimsy. I've got metal showing through on mine, its days may be numbered.
On balance I preferred the angled power connector on my old Asus laptop. But it's way better than the obscene protrusion on my Dell.
Wikipedia has the Apple II, PET and TRS-80 on the leading edge for home computing, all being released in 1977. Those other machines you mentioned (aside from the TRS-80) came later, and were basically just new permutations and incremental enhancements on the same designs.
But of those first three, what I found particularly striking about the Apple II was:
It was the first to be released
It was the last to be discontinued (the Apple IIe, arguably just a revision of the Apple II, was selling into the 90's)
It had the most influence on the IBM PC
Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 1
Java strings, as in instances of class String, are always immutable.
Sure, there's a CharSequence interface that covers both immutable (String) and mutable (e.g. StringBuilder) strings, but it's not good style to take CharSequence arguments rather than String just for the sake of allowing mutable and immutable strings to be interchangeable.
Haskell for one is pure functional but does have support for mutable values - arrays in particular. The catch is that you can only modify them in the context of a monad, which can be pretty inconvenient.
The 123/789 thing doesn't usually bother me, except the time way back when I went to the bank to register for phone banking. They had me enter my initial 5-digit PIN on a PC numeric keypad - 789 on top, rather than the 123-on-top arrangement on the phone. Really did my head in.
And while I'm spouting barely relevant anecdotes around phone banking, for a while as an 'added security measure', my bank would have you enter just digits X and Y, presumably so that someone eavesdropping the call couldn't just replay the DTMF tones to gain access. But never mind that this gives an attacker (much?) better than 1% odds of gaining access with a (well?) educated random guess..!?
That's not a problem if your 'real' email address has its own signature of dots. Unfortunately I wasn't aware of this trick when I created by GMail account, so that horse has already bolted.
Another approach is to not have a 'real' email address at all - instead, use a distinct address for every counterparty.
I don't buy that argument that ray tracing wins because of computational complexity - you won't break even on performance until you're ray tracing far more triangles than you have pixels to display them on.
And that means your average triangle isn't going to contribute to most frames at all, and when it does, you have aliasing!
In terms of the numbers of dollars and shares held by market participants across a trade, it is zero-sum if you ignore the transaction costs.
But that's not what people mean when they talk about 'the market', because there's this fixation on 'capitalization' whereby a value of a parcel of shares is deemed to be the number of shares multiplied by the current price per share. And that's there the zero-sum-ness of the market goes out the window, since the share price can vary independent of trading volume.
You could buy $100K worth of shares in a company today, pushing the share price up and increasing the company's market cap by $1M. Then you could sell those same shares tomorrow, pushing the share price down and reducing the market cap by $5M. There's no getting those numbers to sum to zero.
You're generously assuming that the network latency will substitute for the local latency.
More realistically, you'll get the baseline local latency of around 50-150ms, *plus* the network latency of let's say 50-150ms (don't forget more bandwidth means higher latency), and maybe some extra latency for processing the incoming video.
(Not so sure about the last bit, but I guess you'll be transferring more data from system RAM to the GPU than usual. Normal video rendering performance isn't comparable since that benefits from more buffering and pipelining.)
Anyway, all up you're looking at 100-300ms, which would garner it an *unplayable crap* rating from me 8-) (And yes, I'd apply that standard to console games as well.)
Compare to Smalltalk - once you decide that everything is an object, you pretty much have to either support operator overloading, or not support operators at all.
Java on the other hand has both primitive and object types, each with a distinct set of operators. That is, aside from the awkward case of '+' for string concatenation, and the effects of auto-[un]boxing, and maybe some other bits and bobs I've forgotten about.
But note that the operators provided for primitives are all 'numbers'-related, since the primitive types are numeric. Apart from classes that represent numbers, re-using that same set of operators doesn't make much sense. And in that space there's scope for operator overloading for classes like BigInteger, BigDecimal, a Complex, maybe a Quarternion, and frankly that's about it.
I'd actually prefer that handful of classes to gain operators by making them primitives (like String), rather than allowing those operators to be overloaded for arbitrary classes.
Some languages, like Smalltalk and Haskell, allow new operators to be defined, rather than being restricted to the predefined set as in C++. On the face of it that's a great feature, until you account for operator fixity, which is where it gets really nasty. In Smalltalk, '5 + 4 * 2' mightn't mean what you think it does (it's 18, like on a cheapo desk calculator), and in Haskell 'a =*= [] &&- 4 --+ d' could be a mission to decipher.
This post is proof that Windows XP supports SSDs just fine.
It lacks TRIM support, but that's hardly a deal-breaker - chances are you'd be using the SSD primarily for OS and apps, which are mostly WORM and so mostly won't be subject to the performance degradation that TRIM helps to address.
But replying to the post rather than TFA, laptops already handle 10Gb/s via PCIe, HDMI, Thunderbolt, etc.
On the face of it, that's comparing pipfruit to citrus, as those aren't 'network' adapters as such. But it's reasonable to expect that the number of external connectors is trending to one, and expect that to make our equipment cheaper and more versatile.
I'm pretty happy using Copernic on Windows XP. Not sure how it compares with the built-in searches on Windows 7, but I gave up on both MS and Google desktop searches due to glitches.
With some improvements to the file preview in the search results, to display larger files faster and allow paging through matches in the document, I'd be a very happy Copernic user.
I just don't get all this talk of Apple being a marketing success. On the contrary, if Apple is spending big on marketing, they're doing a lousy job because I'm not seeing it.
Apple is a success as a design and engineering company, not a sales and marketing company.
For marketing successes, try Coca-Cola and McDonalds. That's good money for a mouthful of sugar and a mouthful of salt, respectively. Apple's products are much more wholesome and tasty, and basically sell themselves.
Slashdot Mirror
Thank you for at least trying to form a coherent argument why Chrome has it wrong, rather than just grabbing a pitchfork.
Well, it took me about five seconds to find the cache link in its new location. I don't think it's a bad change overall, since the cached page and the preview are (presumably) the same thing. It also unclutters the main results - the only per-link boilerplate I get now is the inane '+1' button.
If I had scripts blocked for google domains (or in general), I might be a little annoyed. But I also appreciate that if I disable scripts, I shouldn't expect everything to work. In Google's case, I don't block their scripts because I use their site heavily and trust it. It'd be a bit of a double standard to distrust Google's JavaScript but still trust their cache.
The 'farm people for power' storyline did seem pretty weak compared to the 'farm people for processing' alternative - after all, they could've just farmed sheep or hemp without having to worry about an uprising.
But here's a few plausible explanations:
Agreed. Have you ever seen anyone actually hold an iPhone up to their head?
And what's with the lack of custom SMS alert sounds? That's a bog standard phone feature completely absent on the iPhone. I need a loud, long noise to wake me up if I get paged in the middle of the night, not a tootle on a bicycle horn.
The magnetic connector thingy is a nice idea, but the power cord is too short and the join between the connector and the cable is too flimsy. I've got metal showing through on mine, its days may be numbered.
On balance I preferred the angled power connector on my old Asus laptop. But it's way better than the obscene protrusion on my Dell.
Erm, and does 'pixel' mean something different too now? We used to have 24 bits for each of those.
I blame the digital camera manufacturers and their bogus 'megapixel' counts. First we get a million bytes per megabyte and now this 8-(
Wikipedia has the Apple II, PET and TRS-80 on the leading edge for home computing, all being released in 1977. Those other machines you mentioned (aside from the TRS-80) came later, and were basically just new permutations and incremental enhancements on the same designs.
But of those first three, what I found particularly striking about the Apple II was:
Java strings, as in instances of class String, are always immutable.
Sure, there's a CharSequence interface that covers both immutable (String) and mutable (e.g. StringBuilder) strings, but it's not good style to take CharSequence arguments rather than String just for the sake of allowing mutable and immutable strings to be interchangeable.
Haskell for one is pure functional but does have support for mutable values - arrays in particular. The catch is that you can only modify them in the context of a monad, which can be pretty inconvenient.
No need to touch the mouse even, just type <win>search term<enter>.
That convenience makes up for a multitude of Start Menu sins, like wtf organized by publisher by default like I care.
I wouldn't stress about Goggles for facial recognition just yet... according to Goggles, my son is E.T. and my daughter is a wedding cake.
Seconded. VirtualBox now supports branching snapshots, which for my mileage is a killer feature.
A Capella!
http://www.youtube.com/watch?v=yjbpwlqp5Qw
The 123/789 thing doesn't usually bother me, except the time way back when I went to the bank to register for phone banking. They had me enter my initial 5-digit PIN on a PC numeric keypad - 789 on top, rather than the 123-on-top arrangement on the phone. Really did my head in.
And while I'm spouting barely relevant anecdotes around phone banking, for a while as an 'added security measure', my bank would have you enter just digits X and Y, presumably so that someone eavesdropping the call couldn't just replay the DTMF tones to gain access. But never mind that this gives an attacker (much?) better than 1% odds of gaining access with a (well?) educated random guess..!?
This smacks of the old days when people used TV antennas to get "free" TV, and then complained about commercials.
Heh, unlike now, where people get "free" TV and the commercials are the best thing on it.
That's not a problem if your 'real' email address has its own signature of dots. Unfortunately I wasn't aware of this trick when I created by GMail account, so that horse has already bolted.
Another approach is to not have a 'real' email address at all - instead, use a distinct address for every counterparty.
I don't buy that argument that ray tracing wins because of computational complexity - you won't break even on performance until you're ray tracing far more triangles than you have pixels to display them on.
And that means your average triangle isn't going to contribute to most frames at all, and when it does, you have aliasing!
In terms of the numbers of dollars and shares held by market participants across a trade, it is zero-sum if you ignore the transaction costs.
But that's not what people mean when they talk about 'the market', because there's this fixation on 'capitalization' whereby a value of a parcel of shares is deemed to be the number of shares multiplied by the current price per share. And that's there the zero-sum-ness of the market goes out the window, since the share price can vary independent of trading volume.
You could buy $100K worth of shares in a company today, pushing the share price up and increasing the company's market cap by $1M. Then you could sell those same shares tomorrow, pushing the share price down and reducing the market cap by $5M. There's no getting those numbers to sum to zero.
You're generously assuming that the network latency will substitute for the local latency.
More realistically, you'll get the baseline local latency of around 50-150ms, *plus* the network latency of let's say 50-150ms (don't forget more bandwidth means higher latency), and maybe some extra latency for processing the incoming video.
(Not so sure about the last bit, but I guess you'll be transferring more data from system RAM to the GPU than usual. Normal video rendering performance isn't comparable since that benefits from more buffering and pipelining.)
Anyway, all up you're looking at 100-300ms, which would garner it an *unplayable crap* rating from me 8-) (And yes, I'd apply that standard to console games as well.)
Compare to Smalltalk - once you decide that everything is an object, you pretty much have to either support operator overloading, or not support operators at all.
Java on the other hand has both primitive and object types, each with a distinct set of operators. That is, aside from the awkward case of '+' for string concatenation, and the effects of auto-[un]boxing, and maybe some other bits and bobs I've forgotten about.
But note that the operators provided for primitives are all 'numbers'-related, since the primitive types are numeric. Apart from classes that represent numbers, re-using that same set of operators doesn't make much sense. And in that space there's scope for operator overloading for classes like BigInteger, BigDecimal, a Complex, maybe a Quarternion, and frankly that's about it.
I'd actually prefer that handful of classes to gain operators by making them primitives (like String), rather than allowing those operators to be overloaded for arbitrary classes.
Some languages, like Smalltalk and Haskell, allow new operators to be defined, rather than being restricted to the predefined set as in C++. On the face of it that's a great feature, until you account for operator fixity, which is where it gets really nasty. In Smalltalk, '5 + 4 * 2' mightn't mean what you think it does (it's 18, like on a cheapo desk calculator), and in Haskell 'a =*= [] &&- 4 --+ d' could be a mission to decipher.
Well sort of - you can '+' two String instances together, but String is really a primitive type that acts like a class.
So in that sense this is no more overloading than is allowing '+' to stand for both integer and floating-point addition.
This post is proof that Windows XP supports SSDs just fine.
It lacks TRIM support, but that's hardly a deal-breaker - chances are you'd be using the SSD primarily for OS and apps, which are mostly WORM and so mostly won't be subject to the performance degradation that TRIM helps to address.
But replying to the post rather than TFA, laptops already handle 10Gb/s via PCIe, HDMI, Thunderbolt, etc.
On the face of it, that's comparing pipfruit to citrus, as those aren't 'network' adapters as such. But it's reasonable to expect that the number of external connectors is trending to one, and expect that to make our equipment cheaper and more versatile.
Let me introduce my friend the sarcamflex: ~!
I hope you hit it off. Observe:
I'm pretty happy using Copernic on Windows XP. Not sure how it compares with the built-in searches on Windows 7, but I gave up on both MS and Google desktop searches due to glitches.
With some improvements to the file preview in the search results, to display larger files faster and allow paging through matches in the document, I'd be a very happy Copernic user.
I just don't get all this talk of Apple being a marketing success. On the contrary, if Apple is spending big on marketing, they're doing a lousy job because I'm not seeing it.
Apple is a success as a design and engineering company, not a sales and marketing company.
For marketing successes, try Coca-Cola and McDonalds. That's good money for a mouthful of sugar and a mouthful of salt, respectively. Apple's products are much more wholesome and tasty, and basically sell themselves.