It may be because I didn't play Halo 1 (though I didn't play HL1 before HL2 either), but when I went through Halo 2 I didn't think the story was that good. IMO, HL2's story is way better.
No, but you can go to dpreview.com and read the shutter release stats (and usually a full review) for pretty much every camera out there.
I did acknowledge that if you had kept reading. Perhaps you can tell me what shutter release timings correspond to "poor", "okay", "acceptable", and "good", because I certainly don't know that information. Are the times dpreview gives from the shutter button half-way down and AF set, or including the AF time, which will make these comparisons worthless because nothing will focus that fast?
In fact, I'm not even convinced they have that information for everything now. What's the shutter release time of the SD 780 IS?
this only matters in automatic modes, by the way - in manual mode you press the button and the shutter opens
If your camera has one. I was getting a small point-and-shoot to complement my DSLR, and looking for a camera that I could comfortably put in my pocket and carry around everywhere. I've yet to see a camera like this with a real manual mode, though I think that if a company came out with one and was the only one on the market for a while, and was a good camera otherwise too, they could do very well by it. I know I'm not the only non-newb at photography who wants a camera like that.
If I was unhappy with it, I could have returned it within 7 days for a full refund.
And then what, try another? What happens if it takes three or four tries? (Which is not totally unreasonable -- at the time I got mine, there were plenty of cameras on the shelves whose reaction times were not acceptable.) You've now made three or four returns, have cost Amazon a fair bit of money, have still tried far fewer cameras then I did in Circuit City in 10 minutes, and took probably a month to do it.
Online isn't a complete substitute. I can't walk into Amazon's camera department and give a bunch of the cameras a try to test their shutter release delay. I did do that in Circuit City and Best Buy a couple years ago. (Wound up buying a camera from CC; it was on sale, and with the free SD card that came with it, wasn't much worse than online, and I had reason to want it then.)
Of course, you can still go into Best Buy to give it a shot, then buy online, but if the hypothesis of the parent is correct and people would lose selection, that's not great. (You can also look at stuff like dpreview.com, since they actually have these numbers for some cameras, but it's hard to change "0.1sec" to "acceptable/unacceptable".)
I can think of plenty of situations where passing a simple bit of data between two platforms causes all sorts of problems without the added headache of having to deal with different "types" of strings, with different headers etc.
That's precisely why I say strings don't have to be portable. The fact that nothing else is portable either means that you already have to but a fair bit of effort into dealing with serializing the data in some portable fashion, and at that point a little extra effort to deal with the string header isn't going to be a huge deal. If you have some automatic method of dealing with structs already (some sort of IDL for instance), it may even be free.
...but then what do you do a decade from now when a 16bit string is laughably small?
Start tacking 32 bits in front. C has plenty of other places with implementation-dependent limits.
If you always tack the computer's native word size in front, you'll almost always have the maximum length be the size of an address space (which is a length limit already).
You get source code compatibility from this. Actual string objects don't have to be portable between a 16-bit and 32-bit implementation, because you almost never save information simply by doing a memory dump, and there will be lots more which isn't portable between those versions (like an 'int'). And if you could find a platform where moving the compiled code to a new system gives you more memory but still works (which seems unlikely since the pointer size would also have changed), if you've been living for years with shorter strings, you can probably continue to live with shorter strings.
In any case, these problems seem very minor in comparison to the fact that, had Ritchie or whoever made decisions about strings in C provided strings as a first-class type or at least used a better de facto standard (a string = some struct), what I suspect to be the second-most common reason for buffer overflows could have been stopped.
I have had several CD players that got a bit long-in-the-tooth and now have problems with reading some, but not all, CDs. The CDs tend to not be scratched, and will read fine in newer or better players. I don't know if this would help with this problem, but it might. And if it did, I would rather appreciate that, as it would extend the useful life of CD players.
Of course, they're about 5 or 10 years too late to be of much use.
These games work every bit as well and are every bit as fun even if every single song is from an indie group that no one has heard of before.
Well, this is true to an extent, but not entirely; I'll give three (related) reasons why I partially disagree:
1. I tend to have more fun playing songs that I like -- not to say that I won't like whatever indie music they put in (I've found a couple songs I liked because of GH and RB), but you have to be careful about putting a lot of unheard of stuff in there because of #3.
2. #1 goes especially true for the vocalist. I don't particularly enjoy singing songs that I don't know or like, though I don't know if I'm in the minority here.
3. Because of #1, if I'm trying to decide between buying GH or RB, I'm going to look at the set list and pick the one with the better set list. In a world where I were more motivated, I'd go and listen to the songs that I don't know in order to decide which set list is better. In the real world, I'm just going to ignore them, or in fact weight songs I don't know a little negatively. (I actually really don't like a lot of the music that's in either game, and just put up with it because when playing a good song, it's so much fun.) So if you drop out too much stuff that I know and know I like in favor of stuff that I haven't heard of, I'm going to get the other game, even if it's a few bucks more expensive.
If both games are full of stuff I haven't heard of, I may just stick with the games I have and play Cliffs of Dover or Don't Fear the Reaper some more.
So I agree that the GH/RB venue is a potential place for indy bands to break out... but I don't think that the companies would be particularly well-served by cranking up the indy music a lot.
I've been thinking that when I get out of grad school and get a job and a house, I want to plop a server box in the basement somewhere, and just scatter terminals running VLC or something similar around the house. I would think the Jack PC would be great for this, but the wall-wart in this article would be pretty good. Maybe have a separate gaming computer hooked up to the TV or something like that.
No duh. It's still a problem that you have if you put the computer into hibernate that you don't if you can swap batteries with the computer still running.
They don't need to do that anymore. They've implemented something called "Safe Sleep", i.e. hibernation. When the laptop goes to sleep it writes out the contents of RAM to the hard drive. If it loses power completely while it's asleep, like during a battery replacement, it will boot up just fine and reload the RAM contents from disk. And it actually works reliably.
This solves a lot of the problem, but not all of it. For instance, hibernate won't safe network connections, so if you're downloading something, sucks to be you.
The NC also has the effect of accentuating the unequal status of the original copyright holder and later contributors, because the NC isn't intended to make sure no profit can ever be made from the work, it's intended to make sure that nobody else can ever profit from it.
Alternately, you could argue that the NC is intended to make sure that if someone else wants to profit from your work, they have to go through you and give you a cut too.
Personally, I think it sounds way more reasonable when phrased that way.
The "specific types of hardware" mentioned in the latter quote appears to refer to netbooks. (The link is from the/. article on it.)
Even if it doesn't refer specifically to netbooks, the fact remains that the edition with the process limit is no longer the emerging markets edition, so the poster I originally replied to was still misinformed. ("You probably live in a western country - NO PROCESS LIMIT FOR YOU")
We know emerging markets have unique needs and we will offer Windows 7 Home Basic, only in emerging markets, for customers looking for an entry-point Windows experience on a full-size value PC.
We'll also continue to offer Windows Starter edition, which will only be offered pre-installed by an OEM. Windows Starter edition will now be available worldwide. This edition is available only in the OEM channel on new PCs limited to specific types of hardware.
Well, first, he didn't say *quite* that, but yes he did imply it. But he's wrong. There's no indication that such software would be exempt, and as the other poster says, that would be suicide in, e.g., the EU.
Gentoo is actually a distro I recommend sometimes to people who are interested in installing Linux on their computer for the first time, and it's not because I'm some MS shrill who is trying to scare people away from Linux.;-)
But if you have someone who has used Linux on someone else's computer or a school computer or something so that they are not scared of the command line and are have become pretty sure this is something that they'll actually be using rather than just installing because they want to try it and think they might like it, and they are willing to put in a little bit of effort, Gentoo is a really good choice. (That's a lot of 'if's, but Gentoo isn't exactly your typical newbie distro.)
I'm a big Gentoo fan (well, to the extent I'm a fan of any OS, which is not very much), because I think it is a solid distribution, but the main selling point for someone who can and is willing to deal with it is that in my experience, the documentation has just been outstanding. It's been a bit since I have really done any adminning of my own Linux box so this may be out of date, but I would do searches for Linux problems without specifying I was running Gentoo, and it seemed that half the time I would hit something on the Gentoo site anyway.
So - not exactly sad it's not specifically Linux (could be *BSD or any other flavor of *nix).
As someone who has used both Windows and Linux extensively, going to another flavor of *nix for me is like going from the frying pan into the oven, or something like that.
I have a ton of complaints about just about every OS in significant use, they just vary a lot. Windows tends to annoy me a little less than the others usually do, so I use that on my computers.
Slashdot Mirror
Totally agreed on this comment.
It may be because I didn't play Halo 1 (though I didn't play HL1 before HL2 either), but when I went through Halo 2 I didn't think the story was that good. IMO, HL2's story is way better.
No, but you can go to dpreview.com and read the shutter release stats (and usually a full review) for pretty much every camera out there.
I did acknowledge that if you had kept reading. Perhaps you can tell me what shutter release timings correspond to "poor", "okay", "acceptable", and "good", because I certainly don't know that information. Are the times dpreview gives from the shutter button half-way down and AF set, or including the AF time, which will make these comparisons worthless because nothing will focus that fast?
In fact, I'm not even convinced they have that information for everything now. What's the shutter release time of the SD 780 IS?
this only matters in automatic modes, by the way - in manual mode you press the button and the shutter opens
If your camera has one. I was getting a small point-and-shoot to complement my DSLR, and looking for a camera that I could comfortably put in my pocket and carry around everywhere. I've yet to see a camera like this with a real manual mode, though I think that if a company came out with one and was the only one on the market for a while, and was a good camera otherwise too, they could do very well by it. I know I'm not the only non-newb at photography who wants a camera like that.
If I was unhappy with it, I could have returned it within 7 days for a full refund.
And then what, try another? What happens if it takes three or four tries? (Which is not totally unreasonable -- at the time I got mine, there were plenty of cameras on the shelves whose reaction times were not acceptable.) You've now made three or four returns, have cost Amazon a fair bit of money, have still tried far fewer cameras then I did in Circuit City in 10 minutes, and took probably a month to do it.
How is this supposed to be better?
"Define 'portable'"? Is "I need something I can carry in my pocket, not a micro-notebook with full keyboard" (FTFS) enough of a definition?
Because Moore's law applies only to electronics (specifically, transistors) and not things with moving parts?
That's not totally unlike asking "Why does Moore's Law not apply to cars?"
Online isn't a complete substitute. I can't walk into Amazon's camera department and give a bunch of the cameras a try to test their shutter release delay. I did do that in Circuit City and Best Buy a couple years ago. (Wound up buying a camera from CC; it was on sale, and with the free SD card that came with it, wasn't much worse than online, and I had reason to want it then.)
Of course, you can still go into Best Buy to give it a shot, then buy online, but if the hypothesis of the parent is correct and people would lose selection, that's not great. (You can also look at stuff like dpreview.com, since they actually have these numbers for some cameras, but it's hard to change "0.1sec" to "acceptable/unacceptable".)
Did it really not occur to you that people might rely on the TV for information about whether or not there is actually a tornado coming at them?
Not that there aren't other sources of this information (particularly radio), but TV is often a particularly good one.
Well, to be fair, streets department rarely designate streets as the place to go if you want sex.
I can think of plenty of situations where passing a simple bit of data between two platforms causes all sorts of problems without the added headache of having to deal with different "types" of strings, with different headers etc.
That's precisely why I say strings don't have to be portable. The fact that nothing else is portable either means that you already have to but a fair bit of effort into dealing with serializing the data in some portable fashion, and at that point a little extra effort to deal with the string header isn't going to be a huge deal. If you have some automatic method of dealing with structs already (some sort of IDL for instance), it may even be free.
...but then what do you do a decade from now when a 16bit string is laughably small?
Start tacking 32 bits in front. C has plenty of other places with implementation-dependent limits.
If you always tack the computer's native word size in front, you'll almost always have the maximum length be the size of an address space (which is a length limit already).
You get source code compatibility from this. Actual string objects don't have to be portable between a 16-bit and 32-bit implementation, because you almost never save information simply by doing a memory dump, and there will be lots more which isn't portable between those versions (like an 'int'). And if you could find a platform where moving the compiled code to a new system gives you more memory but still works (which seems unlikely since the pointer size would also have changed), if you've been living for years with shorter strings, you can probably continue to live with shorter strings.
In any case, these problems seem very minor in comparison to the fact that, had Ritchie or whoever made decisions about strings in C provided strings as a first-class type or at least used a better de facto standard (a string = some struct), what I suspect to be the second-most common reason for buffer overflows could have been stopped.
It's solving a problem nobody has.
I have had several CD players that got a bit long-in-the-tooth and now have problems with reading some, but not all, CDs. The CDs tend to not be scratched, and will read fine in newer or better players. I don't know if this would help with this problem, but it might. And if it did, I would rather appreciate that, as it would extend the useful life of CD players.
Of course, they're about 5 or 10 years too late to be of much use.
That's a real good idea... install drivers when they plug in the USB device.
This idea is like autorun, except 10 times worse.
On DDR/HP (I don't know guitar hero and stuff like that) the licensed songs are only covers...
GH started that way almost entirely, but has moved to more master tracks. I think virtually everything in GH4 and RB2 are master tracks.
These games work every bit as well and are every bit as fun even if every single song is from an indie group that no one has heard of before.
Well, this is true to an extent, but not entirely; I'll give three (related) reasons why I partially disagree:
1. I tend to have more fun playing songs that I like -- not to say that I won't like whatever indie music they put in (I've found a couple songs I liked because of GH and RB), but you have to be careful about putting a lot of unheard of stuff in there because of #3.
2. #1 goes especially true for the vocalist. I don't particularly enjoy singing songs that I don't know or like, though I don't know if I'm in the minority here.
3. Because of #1, if I'm trying to decide between buying GH or RB, I'm going to look at the set list and pick the one with the better set list. In a world where I were more motivated, I'd go and listen to the songs that I don't know in order to decide which set list is better. In the real world, I'm just going to ignore them, or in fact weight songs I don't know a little negatively. (I actually really don't like a lot of the music that's in either game, and just put up with it because when playing a good song, it's so much fun.) So if you drop out too much stuff that I know and know I like in favor of stuff that I haven't heard of, I'm going to get the other game, even if it's a few bucks more expensive.
If both games are full of stuff I haven't heard of, I may just stick with the games I have and play Cliffs of Dover or Don't Fear the Reaper some more.
So I agree that the GH/RB venue is a potential place for indy bands to break out... but I don't think that the companies would be particularly well-served by cranking up the indy music a lot.
I've been thinking that when I get out of grad school and get a job and a house, I want to plop a server box in the basement somewhere, and just scatter terminals running VLC or something similar around the house. I would think the Jack PC would be great for this, but the wall-wart in this article would be pretty good. Maybe have a separate gaming computer hooked up to the TV or something like that.
No duh. It's still a problem that you have if you put the computer into hibernate that you don't if you can swap batteries with the computer still running.
When was the last time you saw a 3+ year old Apple laptop in the wild?
Last night.
I thought very long lives were a selling pt of Apple laptops, used as an argument why the extra cost is offset?
They don't need to do that anymore. They've implemented something called "Safe Sleep", i.e. hibernation. When the laptop goes to sleep it writes out the contents of RAM to the hard drive. If it loses power completely while it's asleep, like during a battery replacement, it will boot up just fine and reload the RAM contents from disk. And it actually works reliably.
This solves a lot of the problem, but not all of it. For instance, hibernate won't safe network connections, so if you're downloading something, sucks to be you.
The NC also has the effect of accentuating the unequal status of the original copyright holder and later contributors, because the NC isn't intended to make sure no profit can ever be made from the work, it's intended to make sure that nobody else can ever profit from it.
Alternately, you could argue that the NC is intended to make sure that if someone else wants to profit from your work, they have to go through you and give you a cut too.
Personally, I think it sounds way more reasonable when phrased that way.
And yet, Linux is better for having gone through the BitKeeper fiasco. So how did it demonstrate that?
The "specific types of hardware" mentioned in the latter quote appears to refer to netbooks. (The link is from the /. article on it.)
Even if it doesn't refer specifically to netbooks, the fact remains that the edition with the process limit is no longer the emerging markets edition, so the poster I originally replied to was still misinformed. ("You probably live in a western country - NO PROCESS LIMIT FOR YOU")
Source.
Tepples is the one who corrected me and linked that, from here.
Well, first, he didn't say *quite* that, but yes he did imply it. But he's wrong. There's no indication that such software would be exempt, and as the other poster says, that would be suicide in, e.g., the EU.
Gentoo is actually a distro I recommend sometimes to people who are interested in installing Linux on their computer for the first time, and it's not because I'm some MS shrill who is trying to scare people away from Linux. ;-)
But if you have someone who has used Linux on someone else's computer or a school computer or something so that they are not scared of the command line and are have become pretty sure this is something that they'll actually be using rather than just installing because they want to try it and think they might like it, and they are willing to put in a little bit of effort, Gentoo is a really good choice. (That's a lot of 'if's, but Gentoo isn't exactly your typical newbie distro.)
I'm a big Gentoo fan (well, to the extent I'm a fan of any OS, which is not very much), because I think it is a solid distribution, but the main selling point for someone who can and is willing to deal with it is that in my experience, the documentation has just been outstanding. It's been a bit since I have really done any adminning of my own Linux box so this may be out of date, but I would do searches for Linux problems without specifying I was running Gentoo, and it seemed that half the time I would hit something on the Gentoo site anyway.
So - not exactly sad it's not specifically Linux (could be *BSD or any other flavor of *nix).
As someone who has used both Windows and Linux extensively, going to another flavor of *nix for me is like going from the frying pan into the oven, or something like that.
I have a ton of complaints about just about every OS in significant use, they just vary a lot. Windows tends to annoy me a little less than the others usually do, so I use that on my computers.