Or don't you have a clue what TCO and ROI stand for? STFU, troll.
That's very nice. You have heard a couple of financial sounding acronyms and are now a finance expert? But a business is not just about a balance sheet, it's also about a cash flow. In many cases, Linux might have the same TCO or ROI at the end of the year as another option, but the initial investment would be low and so the I in the ROI would more regular and the TCO would be spread over time rather than requiring a single lump expense at the beginning of the accounting period.
Also, no study has conclusively shown that Linux has a higher TCO than other options. Nor is it at all intuitive that a Free Software would have a lower ROI than a proprietary solution. In fact, quite the opposite, as the potential for minimizing Linux expenses is much greater than the potential for minimizing proprietary expenses.
Lower barriers to entry are essential for the marketplace to work correctly and Free Software (e.g. Linux) does just that.
Um, in this case, you're off base. This was clearly the Linux kernel changelog and RMS has never insisted that the kernel be called GNU anything, except in those cases where the kernel is a GNU project, like HURD.
It would have helped a lot, I think, if people like Bruce Perens had worked as part of the Free Software movement instead of confusing the issues with all the talk about Open Source. Not only that, the FSF has a clearly articulated and prominent (i.e. linked from their front web page) stand against software patents. Perhaps Bruce could work with OSI to get a similarly obvious kind of thing over at opensource.org. In fact, I can't find anything about patents at all on the OSI site. So it's easy to see why this European Open Source guy would be confused.
Otherwise, I guess it's okay to bash RMS all day long, but question the OSI or Bruce Perens (respectfully I might add-- I never once called him a "sell out" or anything like that, but people regularly get modded up for calling RMS a dirty hippy) and I'm a troll?
Or does the OSI have a clearly articulated, readily available stance on software patents and I'm completely stupid and missed it? Well?
Alan Cox less blinded by ideals than RMS? I personally don't think censoring parts of the Linux changelog to make a political statement bear out that assertion. But the point isn't to have a single leader, it's to have an organization with properly selected spokepersons. Personally I'd prefer seeing someone like Bradley Kuhn from the FSF do the public representation.
And when it comes down to it, it's actually good to have more than a few people show up to these things representing various organizations. So it would be better to also get the Tim O'Reillys and the someones from Red Hat and maybe the guys from Lindows-- assuming they are also opposed to software patents-- to show up (just to name a few obvious examples, others that would be good to find would be a professor or two, or a scientist, or some non-software business exec whose business benefits from Free Software). Then it would be clearer that there is a broad base of opposition to this.
It would have helped a lot, I think, if people like Bruce Perens had worked as part of the Free Software movement instead of confusing the issues with all the talk about Open Source. Not only that, the FSF has a clearly articulated and prominent (i.e. linked from their front web page) stand against software patents. Perhaps Bruce could work with OSI to get a similarly obvious kind of thing over at opensource.org. In fact, I can't find anything about patents at all on the OSI site. So it's easy to see why this European Open Source guy would be confused.
No. It doesn't. Auto-preview/auto-execute are much different than "open attachement with application". The only thing saving the file to disk before opening it in an application does is make more work for the user. If the file somehow exploits a weakness in the application it doesn't matter whether you opened "directly" from the mail program, or whether you saved it to your drive first. You'll notice I am not proposing that you be able to run attachments *as* applications. But again, if the user is just going to save it to disk and then execute it from there. You aren't really saving anything.
People are typically the weakest link in any security chain. No amount of inconvenience is going to improve that. In fact, I think if you look closely you'll find that the more annoying security measures are to people, the more risky their behavior will become as they do things that help them through those measures with maximum ease.
This is why so many people choose things like their own name as a password, or simply stick the password next to the computer on a piece of notepaper, or whatever. You aren't going to make a system more secure by making it less convenient unless your users know and understand why it's less convenient.
So just to prove that I did my homework: you can "pin" a version by editing the world file and doing a = before the package name. But I'll have to see how this works out in terms of not causing problems with the dependency tree.
They both happen. Personally I think I got hit by rm -rf.*, but rm -rf / usually happens when you accidentally do something like rm -rf / mydir, where you intended to just remove/mydir.:)
Why shouldn't there be an ability to open the attachment directly with an appropriate application? The extra step of saving the file to disk isn't ultimately making anything safer.
And you're kidding about Outlook right? I've seen email viruses that worked in Pine. All that was required were gullible users who would hasten to forward an email to their entire addressbook because they'd read a line in a scare-mail like "forward this to everyone you know immediately".:)
Just as scary, and not unheard of on Linux is the whole trend towards displaying HTML right in the mailer. What a nightmare.
Thank you. That was excellent. To further your fine work: if Free Software doesn't take the time to reimplement a bunch of this stuff, end users are going to take one look at a GNU/Linux system and say "Well, where's the main/start/apple menu?" and "So where's the spreadsheet program?" The response from the Free Software movement very well can't be, "we didn't want to just be a bunch of copycats". The people I want to encourage to move to Free Software are resistant precisely because no one out there has successfully copied the proprietary software they feel dependent on. No one has ever told me, "Well, your Linux just isn't innovative enough."
Strangely enough, I still think there is plenty of innovation going on in Free Software (and plain old Open Source) at the same time. The changes are often small and subtle, but no less powerful when you stop to look at them.
I know about the ~arch keyword. I don't think I had that or anything else set that should have put me into unstable territory. I don't recall doing anything that would have interfered with my KDE upgrade. And I'm not a newbie-- either to Linux or Gentoo. Generally I can solve the problems I have. But what I'd like is even more control over what gets updated and what doesn't and why. I suppose I could do emerge -p -u world, check the package list against the security updates to find matches, and then emerge inject anything that I don't really want to update, but that could create real problems later if one of those packages is a dependency for something else. So maybe I just follow the security updates list and never do emerge -u world, instead only updating packages with security updates.
So I will reread the portage docs (again), but I don't think there's an answer in there. Which is why I'm whining here (and please note that I'm the only one whining in this Gentoo thread who actually uses Gentoo and has pointed out real problems with the system). I want to have more control over what gets updated, when and why. It would be nice if there were an emerge switch or a USE keyword that one could use to pick up just the security updates or to intelligently defer some updates. Like so I can not update some packages until another package I do want to update needs that first package updated.
Perhaps it's time for me to give back to Gentoo by participating more fully in Bugzilla and the forums. And maybe I can bribe developers to fix my favorite problems and/or implement some of these ideas. I do get the sense they're working in this direction anyway, which is part of why I've decided to stick with Gentoo for the time being.
You're not even listening to me. I love being on the cutting edge. But how is it that when I'm updating Gentoo it goes from working software to completely fucked. I gave three examples of software that was fine before "upgrading" that broke as a result of the upgrade. That's not an upgrade. I can do a better job of keeping up-to-date than that by just maintaining everything by hand. How hard would it be for Gentoo to have some way to tag certain updates as security-related and others as simply improvements? And then maybe a way to defer any or all non-security updates for when I have the time to deal with all the broken packages that have seemed to result from each and every emerge -u world I've run? I think they're moving in that direction, actually... so maybe I'll just chime in somewhere appropriate with this sentiment.
Actually the revenue lost is that which should have been provided by the person distributing the file. Under normal circumstances this person would be paying ASCAP/BMI fees.
What I know is that Gentoo is perma-beta software. When the hell are they going to stop putting updates in the main release and make it possible to get security-only updates as a default?
The other day I did an emerge -u world and an application I'd just installed the day before broke. With an error message that my current version of the nvidia drivers wasn't current enough, which they were, no less.
And this is common. My entire KDE system broke. And kept getting more and more broken over time. dvd::rip also has gotten more broken over time. Not that I want to see Gentoo go as far off the stability deep-end as Debian (where your stable machines are usually running software that was released forever ago). But it would be nice to be able to filter all non-security updates, on the "if it ain't broke don't fix it" principle.
Yeah, with people who know what the words mean, I tend to use the name of the system "Debian", "Gentoo", "Red Hat", etc. In print, when writing for a technical audience, I'd use "GNU/Linux" to refer to any Linux-based OS. When talking to non-techies, I just use "Linux". But I also do my best to mention what the meaning behind the phrase "free software" is at some point. That way people understand that there's more to Linux than just not paying. People tend to think you get what you pay for, and if they think something is just "free" that it can't be worth very much. And in some respect they're right. There's a lot of "missing" free software (like Adobe Photoshop or Adobe Illustrator-- just not quite there yet with the GIMP and ???). So it has to be about something more than saving a few bucks. And it is: freedom.
I usually say Microsoft Windows (or MS Windows) when I am writing about that particular piece of software. Window(s) is a generic term in computer software that could (and often does) refer to products besides MS Windows. I don't think they should get to own that word as a de facto trademark (there is already case history that suggests they won't be allowed to call it a registered trademark). But I'm a GNU/Linux zealot, so obviously I'm a stickler for these kinds of details.:)
That is a good point. I'd forgotten about accesskey. I still don't think making an ASCII GUI is somehow more likely to be productive than a properly constructed actual GUI. A properly constructed set of command line tools, however...:)
Does calling me names make you're argument stronger? No wonder you're posting AC. Your argument is as eloquent as an angry primate flinging dung at tourists in the zoo.
You're full of it. The notion of property depends on scarcity. Notice we don't generally speak of owning air. This is because under normal circumstances (i.e. you're not in an air-tight container or an excessively polluted area) anyone who wants to take a breath can do so without restricting the ability of others to also take a breath. Air is not a rivalrous resource. Neither are ideas.
The four inch stack of papers on your desk is a four inch stack of papers. You could burn them all to a crisp and the ideas would still exist in your mind and in the minds of anyone who'd read the papers. Ideas don't exist except as processes within minds. In order to exchange ideas we sometimes fix their form on some piece of physical property.
If ideas were property, we wouldn't have a second class of legislation which applies to them that is separate from existing legislation which prohibits theft, larceny, vandalism, etc. If I steal your book, I have taken a thing from your possession. If I "take" your idea, how is it that you still have the use of it?
Copyright is the legal recognition of an author's natural property rights over his creation.
BS. A book is a physical object. This can be property. The story in the book is a mental construct that develops as a result of reading the book. This cannot be property because it does not exist except as an experience in the mind of the beholder.
So how hard is it to provide training for slow-movers in faster ways to use the tool and then having contests for cash money or gift certificates or a paid day off or whatever? That's got to be cheaper than attempting to go through an entire development cycle just because your employees don't give enough of a [expletive deleted] to work smarter rather than harder.
You're kidding. That's a horrible idea. How do you implement keyboard shortcuts in links/lynx? How do you do menus and stuff like that? Not to mention the additional security headaches involved in writing web interfaces...
What this person wants to do is get over his leetness and learn to make a GUI application that behaves how he/she thinks a console app should work. There's no reason you can't bind keys the exact same way in a GUI that you would in a console app. And really, a console app created with something like curses is a GUI. Just not a very sophisticated one. A true command line application accepts lines in and spits lines out without mocking up a pathetic GUI using ASCII.
Slashdot Mirror
Or don't you have a clue what TCO and ROI stand for? STFU, troll.
That's very nice. You have heard a couple of financial sounding acronyms and are now a finance expert? But a business is not just about a balance sheet, it's also about a cash flow. In many cases, Linux might have the same TCO or ROI at the end of the year as another option, but the initial investment would be low and so the I in the ROI would more regular and the TCO would be spread over time rather than requiring a single lump expense at the beginning of the accounting period.
Also, no study has conclusively shown that Linux has a higher TCO than other options. Nor is it at all intuitive that a Free Software would have a lower ROI than a proprietary solution. In fact, quite the opposite, as the potential for minimizing Linux expenses is much greater than the potential for minimizing proprietary expenses.
Lower barriers to entry are essential for the marketplace to work correctly and Free Software (e.g. Linux) does just that.
Too bad you got a flamebait moderation. That was hilarious!
Oh sheesh. Now I feel like an overly zealous prick. I knew you were being funny and I should've left it alone. :)
And since Cox actually lives in Europe, in this case he probably would be the better choice between the two.
Um, in this case, you're off base. This was clearly the Linux kernel changelog and RMS has never insisted that the kernel be called GNU anything, except in those cases where the kernel is a GNU project, like HURD.
Someone tell me what is "trolling" in this post:
It would have helped a lot, I think, if people like Bruce Perens had worked as part of the Free Software movement instead of confusing the issues with all the talk about Open Source. Not only that, the FSF has a clearly articulated and prominent (i.e. linked from their front web page) stand against software patents. Perhaps Bruce could work with OSI to get a similarly obvious kind of thing over at opensource.org. In fact, I can't find anything about patents at all on the OSI site. So it's easy to see why this European Open Source guy would be confused.
Otherwise, I guess it's okay to bash RMS all day long, but question the OSI or Bruce Perens (respectfully I might add-- I never once called him a "sell out" or anything like that, but people regularly get modded up for calling RMS a dirty hippy) and I'm a troll?
Or does the OSI have a clearly articulated, readily available stance on software patents and I'm completely stupid and missed it? Well?
Alan Cox less blinded by ideals than RMS? I personally don't think censoring parts of the Linux changelog to make a political statement bear out that assertion. But the point isn't to have a single leader, it's to have an organization with properly selected spokepersons. Personally I'd prefer seeing someone like Bradley Kuhn from the FSF do the public representation.
And when it comes down to it, it's actually good to have more than a few people show up to these things representing various organizations. So it would be better to also get the Tim O'Reillys and the someones from Red Hat and maybe the guys from Lindows-- assuming they are also opposed to software patents-- to show up (just to name a few obvious examples, others that would be good to find would be a professor or two, or a scientist, or some non-software business exec whose business benefits from Free Software). Then it would be clearer that there is a broad base of opposition to this.
It would have helped a lot, I think, if people like Bruce Perens had worked as part of the Free Software movement instead of confusing the issues with all the talk about Open Source. Not only that, the FSF has a clearly articulated and prominent (i.e. linked from their front web page) stand against software patents. Perhaps Bruce could work with OSI to get a similarly obvious kind of thing over at opensource.org. In fact, I can't find anything about patents at all on the OSI site. So it's easy to see why this European Open Source guy would be confused.
No. It doesn't. Auto-preview/auto-execute are much different than "open attachement with application". The only thing saving the file to disk before opening it in an application does is make more work for the user. If the file somehow exploits a weakness in the application it doesn't matter whether you opened "directly" from the mail program, or whether you saved it to your drive first. You'll notice I am not proposing that you be able to run attachments *as* applications. But again, if the user is just going to save it to disk and then execute it from there. You aren't really saving anything.
People are typically the weakest link in any security chain. No amount of inconvenience is going to improve that. In fact, I think if you look closely you'll find that the more annoying security measures are to people, the more risky their behavior will become as they do things that help them through those measures with maximum ease.
This is why so many people choose things like their own name as a password, or simply stick the password next to the computer on a piece of notepaper, or whatever. You aren't going to make a system more secure by making it less convenient unless your users know and understand why it's less convenient.
So just to prove that I did my homework: you can "pin" a version by editing the world file and doing a = before the package name. But I'll have to see how this works out in terms of not causing problems with the dependency tree.
They both happen. Personally I think I got hit by rm -rf .*, but rm -rf / usually happens when you accidentally do something like rm -rf / mydir, where you intended to just remove /mydir. :)
Oh come on. Accidentally deleting the entire filesystem is a rite of passage. :)
Why shouldn't there be an ability to open the attachment directly with an appropriate application? The extra step of saving the file to disk isn't ultimately making anything safer.
:)
And you're kidding about Outlook right? I've seen email viruses that worked in Pine. All that was required were gullible users who would hasten to forward an email to their entire addressbook because they'd read a line in a scare-mail like "forward this to everyone you know immediately".
Just as scary, and not unheard of on Linux is the whole trend towards displaying HTML right in the mailer. What a nightmare.
Thank you. That was excellent. To further your fine work: if Free Software doesn't take the time to reimplement a bunch of this stuff, end users are going to take one look at a GNU/Linux system and say "Well, where's the main/start/apple menu?" and "So where's the spreadsheet program?" The response from the Free Software movement very well can't be, "we didn't want to just be a bunch of copycats". The people I want to encourage to move to Free Software are resistant precisely because no one out there has successfully copied the proprietary software they feel dependent on. No one has ever told me, "Well, your Linux just isn't innovative enough."
Strangely enough, I still think there is plenty of innovation going on in Free Software (and plain old Open Source) at the same time. The changes are often small and subtle, but no less powerful when you stop to look at them.
I know about the ~arch keyword. I don't think I had that or anything else set that should have put me into unstable territory. I don't recall doing anything that would have interfered with my KDE upgrade. And I'm not a newbie-- either to Linux or Gentoo. Generally I can solve the problems I have. But what I'd like is even more control over what gets updated and what doesn't and why. I suppose I could do emerge -p -u world, check the package list against the security updates to find matches, and then emerge inject anything that I don't really want to update, but that could create real problems later if one of those packages is a dependency for something else. So maybe I just follow the security updates list and never do emerge -u world, instead only updating packages with security updates.
So I will reread the portage docs (again), but I don't think there's an answer in there. Which is why I'm whining here (and please note that I'm the only one whining in this Gentoo thread who actually uses Gentoo and has pointed out real problems with the system). I want to have more control over what gets updated, when and why. It would be nice if there were an emerge switch or a USE keyword that one could use to pick up just the security updates or to intelligently defer some updates. Like so I can not update some packages until another package I do want to update needs that first package updated.
Perhaps it's time for me to give back to Gentoo by participating more fully in Bugzilla and the forums. And maybe I can bribe developers to fix my favorite problems and/or implement some of these ideas. I do get the sense they're working in this direction anyway, which is part of why I've decided to stick with Gentoo for the time being.
You're not even listening to me. I love being on the cutting edge. But how is it that when I'm updating Gentoo it goes from working software to completely fucked. I gave three examples of software that was fine before "upgrading" that broke as a result of the upgrade. That's not an upgrade. I can do a better job of keeping up-to-date than that by just maintaining everything by hand. How hard would it be for Gentoo to have some way to tag certain updates as security-related and others as simply improvements? And then maybe a way to defer any or all non-security updates for when I have the time to deal with all the broken packages that have seemed to result from each and every emerge -u world I've run? I think they're moving in that direction, actually... so maybe I'll just chime in somewhere appropriate with this sentiment.
Actually the revenue lost is that which should have been provided by the person distributing the file. Under normal circumstances this person would be paying ASCAP/BMI fees.
Yeah. Ok. You just keep telling yourself that.
What I know is that Gentoo is perma-beta software. When the hell are they going to stop putting updates in the main release and make it possible to get security-only updates as a default?
The other day I did an emerge -u world and an application I'd just installed the day before broke. With an error message that my current version of the nvidia drivers wasn't current enough, which they were, no less.
And this is common. My entire KDE system broke. And kept getting more and more broken over time. dvd::rip also has gotten more broken over time. Not that I want to see Gentoo go as far off the stability deep-end as Debian (where your stable machines are usually running software that was released forever ago). But it would be nice to be able to filter all non-security updates, on the "if it ain't broke don't fix it" principle.
Yeah, with people who know what the words mean, I tend to use the name of the system "Debian", "Gentoo", "Red Hat", etc. In print, when writing for a technical audience, I'd use "GNU/Linux" to refer to any Linux-based OS. When talking to non-techies, I just use "Linux". But I also do my best to mention what the meaning behind the phrase "free software" is at some point. That way people understand that there's more to Linux than just not paying. People tend to think you get what you pay for, and if they think something is just "free" that it can't be worth very much. And in some respect they're right. There's a lot of "missing" free software (like Adobe Photoshop or Adobe Illustrator-- just not quite there yet with the GIMP and ???). So it has to be about something more than saving a few bucks. And it is: freedom.
I usually say Microsoft Windows (or MS Windows) when I am writing about that particular piece of software. Window(s) is a generic term in computer software that could (and often does) refer to products besides MS Windows. I don't think they should get to own that word as a de facto trademark (there is already case history that suggests they won't be allowed to call it a registered trademark). But I'm a GNU/Linux zealot, so obviously I'm a stickler for these kinds of details. :)
That is a good point. I'd forgotten about accesskey. I still don't think making an ASCII GUI is somehow more likely to be productive than a properly constructed actual GUI. A properly constructed set of command line tools, however... :)
Does calling me names make you're argument stronger? No wonder you're posting AC. Your argument is as eloquent as an angry primate flinging dung at tourists in the zoo.
You're full of it. The notion of property depends on scarcity. Notice we don't generally speak of owning air. This is because under normal circumstances (i.e. you're not in an air-tight container or an excessively polluted area) anyone who wants to take a breath can do so without restricting the ability of others to also take a breath. Air is not a rivalrous resource. Neither are ideas.
The four inch stack of papers on your desk is a four inch stack of papers. You could burn them all to a crisp and the ideas would still exist in your mind and in the minds of anyone who'd read the papers. Ideas don't exist except as processes within minds. In order to exchange ideas we sometimes fix their form on some piece of physical property.
If ideas were property, we wouldn't have a second class of legislation which applies to them that is separate from existing legislation which prohibits theft, larceny, vandalism, etc. If I steal your book, I have taken a thing from your possession. If I "take" your idea, how is it that you still have the use of it?
Copyright is the legal recognition of an author's natural property rights over his creation.
BS. A book is a physical object. This can be property. The story in the book is a mental construct that develops as a result of reading the book. This cannot be property because it does not exist except as an experience in the mind of the beholder.
So how hard is it to provide training for slow-movers in faster ways to use the tool and then having contests for cash money or gift certificates or a paid day off or whatever? That's got to be cheaper than attempting to go through an entire development cycle just because your employees don't give enough of a [expletive deleted] to work smarter rather than harder.
You're kidding. That's a horrible idea. How do you implement keyboard shortcuts in links/lynx? How do you do menus and stuff like that? Not to mention the additional security headaches involved in writing web interfaces...
What this person wants to do is get over his leetness and learn to make a GUI application that behaves how he/she thinks a console app should work. There's no reason you can't bind keys the exact same way in a GUI that you would in a console app. And really, a console app created with something like curses is a GUI. Just not a very sophisticated one. A true command line application accepts lines in and spits lines out without mocking up a pathetic GUI using ASCII.