Frankly, you'd better take a little more thought in your answer..
MS: License cost is only a small part of the TOTAL COST OF OWNERSHIP.
Your answer: License cost is a *significant* part of the cost of a PC.
You're answer is 'a cote': the cost of a PC itself is only a small part of the total cost of a PC which include support, training, etc..
Also: ARGUMENT: MS-Office documents may not open properly in Open Office and visa-versa. FACT: Isn't this Microsoft's fault? After all, they are the ones that keep changing their applications to make interoperability more and more difficult with each release.
Your answer is a bit 'childish', you're right of course, it is MS fault but who cares?? The fact is that in any company, there is a huge number of existing DOC files and opening those DOC file in OpenOffice doesn't work very well: often the formatting is garbled, and this is not FUD: I'm using OpenOffice at work, so I know it quite well..
The truth is: OpenOffice adoption could only be fast if it had *near perfect* compatibility with Office in reading AND writing (PDF doesn't do it if you want to share a writeable document), which it is a difficult target.
We're currently buying PC with Linux at work, and you know what? We will spend additional money to allow developpers to have an access to MS Office on their Linux's PC, because OpenOffice is not good enough in an 'MS world'.. It would be good enough in an 'OpenOffice world': sharing only documents with other OpenOffice users, but this is not the current real situation..
>> 5. "Open source benefits from anti-American sentiments." >Personally I'm big into this! I feel that the potential for Europe to regain power and all of that is pretty massive. >However, outside of the USA there is one big problem - language. You may think that this is getting better - go to Brussels, goto Barcelona and >see how many 20-30 year olds speak English; not many.
Uh? We're talking about software developers here! Most of the software developers I know speak English, you simply have to, if you want to be a good software developer.. And I'm from France: a country not especially well-known for the people abilities to speak foreign language.
> All the source packages and RPM's we get come with MD5 sums.
And? This doesn't ensure anything. MD5 sums only assure you that the RPM has transferred correctly, it don't tell you who has created the RPM, it could be an hacker.. Until we use some certificate to authenticate the provenance of a package, MD5 won't protect you if the main distribution box has been hacked..
>but the inet connection of the user isn't disrupted at all.
Depends on the user, personnaly the official bittorrent client for Windows is the only thing which makes me loose my internet connection..
It is not a big problem as: 1) I'm not using bittorrent very often. 2) I have a 'daemon' reconnecting the connection in case of a loss (needed it when the driver wasn't very good).
> Considering the change to the scheduler to make everything seem snappier,
This is not obvious: I thought that Mandrake included already the preempt + low latency patch in 2.4 ? In this case, users won't notice much difference..
>Wouldn't you be a little bit more qualified to comment if you had several years of experience with both functional and imperative languages, first?
You're right of course, but it is notheless true, that personnaly I found learning functionnal language quite difficult instead of imperative language, and I'm not the only one.
Plus in the commercial world of today, you're unlikely to have several years of experience with functionnal language..
Academics can have this experience of course, but for some reason, I tend not to value very much academics point of view in this matter: they tend to do different things that in the "real world", such as IHM, database programming, scripts, etc.. That's why we use C and not Pascal today..
I'm not the one having made the original comment but, using exceptions for break or return? Bleach, this is ugly!
Break and return have a purpose (changing the flow of execution), exceptions a different one (reporting error), mixing both is ugly.
What the point of using a "clean language" if you need to hack around the language?
Me I tried to learn Ocaml and didn't manage to do it.. Now I'm learning Ruby which is much more easy for my poor brain, and it is also the most readable language I know (unfortunately it is slow also).
While your rant started OK (I'm often annoyed by poor doc either in the BIOS or in Windows: the help files treats you like an idiot and don't give you any useful informations), it end up wrong, yes to use Unix you have to know that you get the help with the man command, but I don't think that this is a problem, just a small prerequisite.
There are other many thing wrong in help files on Unix: the info tool is pure hell to use if you're not an emacs user, man is quite primitive (no hyperlink), quite often GUI are interface which call a command-line equivalent in some GUI the name of the command-line is not written..
Sorry but what matters in the end is that the software works, so this means: a) a language which ease programmers life. b) a language which simplifies testing, integrates tests.
If you look only at a), you're missing the whole picture..
A language designer should look at both!
It is no surprise that she works at Sun: I was very surprised that Java doesn't include any 'design by contract' aspect like Eiffel, and when I tried to use it, I undestood: sloppy libraries where Sun had everytime a new library to make the coffee but doesn't take the time to fix the numerous bugs in the existing libraries for years, lots of workaround to use..
I think that some people forgets that even with the most intuitive language in the world, you need tests and the language should support it!
I'm wondering if it depends of the "average density"?
Even if all balls are made from the same materials, you could say that on average large balls are less dense than the small one because when there is more free space between the balls..
I'm curious what would happens if the density of the large balls was increased to compensate for the difference in "average density": what added density should be added so that the large balls sinks or don't move..
>There was _no_ way I was going to let myself stop before she did
No way? You're sure? You know some woman runs marathon!:-)
I happened to run from time to time with my mom who used to run Marathon, what made me giggle some of the time is that passerbys could think something like this "oh, nice guy to run with his mother" while in reality it was the opposite: I had to stop before her!
>You never hear people say "oh, macs are nice, but they're just not desktop machines."
No but they say macs are nice, but if you want to play games on your computer, don't get a Mac.
Lack of gaming on Mac and Linux has and will slow down adoption for some desktop usage, that's a fact and putting your head in the sand won't change it..
I know that is the reason I'm not using Linux.. Does WineX support IL2 Forgotten Battle ? Without slowing down the framerate (it is already a resource hog!) ?
And for the home desktop usage adoption, there must be games first, a classic vicious circle which may slow down Linux adoption on home desktop for quite some times..
And there is the problem of dual boot, many users with Linux dual boot with Windows to play games for example, so the market of Linux's game is likely to stay limited for quite some time..
Dual boot is convenient but has some disadvantages, usually when you use one OS you long for some of the advantage of the other OS. And you must patch twice your computer, backup twice, etc..
> how we have to emulate the Windows/Mac/BeOS/Xbox/Sinclair/whatever desktop experience to have a usable desktop are mistaken.
Well, the one thing I loved about BeOS (R.I.P.) was its responsiveness, you were never blocked, KDE or Gnome always felt sluggish on my old Celeron, if Linux became as responsiveness as BeOS was, I would probably switch back from WindowsXP.
Oh and I was using low latency patch for the 2.4 kernel and didn't notice any difference in speed of the UI..
> I think they underestimate the ability of users to adapt, and overestimate the degree to which familiar = better.
Depends on the user, some users are very easily afraid of *any* change, unfortunately..
Well, you can easily setup WindowsXP to look like the 'old thing', so your parents do not really need to adapt.. I know: I did it, I hate the 'new colors' scheme, it looks like a Fisher price toy..
> Why would you want to have a virtual double processor when... you can actually get a second one?
Because it is cheaper? SMT increase very little the size of the CPU and can give some good improvements (depending of the application, and the OS as said in the article).
SMT can work in the same motherboard as a single CPU contrary as what you said..
And for the same price, the single CPU performance of your dual-CPU setup will be lower..
OK, you're worried because SCO stock valuation is very high and you wonder if investors could somehow knows something we've forgotten.
In one word: bullshit! Or more politely, your fears are unfounded.
Do you remember the "Internet bubble"? Why was the valuation of Internet companies ridiculously high?
Sure, some gullible person really believed in the "Internet economy", but mostly it was because the investors were playing the 'hot potatoe' gamble..
That is to say: the stock value is increasing very fast, if I buy now and sell before it crashes, I'll have made a huge profit. Of course like in casino, the hard part is stopping before loosing all your "virtual money"..
IMHO, SCO's investors are playing exactly the same game..
>I use bayesian spam-filtering to cut down on time waisted by spam
And how is that linked to Windows? I'm using Mozilla on Windows and it has a spam-filtering tool too..
>Sometimes I have even been known to forget about several systems because they just work.
Well for client work, Windows XP works pretty well too..
Yes, you have to do security update quite often, but this is true for Linux too.. If you forget some system without updating you may end up with having your server rooted, currently it is less likely than with Windows but only because Linux market share is much lower than Windows..
Recent security problems on Linux servers have shown that hackers are targetting more and more Linux, which is quite logical: as Linux usage grows, the cracker's threat will grow for unpatched Linux servers..
> I saw a great picture after France unilaterally sent troops in Cote d'Ivoire of Ivoirians holding up signs saying "Bush! Save us from the French!"
Well, if you send soldier when two sides are fighting, at least one of the side won't be happy with it, of course..
> Better yet, tell that to all the struggling 3rd world farmers who are oppressed by subsidized European agricultural goods. Economic imperialism at its finest!
Yes, we all know how the US is not protectionist at all with its own market, *sigh*.
That said, I'm French, and not happy at all with the way France has supported dictators, but frankly do you really believe that the USA are better in this respect??
>[Europe]They've certainly managed to do a bang up job in the middle east! Who is spear heading peace in the Middle East right now? The US!
I agree with you that the US has more "diplomatic" power than the Europe, but talking about its diplomatic success with peace in the middle east is a bit ironic no?
Even if we do not talk about Irak, I'm not sure you could call peacefull the Israelian/Palestinian situation..
You're right, that the memory bandwith and parallelism of GPUs are its strong point.
But everyone here is forgetting its weak points: - limited precision: scientific computing usually use double 64bit for computations, currently GPUs are limited to 32bit.. And I don't foresee GPU using doubles any time soon! This level of precision is not needed for 3D graphics calculations. - when you want to read the data back from the GPU, you get little bandwith, but it may become better with PCI express, we'll see soon.
> Moore's law, yes BeOS can do it all on a 60 MHz Pentium I, but no one is running a 60 MHz Pentium I these days.
Only if the bottleneck is the CPU: if your application is busy waiting for the disk or the network, a faster CPU won't help you.. That's why I'm quite pissed at single threaded applications which freeze the whole window when they access the disk or the network: in Windows explorer if I click on a directory with lots of file inside, the whole windows freeze, if I made a mistake, I cannot choose another directory..
And these single threaded apps won't take advantage of the multi-threaded, multi-core CPUs that are coming around (even already there with the "hyper-threaded" P4).
>What features does it have that are (as good as) impossible to port to Linux?
I don't know if those feature are impossible to port on Linux, but so far they haven't been replicated on Linux: 1) fast boot time: BeOS booted to the GUI on ~10s (to a usable GUI! Not like WinXP..), on Linux it take far longer, booting the kernel is slow, and KDE is quite slow to start too. Usually, after this, someone remarks that with Linux, you don't have to stop your computer, true, but my computer is very noisy and I like to sleep at nights! So fast boot time is quite interesting for desktop users, and no LinuxBiOS or equivalent doesn't count:BeOS used the standard BiOS..
2) responsiveness: BeOS apps felt very responsive, you were never "blocked", I think that the extensive usage of thread in the apps was the reason. As a counter-example, there is Mozilla: if it doesn't manage to reach a server for a new page, the whole window can become freezed, in a responsive application, I should be able to continue browsing with the other tabs..
Unfortunately I don't really think it was a BeOS kernel thing, otherwise it would be easy to replicate, I suspect that BeOS guidelines for programming apps were pushing the usage of thread which explains the smooth end-user experience.. And changing the design of Linux applications to become smooth will take a looonnng time, if ever.
Oh and I'm not trolling against Linux, I just explain what my end-user experience of BeOS was: much better than any Linux so far, but with too few apps!!! All BeOS technical prowess meant nothing as there were far too few apps..:-(
Slashdot Mirror
Frankly, you'd better take a little more thought in your answer..
MS: License cost is only a small part of the TOTAL COST OF OWNERSHIP.
Your answer: License cost is a *significant* part of the cost of a PC.
You're answer is 'a cote': the cost of a PC itself is only a small part of the total cost of a PC which include support, training, etc..
Also:
ARGUMENT: MS-Office documents may not open properly in Open Office and visa-versa.
FACT: Isn't this Microsoft's fault? After all, they are the ones that keep changing their applications to make interoperability more and more difficult with each release.
Your answer is a bit 'childish', you're right of course, it is MS fault but who cares??
The fact is that in any company, there is a huge number of existing DOC files and opening those DOC file in OpenOffice doesn't work very well: often the formatting is garbled, and this is not FUD: I'm using OpenOffice at work, so I know it quite well..
The truth is: OpenOffice adoption could only be fast if it had *near perfect* compatibility with Office in reading AND writing (PDF doesn't do it if you want to share a writeable document), which it is a difficult target.
We're currently buying PC with Linux at work, and you know what? We will spend additional money to allow developpers to have an access to MS Office on their Linux's PC, because OpenOffice is not good enough in an 'MS world'..
It would be good enough in an 'OpenOffice world': sharing only documents with other OpenOffice users, but this is not the current real situation..
>> 5. "Open source benefits from anti-American sentiments."
>Personally I'm big into this! I feel that the potential for Europe to regain power and all of that is pretty massive.
>However, outside of the USA there is one big problem - language. You may think that this is getting better - go to Brussels, goto Barcelona and
>see how many 20-30 year olds speak English; not many.
Uh? We're talking about software developers here!
Most of the software developers I know speak English, you simply have to, if you want to be a good software developer..
And I'm from France: a country not especially well-known for the people abilities to speak foreign language.
> All the source packages and RPM's we get come with MD5 sums.
And? This doesn't ensure anything.
MD5 sums only assure you that the RPM has transferred correctly, it don't tell you who has created the RPM, it could be an hacker..
Until we use some certificate to authenticate the provenance of a package, MD5 won't protect you if the main distribution box has been hacked..
>but the inet connection of the user isn't disrupted at all.
Depends on the user, personnaly the official bittorrent client for Windows is the only thing which makes me loose my internet connection..
It is not a big problem as:
1) I'm not using bittorrent very often.
2) I have a 'daemon' reconnecting the connection in case of a loss (needed it when the driver wasn't very good).
> Considering the change to the scheduler to make everything seem snappier,
This is not obvious: I thought that Mandrake included already the preempt + low latency patch in 2.4 ?
In this case, users won't notice much difference..
>Wouldn't you be a little bit more qualified to comment if you had several years of experience with both functional and imperative languages, first?
You're right of course, but it is notheless true, that personnaly I found learning functionnal language quite difficult instead of imperative language, and I'm not the only one.
Plus in the commercial world of today, you're unlikely to have several years of experience with functionnal language..
Academics can have this experience of course, but for some reason, I tend not to value very much academics point of view in this matter: they tend to do different things that in the "real world", such as IHM, database programming, scripts, etc..
That's why we use C and not Pascal today..
I'm not the one having made the original comment but, using exceptions for break or return?
Bleach, this is ugly!
Break and return have a purpose (changing the flow of execution), exceptions a different one (reporting error), mixing both is ugly.
What the point of using a "clean language" if you need to hack around the language?
Me I tried to learn Ocaml and didn't manage to do it..
Now I'm learning Ruby which is much more easy for my poor brain, and it is also the most readable language I know (unfortunately it is slow also).
While your rant started OK (I'm often annoyed by poor doc either in the BIOS or in Windows: the help files treats you like an idiot and don't give you any useful informations), it end up wrong, yes to use Unix you have to know that you get the help with the man command, but I don't think that this is a problem, just a small prerequisite.
There are other many thing wrong in help files on Unix: the info tool is pure hell to use if you're not an emacs user, man is quite primitive (no hyperlink), quite often GUI are interface which call a command-line equivalent in some GUI the name of the command-line is not written..
Yes, sure CDE is perfect..
Give me a break! CDE is a POS, have you tried using dtfile and dtpad?
They won't even work correctly remotely!!
It is the first time that I see applications not working correctly through an export DISPLAY..
Thanks Sun!
Sorry but what matters in the end is that the software works, so this means:
a) a language which ease programmers life.
b) a language which simplifies testing, integrates tests.
If you look only at a), you're missing the whole picture..
A language designer should look at both!
It is no surprise that she works at Sun: I was very surprised that Java doesn't include any 'design by contract' aspect like Eiffel, and when I tried to use it, I undestood: sloppy libraries where Sun had everytime a new library to make the coffee but doesn't take the time to fix the numerous bugs in the existing libraries for years, lots of workaround to use..
I think that some people forgets that even with the most intuitive language in the world, you need tests and the language should support it!
I'm wondering if it depends of the "average density"?
Even if all balls are made from the same materials, you could say that on average large balls are less dense than the small one because when there is more free space between the balls..
I'm curious what would happens if the density of the large balls was increased to compensate for the difference in "average density": what added density should be added so that the large balls sinks or don't move..
>There was _no_ way I was going to let myself stop before she did
:-)
No way? You're sure? You know some woman runs marathon!
I happened to run from time to time with my mom who used to run Marathon, what made me giggle some of the time is that passerbys could think something like this "oh, nice guy to run with his mother" while in reality it was the opposite: I had to stop before her!
> I can tell you with absolute certainty that 68k is most definately RISC, with it's 16-bit length instruction codes.
Depends on your definition of RISC, but for me RISC corresponds also to Load/Store architectures: easily pipelinable instruction set.
So 68k is definitly not a RISC ISA..
>You never hear people say "oh, macs are nice, but they're just not desktop machines."
No but they say macs are nice, but if you want to play games on your computer, don't get a Mac.
Lack of gaming on Mac and Linux has and will slow down adoption for some desktop usage, that's a fact and putting your head in the sand won't change it..
I know that is the reason I'm not using Linux.. Does WineX support IL2 Forgotten Battle ? Without slowing down the framerate (it is already a resource hog!) ?
I really doubt it..
>If the desktops go Linux, the games will too.
And for the home desktop usage adoption, there must be games first, a classic vicious circle which may slow down Linux adoption on home desktop for quite some times..
And there is the problem of dual boot, many users with Linux dual boot with Windows to play games for example, so the market of Linux's game is likely to stay limited for quite some time..
Dual boot is convenient but has some disadvantages, usually when you use one OS you long for some of the advantage of the other OS.
And you must patch twice your computer, backup twice, etc..
> how we have to emulate the Windows/Mac/BeOS/Xbox/Sinclair/whatever desktop experience to have a usable desktop are mistaken.
Well, the one thing I loved about BeOS (R.I.P.) was its responsiveness, you were never blocked, KDE or Gnome always felt sluggish on my old Celeron, if Linux became as responsiveness as BeOS was, I would probably switch back from WindowsXP.
Oh and I was using low latency patch for the 2.4 kernel and didn't notice any difference in speed of the UI..
> I think they underestimate the ability of users to adapt, and overestimate the degree to which familiar = better.
Depends on the user, some users are very easily afraid of *any* change, unfortunately..
Well, you can easily setup WindowsXP to look like the 'old thing', so your parents do not really need to adapt..
I know: I did it, I hate the 'new colors' scheme, it looks like a Fisher price toy..
>A dual processor Athlon motherboard probably costs more than a single processor Pentium 4 motherboard.
;-)
Yes,dual processor Athlon motherboard must be quite rare, especially since AthlonXP do not work in SMP configuration..
That said, a motherboard for dual P4 costs over 200 in France, so this must be taken into account in a performance/price comparison..
> Why would you want to have a virtual double processor when... you can actually get a second one?
Because it is cheaper?
SMT increase very little the size of the CPU and can give some good improvements (depending of the application, and the OS as said in the article).
SMT can work in the same motherboard as a single CPU contrary as what you said..
And for the same price, the single CPU performance of your dual-CPU setup will be lower..
OK, you're worried because SCO stock valuation is very high and you wonder if investors could somehow knows something we've forgotten.
In one word: bullshit! Or more politely, your fears are unfounded.
Do you remember the "Internet bubble"? Why was the valuation of Internet companies ridiculously high?
Sure, some gullible person really believed in the "Internet economy", but mostly it was because the investors were playing the 'hot potatoe' gamble..
That is to say: the stock value is increasing very fast, if I buy now and sell before it crashes, I'll have made a huge profit.
Of course like in casino, the hard part is stopping before loosing all your "virtual money"..
IMHO, SCO's investors are playing exactly the same game..
>I use bayesian spam-filtering to cut down on time waisted by spam
And how is that linked to Windows?
I'm using Mozilla on Windows and it has a spam-filtering tool too..
>Sometimes I have even been known to forget about several systems because they just work.
Well for client work, Windows XP works pretty well too..
Yes, you have to do security update quite often, but this is true for Linux too..
If you forget some system without updating you may end up with having your server rooted, currently it is less likely than with Windows but only because Linux market share is much lower than Windows..
Recent security problems on Linux servers have shown that hackers are targetting more and more Linux, which is quite logical: as Linux usage grows, the cracker's threat will grow for unpatched Linux servers..
> I saw a great picture after France unilaterally sent troops in Cote d'Ivoire of Ivoirians holding up signs saying "Bush! Save us from the French!"
Well, if you send soldier when two sides are fighting, at least one of the side won't be happy with it, of course..
> Better yet, tell that to all the struggling 3rd world farmers who are oppressed by subsidized European agricultural goods. Economic imperialism at its finest!
Yes, we all know how the US is not protectionist at all with its own market, *sigh*.
That said, I'm French, and not happy at all with the way France has supported dictators, but frankly do you really believe that the USA are better in this respect??
>[Europe]They've certainly managed to do a bang up job in the middle east! Who is spear heading peace in the Middle East right now? The US!
I agree with you that the US has more "diplomatic" power than the Europe, but talking about its diplomatic success with peace in the middle east is a bit ironic no?
Even if we do not talk about Irak, I'm not sure you could call peacefull the Israelian/Palestinian situation..
You're right, that the memory bandwith and parallelism of GPUs are its strong point.
But everyone here is forgetting its weak points:
- limited precision: scientific computing usually use double 64bit for computations, currently GPUs are limited to 32bit..
And I don't foresee GPU using doubles any time soon! This level of precision is not needed for 3D graphics calculations.
- when you want to read the data back from the GPU, you get little bandwith, but it may become better with PCI express, we'll see soon.
> Moore's law, yes BeOS can do it all on a 60 MHz Pentium I, but no one is running a 60 MHz Pentium I these days.
Only if the bottleneck is the CPU: if your application is busy waiting for the disk or the network, a faster CPU won't help you..
That's why I'm quite pissed at single threaded applications which freeze the whole window when they access the disk or the network: in Windows explorer if I click on a directory with lots of file inside, the whole windows freeze, if I made a mistake, I cannot choose another directory..
And these single threaded apps won't take advantage of the multi-threaded, multi-core CPUs that are coming around (even already there with the "hyper-threaded" P4).
>What features does it have that are (as good as) impossible to port to Linux?
:BeOS used the standard BiOS..
:-(
I don't know if those feature are impossible to port on Linux, but so far they haven't been replicated on Linux:
1) fast boot time: BeOS booted to the GUI on ~10s (to a usable GUI! Not like WinXP..), on Linux it take far longer, booting the kernel is slow, and KDE is quite slow to start too.
Usually, after this, someone remarks that with Linux, you don't have to stop your computer, true, but my computer is very noisy and I like to sleep at nights!
So fast boot time is quite interesting for desktop users, and no LinuxBiOS or equivalent doesn't count
2) responsiveness: BeOS apps felt very responsive, you were never "blocked", I think that the extensive usage of thread in the apps was the reason.
As a counter-example, there is Mozilla: if it doesn't manage to reach a server for a new page, the whole window can become freezed, in a responsive application, I should be able to continue browsing with the other tabs..
Unfortunately I don't really think it was a BeOS kernel thing, otherwise it would be easy to replicate, I suspect that BeOS guidelines for programming apps were pushing the usage of thread which explains the smooth end-user experience..
And changing the design of Linux applications to become smooth will take a looonnng time, if ever.
Oh and I'm not trolling against Linux, I just explain what my end-user experience of BeOS was: much better than any Linux so far, but with too few apps!!!
All BeOS technical prowess meant nothing as there were far too few apps..