1) I've never modified kernel code myself. But if I wanted to, or had to, I could. With an open-source kernel, you're free to change things on a whim. With a proprietary kernel, even if you have the technical ability, you're screwed.
2) I don't browse through random source code for fun (though sometimes for profit), but open-source software gives me that warm-and-fuzzy feeling because I prefer the whole community attitude (for the most part) over some monolithic corporation that's more interested in getting me to fork over $100 for their latest app. Others might take a strictly moralistic stance.
3) I suppose they could install anything they like. Choice is good.
4) Sure, Cocoa is nice. GCC is also nice. wxWidgets is even nicer, and easily portable across both OSes. Preferences vary from person to person, and YMMV.
5) Silly me, I didn't realize that modern *nix distros were strictly coding environments. And here I've been mixing audio, putting together home movies, editing photos, and doing my daily email/document/browsing/desktop-yada-yada on Linux. Must be a bug.
6) You misspelled "Mac" as "right".
"Oh? What's this? Red Hat donated $2 million to this project, and now they're getting used over OS X? Ah, that's why. So much for free and open."
Last I checked, if someone gives you $2M and an operating system with no costs attached to it, you aren't paying them anything -- ergo "free". If the source is "open", you can modify it as you like -- ergo "the other free".
Honestly I think the sniping community is working on entirely unscientific guesses.
Well, yes. There's nothing scientific about it, just past experiences and an understanding of human nature. All from a limited sample, of course.
Compare this to the auction where the camera sits at the minimum bid until 5 minutes before the close - It has 30 watchers, all of whom want a piece of this super rich deal. The probability that one of them is willing to spend more, given the high amount of attention, is vastly increased.
I know that experiences differ from person to person, but in my case, it's the exact opposite. If something sits low for the longest time, I find that more people bid at the end, but they all place low bids. When something is slowly increasing in price, people tend to go with the flow, and steadily increase their own bids. Thus a $5 camera might jump to $200 in a few seconds, but the same $5 camera with steady bidding will creep to $300 over a few days.
I have witnessed it several times, and sniping your max bid is almost invariably the way to win and save. If everyone bid their max amount from the start, what you say would be true. But they don't. I see on a regular basis auctions with bid histories that look like this:
user6 wins by sniping a max bid of $40 at the end, and wins it for $16. user1 likely gave up after his max bid of $10 was exceeded. user2-5 are too inexperienced to enter max bids, and keep bidding $1 more.
At least, that's been my experience ~95% of the time. I did, in fact, win my video card this way. YMMV.
I think that this little comment by sydb gives a good explanation as to why this is the case.
Seriously, if someone did a real statistical analysis on it...
I'd be interested in that result as well, and how it compares to, say, an empirical study of N auctions, where N is some random sampling of high-ticket items. I'm curious as to whether my hunches are more applicable to certain items, or to certain price ranges.
I say my max bid is $50. If someone bids $35, then autobid bids $36... for me up until $50. If no one else bids my low bid of $25 stands and that's all I have to pay.
Yes, but see my other comment here. Essentially, if you enter your max bid right away, you have a greater chance of reaching your max bid than if you slip it in toward the end of the auction, because a large amount of bidders will bid low amount instead of their maxes. You'll never go over your max, and you have a greater chance of getting the item for cheaper.
In your example, you bid $50 max, and can get bumped up pretty quickly. If you defer this to the end, there's a better chance you can get the item for, say, $40 after others try to snipe it for $35, $36, $37, etc.
In fact the snipers are often shooting themselves in the foot with their technique... - an item unnaturally low up until the end
Again, this is why sniping works. With a few exceptions, other snipers aren't putting in their max bids. They're putting in bids a few bucks more than the current price, trying to get the best deal. The one or two people that put in the max bids at the last second are the winners, barring any overly-competitive bidders as you suggest.
all of this sniping nonsense just proves how ridiculous people treat an auction like ebay (which usually leads them to grossly overpay).
While a good number of people do indeed overbid because they get drawn into the competitive spirit of winning the auction, there's a simple reason why sniping is still effective: savings. Sure, you can enter your maximum bid straight away. But there's also the psychological advantage of deferring your bid until the end, for items with no bids on them. As a simple example, let's say that I see some widget for which I would pay up to $20. I place that $20 max bid right away, starting at $1. Somebody else sees that there's action on the item, and places their max bid at, say, $10. Instantly the bids increase to $11. Is that someone else going to get caught up in the action and bid $21 or more? Possibly, and in that case, I simply walk away because my max was exceeded. Or I could win it at $11. Or at $12. Or up to $20. Whatever.
On the other hand, I know that a great majority of eBay bidders are into sniping themselves. If there's no bids on that widget with only 30 seconds to go, that other person might try to snipe it for $5. Having sold a few items, and bought quite a few more, I know that the most activity usually comes in the last minute, and usually for fairly low amounts. Me, I'll be entering my $20 maximum in the last 10 seconds. In this case, I'll walk away with my widget for a paltry $6 after bids. I seldom get outbid by someone who has the same strategy as this -- everyone else puts in a low bid, and maybe one more marginally higher bid before the time expires.
Sure, this is all relatively unknown info -- there's no guarantee there will be another bidder, there's no knowledge of what the bids will be, etc. But it's a pretty good approximation of how a lot of eBayers purchase items, so I'll try to use that past behaviour against them to get me the best bang for my buck.
The only way bidding before-hand benefits you is if a second last-minute sniper enters the same or a lower amount as you have previously bid. If you enter your max right at the last minute, you're guaranteed to get the same results as early bidding, but with a greater potential to save a few bucks.
install python, wxpython, pythoncard, egenix python extensions, java, jpype, & a few other python modules in linux. do the same in windows. see which takes longer.
You could do that, but unless you install each of these things on a regular basis instead of once or twice, then it's a fairly useless "benchmark". Something can take 1 hour to install vs. 30 minutes, but if it performs 15 seconds faster on a daily basis, where you use it 4 times a day... well, you onle need to use it for 30 days before the slower install pays off in terms of "performance".
Saying you want new music to always be backwards compatible is like saying you want all new music to play on an old vinyl deck.
As it stands right now, I can't be sure that a compact disc will play in my compact disc player until after I have purchased, opened, and tested it. This, of course, invalidates my return option.
Couldn't it be said that SCO is really asking for future plans on major additions to the kernel in asking for planned additions to 2.7, rather than simply asking for data about a piece of code which does not yet exist?
It could, but when you're dealing with lawyer-speak and all of that "letter of the law" mumbo-jumbo, asking for thw 2.7 kernel means asking for the 2.7 kernel. Otherwise they would have asked for "the latest development kernel".
Oh, of course, it's fully within your rights to remove the "software". Absolutely.
But is deleting the drivers the only way? Couldn't you nail the rootkit if you know it is there, and then get anything with $sys$ in it's name?
You might be able to do so, but it sure doesn't look that easy (ie. use of third-party tools). And even it it was relatively easy, Joe Q. Public would need to be tech-savvy enough to realize that he needed to download a "patch" from the Internet and apply it to his system to restore it. My parents, who just "put the CD in the drive and let it play" would have no clue what to do, short of calling me at 6:00 in the morning to rescue their computer.
So yes, I'm sure it can be done. But there'd still be a number of broken systems out there.
Anyways, nothing is the EULA says that I can't just go and delete it.
Except that, if you read through Mark Russinovich's blog, you'll see that it cripples your system when you do this.
When I logged in again I discovered that the CD drive was missing from Explorer. Deleting the drivers had disabled the CD [drive]. Now I was really mad... I know from my past work with device driver filter drivers that if you delete a filter driver's image, Windows fails to start the target driver.
He goes on to detail the steps that were necessary to bring his computer back to fully-functional condition. It's not for Joe Q. Public.
But a better lesson would probably be to just not carry the damn thing around - how hard is it to memorize 9 digits anyway?
How many times does he actually have to recite his SSN? In the rare instance that he needs it (employer, government) can't he say "I'll get back to you, I don't have it on me"?
My SIN (aka Canadian SSN) card lives at home in a drawer. Apart from tax time once a year, I haven't had to give the number out since I started my last job several years ago.
Why is it that almost more and more news items on slashdot are derived from the BBC?
Pfff, that's typical Slashdot Euro-centrism for you. Why are all of us North Americans constantly pushed to the side? Please, submitters, you must realize that the Internet != Britain.
I see no reason why a user shouldn't be able to download and run any program they find, as they should all be sandboxed appropriately that they cannot cause damage.
Sure, it may be a good start to remove some of the bugs, but who writes the sandbox? In what language? Is the sandbox itself sandboxed, to prevent being comprimised? If so, who writes that sandbox? In what language? Is that sandbox itself sandboxed, to prevent being comprimised? If so...
It's not an "obvious solution." It's an "obvious time-saver" when it comes to having to check for bugs.
Could you explain to me why a program downloaded from the internet has read and write access to every file on my computer?
I think that has more to say about your choice of operating system rather than the program itself.
It essentially means that if you write a program to be used on a network, you have to maintain and patch it forever because you'll never catch all the buffer overflows it contains.
I think you mean:
It essentially means that if you write a program to be used on a network, you have to maintain and patch it forever because you'll never catch all the programming errors, incorrect assumptions, and random unexpected behaviour introduced through unforseen run-time activity it contains.
Whoops. I realized after hitting "Submit" that I had mixed the "more than 4000 genes" and "20% of 24000 genes" (=4800) in my percentages. Using 4800 as the estimated number of gene-related patents, more accurate numbers are:
Universities: 28% of all gene-related patents
4800*0.28=1344 patents held
Private firms: 63% of all gene-related patents
4800*0.63=3024 patents held
2000/3024 = 66% of all firm-held patents held by Incyte
2000/4800 = 41.6% of all gene-related patents held by Incyte (not 50% as stated)
The web people believe (reasonably or not) that the form fields will be cleaned up by the backend people... The backend people believe (reasonably or not) that the data will be cleaned up by the web people.
Shouldn't they test that, to verify that at least one group has done it?
But yes, the manager should verify that it has, indeed, been tested.
Are they all needed? Probably not. But then again, they're not all installed by default anyway. I had to add mplayer to my setup, for example, because SuSE installed kaffeine with KDE. With Windows, I don't even have that option. I get Media Player, and nothing else.
Most Linux distros include XMMS, and nobody cries.
When it comes to packages, Microsoft is either Microsoft-centric, or just plain lacking.
Most Linux distros include a number of media players. For instance, my SuSE 9.3 install DVD included xmms, amarok, mplayer, kaffeine, totem, xine, etc. Many from different organizations/vendors/projects, etc.
Microsoft Windows ships with Microsoft Media Player. Period.
Most Linux distros include numerous text/document/office editors. Again, SuSE 9.3 ships with OpenOffice, AbiWord, KOffice, a zillion text editors, etc.
Microsoft Windows ships with Microsoft Notepad and Microsoft Wordpad. No office suite is included, and the one recommended is (no surprise) Microsoft Office.
The reason why nobody cries foul about distros shipping AppX is because there's generally AppY and AppZ bundled with it too. That's not to say that Microsoft should have to ship RealPlayer, or Mozilla, or whatever. It just explains why those used to modern Linux distros often feel that Microsoft is rather limiting in its offerings.
Slashdot Mirror
I'll feed the troll...
1) I've never modified kernel code myself. But if I wanted to, or had to, I could. With an open-source kernel, you're free to change things on a whim. With a proprietary kernel, even if you have the technical ability, you're screwed.
2) I don't browse through random source code for fun (though sometimes for profit), but open-source software gives me that warm-and-fuzzy feeling because I prefer the whole community attitude (for the most part) over some monolithic corporation that's more interested in getting me to fork over $100 for their latest app. Others might take a strictly moralistic stance.
3) I suppose they could install anything they like. Choice is good.
4) Sure, Cocoa is nice. GCC is also nice. wxWidgets is even nicer, and easily portable across both OSes. Preferences vary from person to person, and YMMV.
5) Silly me, I didn't realize that modern *nix distros were strictly coding environments. And here I've been mixing audio, putting together home movies, editing photos, and doing my daily email/document/browsing/desktop-yada-yada on Linux. Must be a bug.
6) You misspelled "Mac" as "right".
"Oh? What's this? Red Hat donated $2 million to this project, and now they're getting used over OS X? Ah, that's why. So much for free and open."
Last I checked, if someone gives you $2M and an operating system with no costs attached to it, you aren't paying them anything -- ergo "free". If the source is "open", you can modify it as you like -- ergo "the other free".
Easy peasy.
Honestly I think the sniping community is working on entirely unscientific guesses.
Well, yes. There's nothing scientific about it, just past experiences and an understanding of human nature. All from a limited sample, of course.
Compare this to the auction where the camera sits at the minimum bid until 5 minutes before the close - It has 30 watchers, all of whom want a piece of this super rich deal. The probability that one of them is willing to spend more, given the high amount of attention, is vastly increased.
I know that experiences differ from person to person, but in my case, it's the exact opposite. If something sits low for the longest time, I find that more people bid at the end, but they all place low bids. When something is slowly increasing in price, people tend to go with the flow, and steadily increase their own bids. Thus a $5 camera might jump to $200 in a few seconds, but the same $5 camera with steady bidding will creep to $300 over a few days.
I have witnessed it several times, and sniping your max bid is almost invariably the way to win and save. If everyone bid their max amount from the start, what you say would be true. But they don't. I see on a regular basis auctions with bid histories that look like this:
SUPER-MAGIC-VIDEO-CARD
07-Nov-05 12:05 $ 1.00 user1 (starting bid)
08-Nov-05 13:53 $ 2.00 user2
08-Nov-05 13:55 $ 5.00 user2
07-Nov-05 12:05 $ 6.00 user1
14-Nov-05 11:58 $11.00 user2
14-Nov-05 11:58 $12.00 user3
14-Nov-05 11:59 $13.00 user4
14-Nov-05 11:59 $14.00 user5
14-Nov-05 11:59 $15.00 user2
14-Nov-05 12:00 $16.00 user6 (sold)
user6 wins by sniping a max bid of $40 at the end, and wins it for $16. user1 likely gave up after his max bid of $10 was exceeded. user2-5 are too inexperienced to enter max bids, and keep bidding $1 more.
At least, that's been my experience ~95% of the time. I did, in fact, win my video card this way. YMMV.
I think that this little comment by sydb gives a good explanation as to why this is the case.
Seriously, if someone did a real statistical analysis on it...
I'd be interested in that result as well, and how it compares to, say, an empirical study of N auctions, where N is some random sampling of high-ticket items. I'm curious as to whether my hunches are more applicable to certain items, or to certain price ranges.
I say my max bid is $50. If someone bids $35, then autobid bids $36 ... for me up until $50. If no one else bids my low bid of $25 stands and that's all I have to pay.
Yes, but see my other comment here. Essentially, if you enter your max bid right away, you have a greater chance of reaching your max bid than if you slip it in toward the end of the auction, because a large amount of bidders will bid low amount instead of their maxes. You'll never go over your max, and you have a greater chance of getting the item for cheaper.
In your example, you bid $50 max, and can get bumped up pretty quickly. If you defer this to the end, there's a better chance you can get the item for, say, $40 after others try to snipe it for $35, $36, $37, etc.
In fact the snipers are often shooting themselves in the foot with their technique ... - an item unnaturally low up until the end
Again, this is why sniping works. With a few exceptions, other snipers aren't putting in their max bids. They're putting in bids a few bucks more than the current price, trying to get the best deal. The one or two people that put in the max bids at the last second are the winners, barring any overly-competitive bidders as you suggest.
all of this sniping nonsense just proves how ridiculous people treat an auction like ebay (which usually leads them to grossly overpay).
While a good number of people do indeed overbid because they get drawn into the competitive spirit of winning the auction, there's a simple reason why sniping is still effective: savings. Sure, you can enter your maximum bid straight away. But there's also the psychological advantage of deferring your bid until the end, for items with no bids on them. As a simple example, let's say that I see some widget for which I would pay up to $20. I place that $20 max bid right away, starting at $1. Somebody else sees that there's action on the item, and places their max bid at, say, $10. Instantly the bids increase to $11. Is that someone else going to get caught up in the action and bid $21 or more? Possibly, and in that case, I simply walk away because my max was exceeded. Or I could win it at $11. Or at $12. Or up to $20. Whatever.
On the other hand, I know that a great majority of eBay bidders are into sniping themselves. If there's no bids on that widget with only 30 seconds to go, that other person might try to snipe it for $5. Having sold a few items, and bought quite a few more, I know that the most activity usually comes in the last minute, and usually for fairly low amounts. Me, I'll be entering my $20 maximum in the last 10 seconds. In this case, I'll walk away with my widget for a paltry $6 after bids. I seldom get outbid by someone who has the same strategy as this -- everyone else puts in a low bid, and maybe one more marginally higher bid before the time expires.
Sure, this is all relatively unknown info -- there's no guarantee there will be another bidder, there's no knowledge of what the bids will be, etc. But it's a pretty good approximation of how a lot of eBayers purchase items, so I'll try to use that past behaviour against them to get me the best bang for my buck.
The only way bidding before-hand benefits you is if a second last-minute sniper enters the same or a lower amount as you have previously bid. If you enter your max right at the last minute, you're guaranteed to get the same results as early bidding, but with a greater potential to save a few bucks.
install python, wxpython, pythoncard, egenix python extensions, java, jpype, & a few other python modules in linux. do the same in windows. see which takes longer.
You could do that, but unless you install each of these things on a regular basis instead of once or twice, then it's a fairly useless "benchmark". Something can take 1 hour to install vs. 30 minutes, but if it performs 15 seconds faster on a daily basis, where you use it 4 times a day... well, you onle need to use it for 30 days before the slower install pays off in terms of "performance".
Saying you want new music to always be backwards compatible is like saying you want all new music to play on an old vinyl deck.
As it stands right now, I can't be sure that a compact disc will play in my compact disc player until after I have purchased, opened, and tested it. This, of course, invalidates my return option.
Couldn't it be said that SCO is really asking for future plans on major additions to the kernel in asking for planned additions to 2.7, rather than simply asking for data about a piece of code which does not yet exist?
It could, but when you're dealing with lawyer-speak and all of that "letter of the law" mumbo-jumbo, asking for thw 2.7 kernel means asking for the 2.7 kernel. Otherwise they would have asked for "the latest development kernel".
Anyone else read that "CISCO" instead of "CISO"?
I read it as "CRISCO" and thought, "mmm, lard."
Oh, of course, it's fully within your rights to remove the "software". Absolutely.
But is deleting the drivers the only way? Couldn't you nail the rootkit if you know it is there, and then get anything with $sys$ in it's name?
You might be able to do so, but it sure doesn't look that easy (ie. use of third-party tools). And even it it was relatively easy, Joe Q. Public would need to be tech-savvy enough to realize that he needed to download a "patch" from the Internet and apply it to his system to restore it. My parents, who just "put the CD in the drive and let it play" would have no clue what to do, short of calling me at 6:00 in the morning to rescue their computer.
So yes, I'm sure it can be done. But there'd still be a number of broken systems out there.
Anyways, nothing is the EULA says that I can't just go and delete it.
Except that, if you read through Mark Russinovich's blog, you'll see that it cripples your system when you do this.
He goes on to detail the steps that were necessary to bring his computer back to fully-functional condition. It's not for Joe Q. Public.
But a better lesson would probably be to just not carry the damn thing around - how hard is it to memorize 9 digits anyway?
How many times does he actually have to recite his SSN? In the rare instance that he needs it (employer, government) can't he say "I'll get back to you, I don't have it on me"?
My SIN (aka Canadian SSN) card lives at home in a drawer. Apart from tax time once a year, I haven't had to give the number out since I started my last job several years ago.
Why is it that almost more and more news items on slashdot are derived from the BBC?
Pfff, that's typical Slashdot Euro-centrism for you. Why are all of us North Americans constantly pushed to the side? Please, submitters, you must realize that the Internet != Britain.
> Looks like they over looked a few "things"
/, ?, <, >, :, ;, ", ', ., ,: 18 chars
Well, let's see now...
a-z: 26 chars
A-Z: 26 chars
1-0: 10 chars
!-): 10 chars
~, `, -, _, +, =, |, \,
{, }, [, ]: 4 chars
Tab, CR, LF, space: 4 chars
TOTAL: 98 chars
Unless there's some wonky Unicode comments in the kernel, I'd have to say that SCO accused them twice, and then some.
I see no reason why a user shouldn't be able to download and run any program they find, as they should all be sandboxed appropriately that they cannot cause damage.
Sure, it may be a good start to remove some of the bugs, but who writes the sandbox? In what language? Is the sandbox itself sandboxed, to prevent being comprimised? If so, who writes that sandbox? In what language? Is that sandbox itself sandboxed, to prevent being comprimised? If so...
It's not an "obvious solution." It's an "obvious time-saver" when it comes to having to check for bugs.
Could you explain to me why a program downloaded from the internet has read and write access to every file on my computer?
I think that has more to say about your choice of operating system rather than the program itself.
It essentially means that if you write a program to be used on a network, you have to maintain and patch it forever because you'll never catch all the buffer overflows it contains.
I think you mean:
It essentially means that if you write a program to be used on a network, you have to maintain and patch it forever because you'll never catch all the programming errors, incorrect assumptions, and random unexpected behaviour introduced through unforseen run-time activity it contains.
Obviously the classes would have to change slightly for internet retailers, mentioning talking fast is sort of moot if this were to go ahead.
Bah, piece of cake.
"Thankyouforpurchasingyouritemsatourstore, wereallyREALLYappreciateyourbusiness..."
Thank you, Mavis Beacon!
Whoops. I realized after hitting "Submit" that I had mixed the "more than 4000 genes" and "20% of 24000 genes" (=4800) in my percentages. Using 4800 as the estimated number of gene-related patents, more accurate numbers are:
Universities: 28% of all gene-related patents
4800*0.28=1344 patents held
Private firms: 63% of all gene-related patents
4800*0.63=3024 patents held
2000/3024 = 66% of all firm-held patents held by Incyte
2000/4800 = 41.6% of all gene-related patents held by Incyte (not 50% as stated)
How can someone spell Tetris wrong three times in a row?
I don't know, but I suspect that had it been spelled incorrectly a fourth time, there would have been some bonus points awarded.
D'oh! Good thing I'm not in charge of optimization...
...Google aims to maim AIM as it gains GAIM's main brain?
Wow, I used to love playing "ladders" as a kid.
ajax (start)
ajar
alar
alan
plan
pean
peat
pent
rent
rest (end)
9 steps isn't elegant, but works.
I've always wanted an iPod Kilo.
The web people believe (reasonably or not) that the form fields will be cleaned up by the backend people ... The backend people believe (reasonably or not) that the data will be cleaned up by the web people.
Shouldn't they test that, to verify that at least one group has done it?
But yes, the manager should verify that it has, indeed, been tested.
Well, it asks the question.
Are they all needed? Probably not. But then again, they're not all installed by default anyway. I had to add mplayer to my setup, for example, because SuSE installed kaffeine with KDE. With Windows, I don't even have that option. I get Media Player, and nothing else.
Most Linux distros include XMMS, and nobody cries.
When it comes to packages, Microsoft is either Microsoft-centric, or just plain lacking.
Most Linux distros include a number of media players. For instance, my SuSE 9.3 install DVD included xmms, amarok, mplayer, kaffeine, totem, xine, etc. Many from different organizations/vendors/projects, etc.
Microsoft Windows ships with Microsoft Media Player. Period.
Most Linux distros include numerous text/document/office editors. Again, SuSE 9.3 ships with OpenOffice, AbiWord, KOffice, a zillion text editors, etc.
Microsoft Windows ships with Microsoft Notepad and Microsoft Wordpad. No office suite is included, and the one recommended is (no surprise) Microsoft Office.
The reason why nobody cries foul about distros shipping AppX is because there's generally AppY and AppZ bundled with it too. That's not to say that Microsoft should have to ship RealPlayer, or Mozilla, or whatever. It just explains why those used to modern Linux distros often feel that Microsoft is rather limiting in its offerings.