Pardon my ignorance. I thought reference counting was a form of GC, cf "Mark and Sweep".
This is incorrect. Reference counting is quite different to mark and sweep gc. The idea behind mark and sweep garbage collection is you look through the root set - all the pointers allocated on the stack in c parlance - and examine where they point to, and where pointers in dynamically allocated memory point to. Each piece of memory that's pointed to is marked (this can be done with a single bit, since we don't care how many times its pointed to). This is the mark phase. Then we look through the dynamically allocated memory and free up anything that's not pointed to. This is the sweep phase.
(There's a different gc technique called copying gc. I won't describe it here, but it's apparently about as fast as mark + sweep - both blow away reference counting - despite its rather counter-intuitive nature.)
When you're using a language like C or C++ where pointers and integers often look the same, a special variant known as conservative collection (which can be applied to mark + sweep and copying gc) is necessary. The idea here is that if something may point to a piece of allocated memory, don't deallocate it. That means, if we're unlucky, an unrelated field (eg bob.age) might happen to "point" to some allocated memory.
GC is really interesting stuff, and I've not even scratched the surface. For a good survey on GC techniques - fascinating reading - see the gc survey for a much better treatment.
It continues to amaze me over and over, how uninformed people attack Richard Stallman not substantively, but personally - attacking the way he looks, the way he talks, but never substantively refuting what he says. It amazes me even more, how these ad hominem attacks get up-modded. Apparently there is a lot of hatred out there for people of principle.
Apparently there is. It is quite odd.
You defense of RMS is mostly on the ball. I'm glad some people can be honest about him.
You refer to ad hominem attacks against RMS, yet part of your defense of him seems like a "pro hominem" defense. That is, instead of bashing his arguments based on personal attributes, you're defending his arguments based on other personal attributes. I mention this because I've never seen "pro hominem" defenses used. (It's probably a bogus use of Latin, too. My apologies.)
I think one thing RMS could use is an official biography available on his website. So much misinformation is really sad. (There are already some rather funny anecdotes available; I'd love to read more.)
Many small companies (or individuals) can make money by providing tech support for Linux, and many more can save money by using Linux (which in many cases will include developing their own improvements to the system, which under the GPL will then be released to benefit the community)
Incorrect. Changes made to software under the GPL does not necessarily go back to the originating group. You're allowed to keep private changes private.
Most people think this is a good thing. What would happen if all changes to a GPL program needed to be sent back to the originator? Imagine hacking at a program trying to figure out how it works. Chances are good you'd throw in a lot of print statements, maybe change stuff around, possibly rename things or add your own versions of components. All perfectly legitimate things to do, but also useless to anyone else. Should you have to send those changes back to the originator? Would the originator want a patch to his program that amounted to the program printing "Hello world, I am number X" in addition to whatever else it did?
Unlikely.
You're right that lots of useful changes would be given to the originator, but it won't always be the case.
RMS represents pretty much the extreme of trying to prevent people from using what he considers "non-free" software. He routinely uses both license and PR to try to discourage "non-free" software.
It is true that RMS discourages everyone to avoid using free software. It is also true that he sometimes tries to prevent people from creating non-free software. However, it is incorrect to state that he tries to prevent people from using non-free software.
If he were in a position to do so, I'd expect him to work to work hard to prevent or at least strongly discourage "non-free" software from being able to use GNOME, for example by trying to force changes in the licensing.
I don't think that's likely at all. He has made no moves towards such changes with glibc.
Look at the GNU libc LGPL isssue recently for a fine example of him trying to push the libc in that direction.
Either I missed something big, or you misunderstood what happened. glibc has always been available under the LGPL. Changing to a newer version of the same license was a wholly political move. The terms and conditions of the license are the same.
I'm not sure where all of the anti-RMS vitriol comes from. Certainly people who post here sans clue don't help a lot.:( As at least one other poster has suggested (not to you), go have a look at [gnu.org] to get an accurate picture of his views - you won't often see them here on slashdot, I'm afraid.
Just bear in mind that what RMS claims to want most is freedom, and his actions are consistent with that. He doesn't support stuff that limits other people's freedom, but he never denies the right of people to voluntarily do stupid stuff (which is how he regards using non-free software).
Well, I hope that never happens. (Although other posters have claimed that it has. Oh well.)
It's to prevent such problems that the FSF requires people making significant contributions to a GNU Project project to sign their copyrights (for just that specific project, of course) to the FSF. Everyone agrees that it's a pain, but most people (but not all, such as some XEmacs hackers) agree with the reasoning behind it.
It would be a shame if someone tried to pull a stunt like that on the Linux kernel. If it did happen, you can bet that RMS would be jumping up and down and waving his arms around frantically explaining why people running free software projects should collect copyright assignments. As usual, he'd be right.
As I possibly didn't make clear before, I agree with you in general. However,
Tyrrany must be opposed, using any and all tools at our disposal.
is, I'm afraid, an irreconcilable difference between us. I guess it's a case of two wrongs not making a right: I consider the national ID card to be a wrong [1], and I consider making arguments based on emotion a wrong too. I'm aware that many people don't feel the same way.:(
[1] I'm not entirely opposed to a national ID card; provided that it were done properly, I think it could actually be a useful alternative to the SSN system. There would need to be some form of public key crypto used (including private key escro [2]), and strong controls on just who was allowed to access what. But there's no indication that anyone with the means to implement this nonsense has either the desire (Ellison) or ability (Feinstein and Ashcroft) to do it properly.:(
I wouldn't want all private keys escrowed, of course! Just the ones used to implement a national ID system. And then, to get access to the private key, basically, the owner of the ID would have to testify somehow that the forgot his private key... which of course raises other issues.
Re:Why prefer GNU Emacs over XEmacs?
on
GNU Emacs 21
·
· Score: 1
I have the following code run every time Emacs 21 starts:
(defun 21-stuff ()
"Does stuff related to emacs 21"
(blink-cursor-mode -1)
(tool-bar-mode -1))
(if (- emacs-major-version 21)
(21-stuff))
Elsewhere, I have
(menu-bar-mode -1)
(if (equal window-system 'x)
(scroll-bar-mode -1))
The -1 arguments used aren't strictly necessary if you're going to run this code only once per session, but if you might run it more than once, the -1 effectively forces idempotency.
BTW, the primary reason that I use GNU Emacs instead of XEmacs is a small one: in GDB, one can use tab completion to good effect. For example, if I want to set a breakpoint at the function main, I can type break 'mai and GDB will automagically expand it to main. In GUD mode (debugger such as gdb as an inferior process of emacs) in GNU Emacs, I get the same behaviour. In XEmacs' GUD mode, the will cause an expansion to main.c! A minor issue, but one that's annoying enough (I use GDB all the time) to be the deciding factor in my choice of editor.
If Ashcroft and Fienstien both like it, it HAS to be a REALLY bad idea.
As worrying as the idea of a national ID card is, the argument that it's bad because Ashcroft and Feinstein like it holds no water. It's a case of the ad hominem fallacy.
If you're going to argue against national ID cards (a good idea), please do so without undermining the efforts of its opponents who do so without resorting to intellectually dishonest tactics (a bad idea).
(We, as opponents of a national ID card, have the high moral ground here; it would be a pity to squander it.)
No, since Qt is available under the GPL, you CAN use it for corporate, for profit use.
You just can't incorporate it into proprietary products without paying for a different license.
I'm not wild about proprietary software (although I accept it as a fact of life at least for now), but I hardly think paying for a proprietary development license would bother serious proprietary app developers.
I've told a few telemarketers that I don't have a phone.
Some actually sort of argue. "But, uhm. I'm talking to you now!" "Yes, you are, it's a common mistake. I'm not really sure how this happens with my having a phone."
In some ways, it's pretty handy to (now) live in a country that charges for local calls. It apparently makes telemarketing prohibitively expensive; I've received only one telemarketing call (I strongly suspect that the phone company is the one that sold them my details, btw) in the more than a year that I've had my current landline number. If you make few calls - I have made literally only one call from my landline this month - charged local calls aren't very expensive.
Linux a religion? Bah! I say let's have a real religion like Emacs! It's not just an editor, it's an operating system, a philosophy, a way of life! Worship in the Church of Emacs!
(We're a tolerant bunch, and only snicker a little bit when vi users talk crap about Emacs and then start making all sorts of amusing typos and fogetting what mode they're in and hitting the wrong key and deleting whole paragraphs and stuff. (I've seen this happen, btw. No exageration. I did snicker a little bit.))
Yeah, well obviously they're going to make up for it in volume!
In 98 or 99, I remember pointing out to a stockbrokertype uncle of mine that Amazon loses money for every book they sell (I don't know if this is still true, btw) and his response was "they'll make up for it in volume." The unsustainability didn't seem to worry him at all. Haven't had a chance to ask him about it lately, though.
I'm really curious how a regex editor would work. Do you get to make a DFA or NDFA, move the states around, drag arcs from one to another, and specify transition rules? Or maybe is it more like dragging "building blocks" (such as a Kleene star or a + (what's that called?)) around and specifying their order somehow? It just seems rather implausable to me.
On the other hand, I think it would be great if they could put a good interface on it! I could see it being useful for someone who can never remember the specifics of regex syntax depending on what language he's using(like me)and great for people who would otherwise be forced to comb through documents word-by-word to fix - or tag - certain classes of mistakes (like my girlfriend, who will likely be a professional editor in a few years).
Is anyone out there in slashdotland better acquainted with kregexpeditor?
What's more dangerous than a flawed analogy or two?
..Which is a flaming pile of bullshit, logically, because it immediately commits you (and your project) to a life of constantly playing second fiddle..Nobody ever won a marathon by saying "I'm just going to run right behind the leader for the whole race..."
Actually, many races are won by following a naive leader who does the hard work blazing the trail, following in his wake until nearly the end, and surging ahead at the last moment. Of course, I don't think it quite applies to the situation at hand, but that's ok; it is a flawed analogy.
In any case, it's quite unlikely that implementing radical ideas in a direct attempt to improve user interface will have much impact at all. 1: Many people want something similar to what they know. 2: Most "radical ideas" are stupid (3d interfaces and pie-chart style menus are two proposed, very silly, ideas that spring to mind immediately). What ties these two facts together is that incremental improvement is the way to go. Using colour as a visual cue is a good idea. I thought so years ago when I first heard of the idea. Hardly radical, it would probably make a great, yet incremental, improvement.
Whether the people working on GNOME (or KDE) are making the incremental improvements is debatable. I agree with your premise that continually playing catch-up is no good. But the notion of making radical departures from today's accepted norm seems, given GNOME's goals, quite irresponsible. Great research idea, but very risky given more practical goals. --
I recall a statistic someware that only 2% of companies ever really succeed, beyond 3 years. I wonder whatr that percentage is for OSS projects...
It's probably nearly as low. Consider how many dead projects there are at sourceforge. (It's always frustrating to find something that looks promising, only to discover that it's at.002 and has been inactive since early 1999.) Lots of people have great ideas, but there simply isn't enough nosphere (actually quite a good term that) for everyone to build his own text editor or IRC client.
Now, of course, there's nothing wrong with that. I'm not aware of many open source projects that really haven't lived up to their potential - I've never heard anyone say "Gee, it's too bad that AIRC never got off the ground."
Also note that I don't really know how few projects ever make it, but would like to too! --
I sort my garbage, I don't drive unless I have to. I'd like to do it less, but since it's so declasse to use public transportation, no one wants to increase it's availability.
Hmph. You're forgetting the definition of white trash, which is someone who drives a car voluntarily. This is the nineties; you don't need to be white to be white trash anymore!
Seriously, pollution is a serious problem. I'd like to say "it's all your fault" to the fuckwits who continue to do it on a daily basis. Someday, I probably will.
but the right claims that cutting pollution will destroy our economy
They do, and they couldn't be further from the truth. Creating cleaner technologies will improve the economy, because it will put people to work designing, building, and maintaining the new clean equipment. I expect most people who have seriously considered the issue came to that conclusion. But when the President of the US is an oil man, don't expect much in the way of honesty, intellectual or otherwise. We're talking about a President of the US who was elected because he managed to look less intelligent than his opponent. --
Some of us want it both ways: we want a relatively independent source of news provided by relatively clueful people, and we want it to be well written.
Slashdot is a good source of relatively independent and informed (they could do better in this department, but as other media outlets show, they could also do much, much worse) news. On the other hand, sometimes the grammar and spelling is so bad that the writing verges on the unintelligible.
Professionalism and integrity are not mutually exclusive; really, it's plain sad to see Malda say he doesn't care about the professionalism side of it, because, if nothing else, it means that slashdot will never reach its full potential. --
Re:Don't mean to sound the troll here but...
on
SJGames Layoffs
·
· Score: 1
And especially, for me anyway....
Oops, make that ``...especially apropos...''. There is no such thing as ``especially true''.:) --
Re:Don't mean to sound the troll here but...
on
SJGames Layoffs
·
· Score: 1
Deception starts with the individual, and feeds on itself. Note the drug addict.
Too true. (And especially, for me anyway, since I've been dealing with someone who deludes herself regarding some of her, err, habits.)
However, the guy I was trying to convince knew that Windows was crap and crashed and caused him all sorts of grief. He was/is an structural engineer for a small company. Either by choice or by fiat (I think it was the former, but can't remember) he became his firm's ``computer guy'' - he knew first hand just how irritating Windows' fallability was.
One excuse he used was that clients were billed for time working, even if that time working had to be done more than once due to computer unreliability, but he was only rationalising by that point, trying to get something good out of the situation.
Still leaves me puzzled that he would prefer to be lied to. --
True, none of them seems particularly knowledgable. However, Natasha Stott Despoja (who, interestingly, seems to be as despised by the hard left as she is by the right) did submit a fairly insightful dissenting opinion in response to the net censorship bill. I wonder how much of it was actually written by her.
She seems to be considerably better informed than the current minister for telecommunications and IT, Richard Alston. I hesitate to use such strong language, but the best way to describe Alston is to call him a menace. (To be fair, I expect that term can be used for most people with a cabinet position, but I'm not sufficiently qualified in their areas to describe them as such.) Anyone with the ability to sound authoritative about something he knows virtually nothing about is a dangerous person indeed. Doubly so when that person has authority in that area.
I read recently that the Labour party was considering coming out in favour of open source. However, the statement I read seemed to be entirely content-free, or at least entirely non-commital, certainly not a sufficient reason to vote for Labour.
It certainly is a discouraging situation, all the more so because I haven't a clue what can realistically be done about it. --
Slashdot Mirror
What was funny was wneh Kevin McHale clotheslined him. Hehehe.
This is incorrect. Reference counting is quite different to mark and sweep gc. The idea behind mark and sweep garbage collection is you look through the root set - all the pointers allocated on the stack in c parlance - and examine where they point to, and where pointers in dynamically allocated memory point to. Each piece of memory that's pointed to is marked (this can be done with a single bit, since we don't care how many times its pointed to). This is the mark phase. Then we look through the dynamically allocated memory and free up anything that's not pointed to. This is the sweep phase.
(There's a different gc technique called copying gc. I won't describe it here, but it's apparently about as fast as mark + sweep - both blow away reference counting - despite its rather counter-intuitive nature.)
When you're using a language like C or C++ where pointers and integers often look the same, a special variant known as conservative collection (which can be applied to mark + sweep and copying gc) is necessary. The idea here is that if something may point to a piece of allocated memory, don't deallocate it. That means, if we're unlucky, an unrelated field (eg bob.age) might happen to "point" to some allocated memory.
GC is really interesting stuff, and I've not even scratched the surface. For a good survey on GC techniques - fascinating reading - see the gc survey for a much better treatment.
Apparently there is. It is quite odd.
You defense of RMS is mostly on the ball. I'm glad some people can be honest about him.
You refer to ad hominem attacks against RMS, yet part of your defense of him seems like a "pro hominem" defense. That is, instead of bashing his arguments based on personal attributes, you're defending his arguments based on other personal attributes. I mention this because I've never seen "pro hominem" defenses used. (It's probably a bogus use of Latin, too. My apologies.)
I think one thing RMS could use is an official biography available on his website. So much misinformation is really sad. (There are already some rather funny anecdotes available; I'd love to read more.)
New code. Coolness. Now where's that remove-double-posts feature? =)
Incorrect. Changes made to software under the GPL does not necessarily go back to the originating group. You're allowed to keep private changes private.
Most people think this is a good thing. What would happen if all changes to a GPL program needed to be sent back to the originator? Imagine hacking at a program trying to figure out how it works. Chances are good you'd throw in a lot of print statements, maybe change stuff around, possibly rename things or add your own versions of components. All perfectly legitimate things to do, but also useless to anyone else. Should you have to send those changes back to the originator? Would the originator want a patch to his program that amounted to the program printing "Hello world, I am number X" in addition to whatever else it did? Unlikely.
You're right that lots of useful changes would be given to the originator, but it won't always be the case.
It is true that RMS discourages everyone to avoid using free software. It is also true that he sometimes tries to prevent people from creating non-free software. However, it is incorrect to state that he tries to prevent people from using non-free software.
If he were in a position to do so, I'd expect him to work to work hard to prevent or at least strongly discourage "non-free" software from being able to use GNOME, for example by trying to force changes in the licensing.
I don't think that's likely at all. He has made no moves towards such changes with glibc.
Look at the GNU libc LGPL isssue recently for a fine example of him trying to push the libc in that direction.
Either I missed something big, or you misunderstood what happened. glibc has always been available under the LGPL. Changing to a newer version of the same license was a wholly political move. The terms and conditions of the license are the same.
I'm not sure where all of the anti-RMS vitriol comes from. Certainly people who post here sans clue don't help a lot. :( As at least one other poster has suggested (not to you), go have a look at [gnu.org] to get an accurate picture of his views - you won't often see them here on slashdot, I'm afraid.
Just bear in mind that what RMS claims to want most is freedom, and his actions are consistent with that. He doesn't support stuff that limits other people's freedom, but he never denies the right of people to voluntarily do stupid stuff (which is how he regards using non-free software).
It's to prevent such problems that the FSF requires people making significant contributions to a GNU Project project to sign their copyrights (for just that specific project, of course) to the FSF. Everyone agrees that it's a pain, but most people (but not all, such as some XEmacs hackers) agree with the reasoning behind it.
It would be a shame if someone tried to pull a stunt like that on the Linux kernel. If it did happen, you can bet that RMS would be jumping up and down and waving his arms around frantically explaining why people running free software projects should collect copyright assignments. As usual, he'd be right.
Tyrrany must be opposed, using any and all tools at our disposal.
is, I'm afraid, an irreconcilable difference between us. I guess it's a case of two wrongs not making a right: I consider the national ID card to be a wrong [1], and I consider making arguments based on emotion a wrong too. I'm aware that many people don't feel the same way. :(
[1] I'm not entirely opposed to a national ID card; provided that it were done properly, I think it could actually be a useful alternative to the SSN system. There would need to be some form of public key crypto used (including private key escro [2]), and strong controls on just who was allowed to access what. But there's no indication that anyone with the means to implement this nonsense has either the desire (Ellison) or ability (Feinstein and Ashcroft) to do it properly. :(
I wouldn't want all private keys escrowed, of course! Just the ones used to implement a national ID system. And then, to get access to the private key, basically, the owner of the ID would have to testify somehow that the forgot his private key... which of course raises other issues.
(defun 21-stuff ()
"Does stuff related to emacs 21"
(blink-cursor-mode -1)
(tool-bar-mode -1))
(if (- emacs-major-version 21)
(21-stuff))
Elsewhere, I have
(menu-bar-mode -1)
(if (equal window-system 'x)
(scroll-bar-mode -1))
The -1 arguments used aren't strictly necessary if you're going to run this code only once per session, but if you might run it more than once, the -1 effectively forces idempotency.
BTW, the primary reason that I use GNU Emacs instead of XEmacs is a small one: in GDB, one can use tab completion to good effect. For example, if I want to set a breakpoint at the function main, I can type break 'mai and GDB will automagically expand it to main. In GUD mode (debugger such as gdb as an inferior process of emacs) in GNU Emacs, I get the same behaviour. In XEmacs' GUD mode, the will cause an expansion to main.c! A minor issue, but one that's annoying enough (I use GDB all the time) to be the deciding factor in my choice of editor.
As worrying as the idea of a national ID card is, the argument that it's bad because Ashcroft and Feinstein like it holds no water. It's a case of the ad hominem fallacy.
If you're going to argue against national ID cards (a good idea), please do so without undermining the efforts of its opponents who do so without resorting to intellectually dishonest tactics (a bad idea).
(We, as opponents of a national ID card, have the high moral ground here; it would be a pity to squander it.)
No, since Qt is available under the GPL, you CAN use it for corporate, for profit use.
You just can't incorporate it into proprietary products without paying for a different license.
I'm not wild about proprietary software (although I accept it as a fact of life at least for now), but I hardly think paying for a proprietary development license would bother serious proprietary app developers.
Some actually sort of argue. "But, uhm. I'm talking to you now!" "Yes, you are, it's a common mistake. I'm not really sure how this happens with my having a phone."
In some ways, it's pretty handy to (now) live in a country that charges for local calls. It apparently makes telemarketing prohibitively expensive; I've received only one telemarketing call (I strongly suspect that the phone company is the one that sold them my details, btw) in the more than a year that I've had my current landline number. If you make few calls - I have made literally only one call from my landline this month - charged local calls aren't very expensive.
(We're a tolerant bunch, and only snicker a little bit when vi users talk crap about Emacs and then start making all sorts of amusing typos and fogetting what mode they're in and hitting the wrong key and deleting whole paragraphs and stuff. (I've seen this happen, btw. No exageration. I did snicker a little bit.))
In 98 or 99, I remember pointing out to a stockbrokertype uncle of mine that Amazon loses money for every book they sell (I don't know if this is still true, btw) and his response was "they'll make up for it in volume." The unsustainability didn't seem to worry him at all. Haven't had a chance to ask him about it lately, though.
Actually, Charles will take to the throne as George. It's one of his four names. The previous Charleses apparently weren't such great kings.
Doh! So much for my glorious uptime! :(
Yeah, right, like Jon Katz has ever heard of Donald Knuth.. :)
On the other hand, I think it would be great if they could put a good interface on it! I could see it being useful for someone who can never remember the specifics of regex syntax depending on what language he's using(like me)and great for people who would otherwise be forced to comb through documents word-by-word to fix - or tag - certain classes of mistakes (like my girlfriend, who will likely be a professional editor in a few years).
Is anyone out there in slashdotland better acquainted with kregexpeditor?
Actually, many races are won by following a naive leader who does the hard work blazing the trail, following in his wake until nearly the end, and surging ahead at the last moment. Of course, I don't think it quite applies to the situation at hand, but that's ok; it is a flawed analogy.
In any case, it's quite unlikely that implementing radical ideas in a direct attempt to improve user interface will have much impact at all. 1: Many people want something similar to what they know. 2: Most "radical ideas" are stupid (3d interfaces and pie-chart style menus are two proposed, very silly, ideas that spring to mind immediately). What ties these two facts together is that incremental improvement is the way to go. Using colour as a visual cue is a good idea. I thought so years ago when I first heard of the idea. Hardly radical, it would probably make a great, yet incremental, improvement.
Whether the people working on GNOME (or KDE) are making the incremental improvements is debatable. I agree with your premise that continually playing catch-up is no good. But the notion of making radical departures from today's accepted norm seems, given GNOME's goals, quite irresponsible. Great research idea, but very risky given more practical goals.
--
It's probably nearly as low. Consider how many dead projects there are at sourceforge. (It's always frustrating to find something that looks promising, only to discover that it's at .002 and has been inactive since early 1999.) Lots of people have great ideas, but there simply isn't enough nosphere (actually quite a good term that) for everyone to build his own text editor or IRC client.
Now, of course, there's nothing wrong with that. I'm not aware of many open source projects that really haven't lived up to their potential - I've never heard anyone say "Gee, it's too bad that AIRC never got off the ground."
Also note that I don't really know how few projects ever make it, but would like to too!
--
Hmph. You're forgetting the definition of white trash, which is someone who drives a car voluntarily. This is the nineties; you don't need to be white to be white trash anymore!
Seriously, pollution is a serious problem. I'd like to say "it's all your fault" to the fuckwits who continue to do it on a daily basis. Someday, I probably will.
but the right claims that cutting pollution will destroy our economy
They do, and they couldn't be further from the truth. Creating cleaner technologies will improve the economy, because it will put people to work designing, building, and maintaining the new clean equipment. I expect most people who have seriously considered the issue came to that conclusion. But when the President of the US is an oil man, don't expect much in the way of honesty, intellectual or otherwise. We're talking about a President of the US who was elected because he managed to look less intelligent than his opponent.
--
Slashdot is a good source of relatively independent and informed (they could do better in this department, but as other media outlets show, they could also do much, much worse) news. On the other hand, sometimes the grammar and spelling is so bad that the writing verges on the unintelligible.
Professionalism and integrity are not mutually exclusive; really, it's plain sad to see Malda say he doesn't care about the professionalism side of it, because, if nothing else, it means that slashdot will never reach its full potential.
--
Oops, make that ``...especially apropos...''. There is no such thing as ``especially true''. :)
--
Too true. (And especially, for me anyway, since I've been dealing with someone who deludes herself regarding some of her, err, habits.)
However, the guy I was trying to convince knew that Windows was crap and crashed and caused him all sorts of grief. He was/is an structural engineer for a small company. Either by choice or by fiat (I think it was the former, but can't remember) he became his firm's ``computer guy'' - he knew first hand just how irritating Windows' fallability was.
One excuse he used was that clients were billed for time working, even if that time working had to be done more than once due to computer unreliability, but he was only rationalising by that point, trying to get something good out of the situation.
Still leaves me puzzled that he would prefer to be lied to.
--
She seems to be considerably better informed than the current minister for telecommunications and IT, Richard Alston. I hesitate to use such strong language, but the best way to describe Alston is to call him a menace. (To be fair, I expect that term can be used for most people with a cabinet position, but I'm not sufficiently qualified in their areas to describe them as such.) Anyone with the ability to sound authoritative about something he knows virtually nothing about is a dangerous person indeed. Doubly so when that person has authority in that area.
I read recently that the Labour party was considering coming out in favour of open source. However, the statement I read seemed to be entirely content-free, or at least entirely non-commital, certainly not a sufficient reason to vote for Labour.
It certainly is a discouraging situation, all the more so because I haven't a clue what can realistically be done about it.
--