It's a good idea because OS's may need to change and you can easily keep all of your data on the HD (on a widely supported file system, on its own primary DOS partition) across OS changes.
It's a bad idea because of performance (as you note). On a live-CD system I'm familiar with, FreeSBIE, the files are compressed to save space so access is slowed by both the CD drive and decompression (first access of each file, at least).
It's also a bad idea because you lack the flexibility of easily updating the OS and third-party SW, and adding third-party SW. It's doable, of course, by using the hard drive, but then you're kind of defeating the purpose of CD-based OS's in the first place and adding a small management headache.
A good way, given enough HD space, might be to have a data partition plus one or two OS partitions. You could still run live-CD OS's, too.
Not weird. It's just a live-CD version OpenBSD, so it's following their versions. Definitely the best way to do it. PC-BSD (based on FreeBSD) is doing that now, too.
I've done the toothpaste/T-shirt method and it worked very well. But, it was a lot of work. 20 minutes of polishing and buffing that gooey toothpaste off. If I ever have to repair more than one disk, I will consider your advice.
Well put. This technology has great possibilities. It's up to the legal system what's done with it.
As for the idea of diffing two watermarked files and flipping the non-matching bits, that's an interesting idea and might work. But, it depends on how they encode it. There's a lot of bits there to play with. There no reason the location of the shifted bits needs to be the same. For that matter, there's no reason why each watermark couldn't include a shifted bit somewhere in the stream that is shifted on no other watermarked file. So, one possible result is that the new, mangled file can identify BOTH of the original files.
Now, I can see if you have current applications that you could not switch to OSS. But, the training issue I believe you are exaggerating greatly. It's hardly a "whole new way of doing things". We're talking about desktops, right? A point is still a point and a click is still a click. (I use FreeBSD. PC-BSD is reported to be a very user-freindly variant.)
And, if we're not talking about desktops, then I'm even more baffled. I've always found network administration and servers to be way easier to manage on unix than on w1ndows.
... My company recently announced that we're making the switch -- from win2000 to WindowsXP !! Woohoo. I'll probably have retired before we see Vista.
I think M$ is in trouble. Their business model seems to require churning perfectly good SW. Businesses have caught on. If it aint' broke.... (Or, if it ain't more broke than the upgrade.)
Tribes 2, Tribes 2, and Tribes 2. It's the 2++++++++ player game and completely awesome. (As I have mentioned before, and before:). )
I also like great games that I can play with my kids and still have a lot of fun. Worms Armageddon is the best (and still available). You can even handicap it (because there is a lot of strategy once you learn it). And, the old "Crash Team Racing" for the old PlayStation. It has a battle mode for up to four people (with mutli-tap) that is a blast. Hot damn, I think I'm going to play it tonight.
When people discuss this issue they usually forget to make a distinction between fault monitoring and data gathering for historical trending (like, what has my traffic looked like this past year). Most tools are only very good at one of these tasks, while the other is a so-so add-on.
For data collection and graphing, I've found cricket (link) to be very good. Once you've learned it, you can easily add new snmp OIDs into monitoring. In my experience that's been important because there are often new, sometimes proprietary, OIDs that need to be polled. I think it beats cacti for ease of use and clarity. It uses rrdtool for storage, so you can easily keep / roll-up data for a very long period of time without running out of disk space. Its "config tree" concept is great. It is the MRTG replacement, par excellence.
It has some trapping functionality, but it doesn't really seem to be equal to that of other tools that focus on fault monitoring. It's front-end/display is somewhat limited (but not hard to modify). But, I use it just for gathering the data and my colleagues have written a totally different front-end/display for it.
If your religion is really that important to you, are you really going to be happy with a career that conflicts with it so frequently? Geez, first they shove that evil-ution stuff down your throat in pre-med, then, before you know it, you're being asked to help rape victims.
Should you be "free" to drown your kids in the bathtub, if the Invisible Sky Fairy tells you to? No? Gee, maybe there really ought to be limits to freedom.
Wow, you really don't get the freedom thing. Do people really need you to tell them what their career should be?...
Anyone who is religious or believes in God is also a slack-jawed creationist?? You must sense you're on the losing side on this one, so you're playing the universal-slashdot-creationist trump card....
You frame the issue deceptively, it's only about "helping victims" -- and not about making new victims.
Again, freedom: If the person or organization is truly only being ASKED to "help", then they may say no and the matter is settled. But, the postion you've chosen to defend is to coerce people into doing what they believe is wrong.
Your last paragraph is really too much. Isn't the whole issue that some people don't want to kill kids!? Few people could have picked a more backwards counter-example.
An old teacher of mine related in class once that back in the 50's -- or 60's or something like that -- when the movie rating system was new, some theaters would print in advertisements that: "This movie is rated 'R' for violence, so it's OK to bring the kids!"
To me, this implies that in a large program, you have very large blocks.
Why don't you just divide them up into very small functions...?
You're right in that the issue that I'm addressing is really large blocks, particularly when there are nested blocks of any size as well. My tendency is to use functions for code that will be repeated at some point. My concern with your suggestion is that arbitrarily dividing up a block of related code into functions would just cause other readability problems (having, maybe, to go back and forth between functions).
But, it's a fair suggestion. In many cases, it starts off as a simple program that grows over time and could be cleaned up considerably with a judicious use of functions.
Now that we've established that I'm a hack (thanks), I still maintain that you can't have too many comments. I would recommend commenting the end of any block, including a function block, that might often have the top of the block not on the same screen, just to make quick navigation easier.
Amen.
You've got to have whitespace in any program anyway just to make it readable -- why shouldn't it do some work?!
And, please, I'd rather crawl through a ditch full of broken bottles than suffer through even one more "bracket audit." "... four, five,... uhm,.... One, two, three,...."
In a large program, though, I do tend to use comments to identify the end of blocks.
Python is a great language. People who get hung up on a piddling syntax thing like this are missing the point.
>>... they ought not be forced to either provide said procedure or to not treat anyone at all.
> No, they ought to go flip burgers or write AJAX code or do something besides work in the medical field under a professional oath and state-sanctioned license.
You really don't get this whole freedom thing, do you?
>... we have a single graph of a temperature proxy in the Sargasso sea, which may indicate a temperature change. On the other hand we have a several, independant, large studies of different proxies
Where is your data, schmuck? I put some out there and it is damn good data. Certainly, I don't claim that the surface temperature of the Sargasso sea is a perfect proxy for average global temperatures. But, it is an excellent indicator. You, on the other hand, want to claim that it means nothing and is -- what? -- a freak accident? Just like all of those Vikings farming in Greenland (in shorts!:) ) 500 years ago. Another quirk of our complex global environment? Get a life.
Slashdot Mirror
So they went from being the Linux with the best package manager to being the BSD with the worst package manager?
6.4 included official DVD images! (For i386 and amd64.) See the release announcement. You can get them via ftp. Or some torrents are here.
.
Why didn't he wear pajamas?!
It's a good idea because OS's may need to change and you can easily keep all of your data on the HD (on a widely supported file system, on its own primary DOS partition) across OS changes.
It's a bad idea because of performance (as you note). On a live-CD system I'm familiar with, FreeSBIE, the files are compressed to save space so access is slowed by both the CD drive and decompression (first access of each file, at least).
It's also a bad idea because you lack the flexibility of easily updating the OS and third-party SW, and adding third-party SW. It's doable, of course, by using the hard drive, but then you're kind of defeating the purpose of CD-based OS's in the first place and adding a small management headache.
A good way, given enough HD space, might be to have a data partition plus one or two OS partitions. You could still run live-CD OS's, too.
Not weird. It's just a live-CD version OpenBSD, so it's following their versions. Definitely the best way to do it. PC-BSD (based on FreeBSD) is doing that now, too.
:-P
Oh, and RTFA.
I've done the toothpaste/T-shirt method and it worked very well. But, it was a lot of work. 20 minutes of polishing and buffing that gooey toothpaste off. If I ever have to repair more than one disk, I will consider your advice.
Thank you. Yes, the bloody summary is wrong. ... Let's call it 1700 UTC, shall we?
So I take it that 2008 will be the Year of the BSD Desktop?
I'm thinking yes and hell yes. PC-BSD is going to be carried in Fry's and Microcenter (for starters).
And, whenever one is choosing an OS, even for the desktop, you've got consider what sort of crowd you'll be getting mixed up with.
"Unleash your desktop with PC-BSD!"
There have been some tests done to compare FreeBSD 7 performance to FreeBSD 6, and the gains are impressive.
See these slides by Kris Kennaway for more details on that.
Well put. This technology has great possibilities. It's up to the legal system what's done with it.
As for the idea of diffing two watermarked files and flipping the non-matching bits, that's an interesting idea and might work. But, it depends on how they encode it. There's a lot of bits there to play with. There no reason the location of the shifted bits needs to be the same. For that matter, there's no reason why each watermark couldn't include a shifted bit somewhere in the stream that is shifted on no other watermarked file. So, one possible result is that the new, mangled file can identify BOTH of the original files.
Only if the pirate has access to the reference file. Without that, he's SOL.
From: Ken Smith
....
l -- not yet working as I write this!
Date: Jan 15, 2007 12:29 AM
Subject: [FreeBSD-Announce] FreeBSD 6.2 Released
To: freebsd-announce@freebsd.org
So, wow, Slashdot was only an hour and eleven minutes ahead of the announcement.
If you're not on the announce mailing list, the full text should appear at this URL soon: http://www.freebsd.org/releases/6.2R/announce.htm
Now, I can see if you have current applications that you could not switch to OSS. But, the training issue I believe you are exaggerating greatly. It's hardly a "whole new way of doing things". We're talking about desktops, right? A point is still a point and a click is still a click. (I use FreeBSD. PC-BSD is reported to be a very user-freindly variant.)
And, if we're not talking about desktops, then I'm even more baffled. I've always found network administration and servers to be way easier to manage on unix than on w1ndows.
I think M$ is in trouble. Their business model seems to require churning perfectly good SW. Businesses have caught on. If it aint' broke
Tribes 2, Tribes 2, and Tribes 2. It's the 2++++++++ player game and completely awesome. (As I have mentioned before, and before :). )
I also like great games that I can play with my kids and still have a lot of fun. Worms Armageddon is the best (and still available). You can even handicap it (because there is a lot of strategy once you learn it). And, the old "Crash Team Racing" for the old PlayStation. It has a battle mode for up to four people (with mutli-tap) that is a blast. Hot damn, I think I'm going to play it tonight.
This was a two man show as I heard it. Senator Byrd was equally involved in the 'secret hold'.
... I guess as a democrat he "did it for the children" and that makes it OK.
When people discuss this issue they usually forget to make a distinction between fault monitoring and data gathering for historical trending (like, what has my traffic looked like this past year). Most tools are only very good at one of these tasks, while the other is a so-so add-on.
For data collection and graphing, I've found cricket (link) to be very good. Once you've learned it, you can easily add new snmp OIDs into monitoring. In my experience that's been important because there are often new, sometimes proprietary, OIDs that need to be polled. I think it beats cacti for ease of use and clarity. It uses rrdtool for storage, so you can easily keep / roll-up data for a very long period of time without running out of disk space. Its "config tree" concept is great. It is the MRTG replacement, par excellence.
It has some trapping functionality, but it doesn't really seem to be equal to that of other tools that focus on fault monitoring. It's front-end/display is somewhat limited (but not hard to modify). But, I use it just for gathering the data and my colleagues have written a totally different front-end/display for it.
Catering to adults is a good plan. Otherwise you end up with this. ... Think it over, dude.
Wow, you really don't get the freedom thing. Do people really need you to tell them what their career should be?
Again, freedom: If the person or organization is truly only being ASKED to "help", then they may say no and the matter is settled. But, the postion you've chosen to defend is to coerce people into doing what they believe is wrong.
Your last paragraph is really too much. Isn't the whole issue that some people don't want to kill kids!? Few people could have picked a more backwards counter-example.
An old teacher of mine related in class once that back in the 50's -- or 60's or something like that -- when the movie rating system was new, some theaters would print in advertisements that: "This movie is rated 'R' for violence, so it's OK to bring the kids!"
You're right in that the issue that I'm addressing is really large blocks, particularly when there are nested blocks of any size as well. My tendency is to use functions for code that will be repeated at some point. My concern with your suggestion is that arbitrarily dividing up a block of related code into functions would just cause other readability problems (having, maybe, to go back and forth between functions).
But, it's a fair suggestion. In many cases, it starts off as a simple program that grows over time and could be cleaned up considerably with a judicious use of functions.
Now that we've established that I'm a hack (thanks), I still maintain that you can't have too many comments. I would recommend commenting the end of any block, including a function block, that might often have the top of the block not on the same screen, just to make quick navigation easier.
Amen. You've got to have whitespace in any program anyway just to make it readable -- why shouldn't it do some work?!
... uhm, .... One, two, three, ...."
And, please, I'd rather crawl through a ditch full of broken bottles than suffer through even one more "bracket audit."
"... four, five,
In a large program, though, I do tend to use comments to identify the end of blocks.
Python is a great language. People who get hung up on a piddling syntax thing like this are missing the point.
>> ... they ought not be forced to either provide said procedure or to not treat anyone at all.
> No, they ought to go flip burgers or write AJAX code or do something besides work in the medical field under a professional oath and state-sanctioned license.
You really don't get this whole freedom thing, do you?
>... we have a single graph of a temperature proxy in the Sargasso sea, which may indicate a temperature change. On the other hand we have a several, independant, large studies of different proxies
:) ) 500 years ago. Another quirk of our complex global environment? Get a life.
Where is your data, schmuck? I put some out there and it is damn good data. Certainly, I don't claim that the surface temperature of the Sargasso sea is a perfect proxy for average global temperatures. But, it is an excellent indicator. You, on the other hand, want to claim that it means nothing and is -- what? -- a freak accident? Just like all of those Vikings farming in Greenland (in shorts!
Hottest in 400 years, yes. In several millenia? No f&)#ing way. Past temperatures are not a mystery. See the graph in this post:
http://s405.blogspot.com/2006/03/globaloney.html
Information will free you. Stop listening to experts who make money on your fear and LOOK INTO IT FOR YOURSELF.