would argue that no one owns linux, well, not all of it anyway. The individual bits of code are owned by their authors, no single entity owns the whole thing.
Like I said, I think you have a valid point.
I just don't see how the distinction is likely impinge on the user, unless of course the user
plans to modify the source code and make a binary only release.
mmm... I was thinking of "illegal monopoly" more as shorthand for "a company found to employ tactics that are illegal in light of it's status as a monopoly", rather than tautologically for emphasis.
Sloppy wording, all the same. I'll have to try and find something better.
This is not targeted specifically at you, NickFortune
Um... the statistics from the U.S. Department of Justice indicating that
youth crime has been declining since 1993, as cited by Shky about four posts earlier in the thread. Was this in some way unclear?
And which version of the Linux kernel are you running? How productive and secure do you think you would be if you were still using version 1.2 (1995) or 2.1 (1998)?
mmm.. call me thick if you must, but what's your point?
And if I thought that your personal serenity
was the key to securing all those windows boxes out there,
I would fall silent henceforth. Alas, I
fear less whimsical solutions are required. Discussion seems appropriate.
The software world in general (notwithstanding governments) really
never cared too much for firewalls. In fact, only websites did. Yet now,
we have everyone under the sun bitching about how insecure Windows is
because of this or that.
It's jolly nice of you to speak up on behalf of all the Microsoft devs
like this. I'm sure that they are all nice guys, and I'll even go so
far as to concede that it's a little unfair to blame them personally
for the majority of the weaknesses in windows security model.
However, that does little to alter the fact that the most widely deployed
computer platform on Planet Earth leaks like a bloody sieve.
It's not just firewalls. It's a security model and worse, a corporate
culture, that places a higher
premium on ease-of-use than it does on actual security.
Incidentally, the "it wasn't designed with security in mind" defence was true (if unhelpful) for Win95 through to ME.
But NT is supposed to be a rewrite from
the ground up, and XP is NT derived, is it not?
So there really isn't any excuse for not getting it right.
Newsflash, they weren't thinking of security at the time, and I'm sure
you weren't either.
That's not news, not relevant and also not accurate.
There's more to security than firewalls. Having a security model so
that the first item of malware to come along doesn't infect the
entire system is also important. XP has this capability, but it's a
recent development in the windows world.
Unix on the other hand, has had a security model for ages. I wrote my first Unix shell script in 1980. Back then, trying to figure out how to hack the root password was the major undergrad sport in the computing department. So Unix devs were thinking of security back then. And so, as it happens, was I, albeit from more of a cracker's perspective than that of a dev.
But the important thing here is that a history of security problems have to be taken into account when you consider the drawbacks of a windows box. It's not a purely historic issue - it's a problem rooted in (IMHO) MS corporate culture, whereby ease of use takes precedence over security, and exacerbated by MS' history of responding to security alerts with denial and disinformation and only with patches as a last resort. Happily, this attitude seems to be on the wane at Redmond,
but corporate cultures run deep and and it's too soon to say if the leopard has changed this particular set of spots.
I suppose it depends how highly you value things like security patches, USB drivers, versions of MS word that can read contemporary documents. Or even an OS that can use current scanners and printers.
Still, you can always do that with an up to date desktop installation, right?
Of course they can't make you upgrade. But they certainly can make life inconvenient for the majority of those who decline.
It was the fourfold duplication of the data which bothered me.
I can see a combination of development methodology and IDE having this effect. Especially when you combine it with deadline pressures and the trade off between development time and tight code.
That said, any development methodolgy that wastes my computer resources like that, I'd have to call broken. And I never heard about anyone at Redmond getting carpetted over that flight sim easter egg...
The upgrade thing is annoying, but you can turn it off.
I was thinking more of the business where MS withdraw support for an version of windows and then if you want security fixes, updated drivers and what have you, then you have to buy a new version of windows. And typically a new computer to handle the demands of the new OS.
I wouldn't mind if it wasn't so artificial.
I guess OS really comes down to Prefrence.
A sane and sensible attitude to take. What you doing on Slashdot?;)
I find this "software bloat to drive hardware upgrades" thing hard to believe.
MS and Intel were hand in glove for years. If the romaince is beginning to fade, they still scratched one another's backs for years. Intel made sure MS got an architecture that favoured Windows and MS released software that required new hardware to make it run.
I don't see what strains the credulity about this scenario.
Of course, maybe it's the bloat and not the relationship to which you object.
I remember reading about an MS command line utility that had four non-functional windows menus embedded in it. And then of course there was flight sim embedded in Exel. I'd call that bloat.
Doesn't that mean that similar applications (on Linux for example) to most Microsoft created applications should be significantly faster?
It means that they could be significantly smaller, and that the system could be faster overall. I have a perfectly functional modern Linux laptop running on a P166 with 2 gig of disk space. It uses FVWM for the desktop and AbiWord for WP, but it does everything I'd want from a windows box except play Far Cry.
The laptop originally ran Win95 and did it quite well. It runs Linux (2.6 kernel, latest software) just as well. I don't think it has enough disk space for XP, let alone processor power.
Last I checked, OpenOffice.org was generally the same speed and level of "bloatness" as Microsoft Word.
No argument there. MS have encouraged a "throw hardware at it" philosophy of software development, and a lot of projects now cheerfully sacrifice speed and size for short term delvelopmental convenience. It's not a philosophy of which I approve, particularly, although I'll grant the approach can be useful in certain circumstances.
But it's not the only way, and there should be alternatives. Linux has those alternatives, Windows (unless you can get a CE install for your old x86 hardware) does not.
Arguably, everyone owns Linux. There are a few rights reserved, mainly restricting my rights to restrict those of others in respect to Linux, and specifically in respect to source code if I choose to distribute a modified version of the software. But it seems clear that the intent is that Linux/Gnu should be a Commonewealth. The point is, I suppose, debatable.
Beyond question however, is that I have vastly more rights regarding my Linux software than I have over any windows system I've ever used or that I ever expect to use.
The only problem with windows, is that it is designed for the Average User.
And that it has this pesky forced upgrade cycle thing. Oh, software bloat to drive hardware upgrades. And serious secuity issues stemming from an unwise level of integration. Which is an instance of a larger problem: fundamental technical decisions being made by marketing staff rather than techies. And then there are the licencing terms which mean that you don't own your OS, and which reserve for MS the right to monkey around with your system at any future time. And there's proprietory file formats, and vendor lock in.
And some of us have concerns about how spyware-friendly windows appears to be, especially in the light of the Claria/Gator aquisition.
But apart from that, you're dead right. The only problem with windows is that it's designed for the Average User. Oh, and a few of us dislike windows because it supports an illegal monopoly which has a well documented history of unethical and anti-competitive business strategies,
But if you link against them, why not just link against a file in/usr/lib?
There are advantages both ways. If you keep everything in whatever/lib then
you have everything (more or less) centralised so you know where to look for it.
On the other hand, individual/opt dirs allow applications to install without (necessarily) needing root, they allow apps to create the same */lib filenames without overwriting each other.
The other solution to the overwriting problem is subdirs under path/lib, but if you're going to do that, you might as well put them in/opt.
Interestingly,/usr/lib and/usr/local/lib both started life as non-system locations for user resources and have become co-opted by the system (/usr/lib more so) as their use was formalised. The/opt system just takes that idea to its logical extreme.
I should also say that I don't personally like/opt. Partly that's because I remember commercial systems where/opt package directories were obfuscated as far as possible - a right pain when you need the release notes for one of fifty packages and upgrades lurking there.
But that doesn't make it a bad way to do things - just different
As others I'm sure have allready mentioned, the personal and covertly threatening attack on Pamela Jones privacy is completely and forever unforgivable.
True enough. Didio went well beyone "credulous" and into personal and vindicitve.
Wow, for those of you with short memories, at one point Apple sued Microsoft essentially claiming they OWNED the windowed desktop metaphor. Suit was dismissed.
Some of us who aren't not trolling as ACs may occasionally reserve the right to disapprove of both of these cases. I appreciate that this makes for very poor flamebait.
Apple butt-kissers can be extremely hypocritical. Their precious company can do no wrong, no matter what, while everyone else is EVIL!!!
I wouldn't know about that, personally, never having used a mac. I do oppose software patents however. Hope that helps.
Now: back under your bridge before the Billy Goats Gruff turn up.
I can't see adding $10 to the price of an iPod giving Microsoft much of a competitive advantage - unless their product is superior, which it probably won't be.
You're probably right. Well, unless MS produce one of their marketing coups - about the one thing MS consistently get right.
But the important thing here is whether MS have done anything to deserve even the option of taking a slice of the revenuse from the iPod. They didn't develop anything.
And for that matter, should they ever have this power?
well in proving that i am thick, and in doing so hopefully becoming not quite so thick.. i was imagining x being the number of hits and y being the number of queries returning this number of hits...
That seems reasonable to me. I think I'd still expect the amplitude of the curve to be proportionate to the number of pages indexed, though.
Agreed, you can question the claims that yahoo has double the indexing of google. If you assume that the returned result set returned from any two keywords is always linear.
At the risk of seeming thick, linear with respect to what? What are the axes on this curve?
I expect the Y-Axis is going to be number of hits. What's the X-Axis? number of pages indexed? I just can't visualise it...
In any case,
I'll grant that one question raised is most certainly "how useful is index size as a metric for search engine usefulness?"
My point wasn't that I think Yahoo is better than, or even as good as, Google -- just that I don't find the experimental results from this study to be meaningful in the debate.
Fair enough. I'll cheerfully conceed that this says nothing about which is the better engine. I just have a hard time squaring the results with Yahoo's claims re: index size.
It's all a bit beside the point really. My choice of search engine owes nothing to reported index size. I just don't like being lied to, and there seems a real possibility that Yahoo lied about those indices.
Slashdot Mirror
Like I said, I think you have a valid point.
I just don't see how the distinction is likely impinge on the user, unless of course the user plans to modify the source code and make a binary only release.
Sloppy wording, all the same. I'll have to try and find something better.
This is not targeted specifically at you, NickFortune
Taken in the spirit intended :)
Um... the statistics from the U.S. Department of Justice indicating that youth crime has been declining since 1993, as cited by Shky about four posts earlier in the thread. Was this in some way unclear?
Which "body of evidence" were you referring to?
mmm.. call me thick if you must, but what's your point?
Frankly they're annoying.
And if I thought that your personal serenity was the key to securing all those windows boxes out there, I would fall silent henceforth. Alas, I fear less whimsical solutions are required. Discussion seems appropriate.
The software world in general (notwithstanding governments) really never cared too much for firewalls. In fact, only websites did. Yet now, we have everyone under the sun bitching about how insecure Windows is because of this or that.
It's jolly nice of you to speak up on behalf of all the Microsoft devs like this. I'm sure that they are all nice guys, and I'll even go so far as to concede that it's a little unfair to blame them personally for the majority of the weaknesses in windows security model.
However, that does little to alter the fact that the most widely deployed computer platform on Planet Earth leaks like a bloody sieve. It's not just firewalls. It's a security model and worse, a corporate culture, that places a higher premium on ease-of-use than it does on actual security.
Incidentally, the "it wasn't designed with security in mind" defence was true (if unhelpful) for Win95 through to ME. But NT is supposed to be a rewrite from the ground up, and XP is NT derived, is it not? So there really isn't any excuse for not getting it right.
Newsflash, they weren't thinking of security at the time, and I'm sure you weren't either.
That's not news, not relevant and also not accurate.
There's more to security than firewalls. Having a security model so that the first item of malware to come along doesn't infect the entire system is also important. XP has this capability, but it's a recent development in the windows world.
Unix on the other hand, has had a security model for ages. I wrote my first Unix shell script in 1980. Back then, trying to figure out how to hack the root password was the major undergrad sport in the computing department. So Unix devs were thinking of security back then. And so, as it happens, was I, albeit from more of a cracker's perspective than that of a dev.
But the important thing here is that a history of security problems have to be taken into account when you consider the drawbacks of a windows box. It's not a purely historic issue - it's a problem rooted in (IMHO) MS corporate culture, whereby ease of use takes precedence over security, and exacerbated by MS' history of responding to security alerts with denial and disinformation and only with patches as a last resort. Happily, this attitude seems to be on the wane at Redmond, but corporate cultures run deep and and it's too soon to say if the leopard has changed this particular set of spots.
Still, you can always do that with an up to date desktop installation, right?
Of course they can't make you upgrade. But they certainly can make life inconvenient for the majority of those who decline.
I have to confess, I only read the first page before the slashdot effect clobbered the site. But based upon what I read, I can't say I'm surprised.
I can see a combination of development methodology and IDE having this effect. Especially when you combine it with deadline pressures and the trade off between development time and tight code.
That said, any development methodolgy that wastes my computer resources like that, I'd have to call broken. And I never heard about anyone at Redmond getting carpetted over that flight sim easter egg...
I was thinking more of the business where MS withdraw support for an version of windows and then if you want security fixes, updated drivers and what have you, then you have to buy a new version of windows. And typically a new computer to handle the demands of the new OS.
I wouldn't mind if it wasn't so artificial.
I guess OS really comes down to Prefrence.
A sane and sensible attitude to take. What you doing on Slashdot? ;)
FVWM and Vi. Pleased to meet you, hope you guessed my name :D
And yes, OOo for MS compatibility issues. Although AbiWord does pretty much as well and is a lot trimmer than OOo writer.
MS and Intel were hand in glove for years. If the romaince is beginning to fade, they still scratched one another's backs for years. Intel made sure MS got an architecture that favoured Windows and MS released software that required new hardware to make it run.
I don't see what strains the credulity about this scenario.
Of course, maybe it's the bloat and not the relationship to which you object. I remember reading about an MS command line utility that had four non-functional windows menus embedded in it. And then of course there was flight sim embedded in Exel. I'd call that bloat.
Doesn't that mean that similar applications (on Linux for example) to most Microsoft created applications should be significantly faster?
It means that they could be significantly smaller, and that the system could be faster overall. I have a perfectly functional modern Linux laptop running on a P166 with 2 gig of disk space. It uses FVWM for the desktop and AbiWord for WP, but it does everything I'd want from a windows box except play Far Cry.
The laptop originally ran Win95 and did it quite well. It runs Linux (2.6 kernel, latest software) just as well. I don't think it has enough disk space for XP, let alone processor power.
Last I checked, OpenOffice.org was generally the same speed and level of "bloatness" as Microsoft Word.
No argument there. MS have encouraged a "throw hardware at it" philosophy of software development, and a lot of projects now cheerfully sacrifice speed and size for short term delvelopmental convenience. It's not a philosophy of which I approve, particularly, although I'll grant the approach can be useful in certain circumstances.
But it's not the only way, and there should be alternatives. Linux has those alternatives, Windows (unless you can get a CE install for your old x86 hardware) does not.
Arguably, everyone owns Linux. There are a few rights reserved, mainly restricting my rights to restrict those of others in respect to Linux, and specifically in respect to source code if I choose to distribute a modified version of the software. But it seems clear that the intent is that Linux/Gnu should be a Commonewealth. The point is, I suppose, debatable.
Beyond question however, is that I have vastly more rights regarding my Linux software than I have over any windows system I've ever used or that I ever expect to use.
That's close enough for me.
I stand corrected :D
And that it has this pesky forced upgrade cycle thing. Oh, software bloat to drive hardware upgrades. And serious secuity issues stemming from an unwise level of integration. Which is an instance of a larger problem: fundamental technical decisions being made by marketing staff rather than techies. And then there are the licencing terms which mean that you don't own your OS, and which reserve for MS the right to monkey around with your system at any future time. And there's proprietory file formats, and vendor lock in. And some of us have concerns about how spyware-friendly windows appears to be, especially in the light of the Claria/Gator aquisition.
But apart from that, you're dead right. The only problem with windows is that it's designed for the Average User. Oh, and a few of us dislike windows because it supports an illegal monopoly which has a well documented history of unethical and anti-competitive business strategies,
But apart from all that...
Indeed.
Similarly, there is no proof for gravity - although the statistical correlation is somewhat overwhelming.
This case lacks the same statistical depth, but the facts are still against games-cause-violence theory.
There are advantages both ways. If you keep everything in whatever/lib then you have everything (more or less) centralised so you know where to look for it.
On the other hand, individual /opt dirs allow applications to install without (necessarily) needing root, they allow apps to create the same */lib filenames without overwriting each other.
The other solution to the overwriting problem is subdirs under path/lib, but if you're going to do that, you might as well put them in /opt.
Interestingly, /usr/lib and /usr/local/lib both started life as non-system locations for user resources and have become co-opted by the system (/usr/lib more so) as their use was formalised. The /opt system just takes that idea to its logical extreme.
I should also say that I don't personally like /opt. Partly that's because I remember commercial systems where /opt package directories were obfuscated as far as possible - a right pain when you need the release notes for one of fifty packages and upgrades lurking there.
But that doesn't make it a bad way to do things - just different
Nope. Although you do need to know where they are to link against them/load them.
Does it also install its executables into its own subdirectory?
Wouldn't matter as long as the user has read access. Just add it to your path.
Conceeded. Like you say, they had a lot in common, but I'll grant that DiDio never stooped so low as that.
True enough. Didio went well beyone "credulous" and into personal and vindicitve.
I notice the article was submitted by "an anonymous reader".
Maybe the person who wrote this has been writing her previous work.
Some of us who aren't not trolling as ACs may occasionally reserve the right to disapprove of both of these cases. I appreciate that this makes for very poor flamebait.
Apple butt-kissers can be extremely hypocritical. Their precious company can do no wrong, no matter what, while everyone else is EVIL!!!
I wouldn't know about that, personally, never having used a mac. I do oppose software patents however. Hope that helps.
Now: back under your bridge before the Billy Goats Gruff turn up.
You're probably right. Well, unless MS produce one of their marketing coups - about the one thing MS consistently get right.
But the important thing here is whether MS have done anything to deserve even the option of taking a slice of the revenuse from the iPod. They didn't develop anything.
And for that matter, should they ever have this power?
That seems reasonable to me. I think I'd still expect the amplitude of the curve to be proportionate to the number of pages indexed, though.
At the risk of seeming thick, linear with respect to what? What are the axes on this curve?
I expect the Y-Axis is going to be number of hits. What's the X-Axis? number of pages indexed? I just can't visualise it...
In any case, I'll grant that one question raised is most certainly "how useful is index size as a metric for search engine usefulness?"
Fair enough. I'll cheerfully conceed that this says nothing about which is the better engine. I just have a hard time squaring the results with Yahoo's claims re: index size.
It's all a bit beside the point really. My choice of search engine owes nothing to reported index size. I just don't like being lied to, and there seems a real possibility that Yahoo lied about those indices.