In seventh and eighth grade I had to fight tooth and nail with my teachers to accept my typewritten essays and original fiction writings in English class because, since nobody else produced such work, they were convinced that, if I was using a computer, I was probably cheating and downloading the material from local BBSs.
I'm pretty sure you were in the minority. Was the fighting worth the time to handwrite it?
Life sucks when you know more than, or have it figured out more thoroughly than, 99.9999% of the people around you.
I am curious how you can gauge the ignorance or intelligence of that many people around you (in your case 1million)?
En masse people apply "guilty until proven " to anything which isn't public "knowledge"--even if public knowledge is largely a misconception fed by those who are promoting their own agenda. That sort of thinking is dangerous and can get you run out of town and into homelessness.
That sort of thinking to me seems like common human tendancy. More realistic it is "uncertain" until proven, which is also pretty basic. And since you mention agendas, how can you be certain all the information/education you have acquired is free of any bias?
I have to disagree with your generalizations, not all software induces so many regressions. I'm not saying Mozilla is the worst, but by no means are they the best.... My comment regarding the regressions was based on watching the early development of Mozilla from circa 2000 to about 2004 or so, as of late I do not know if they have taken better step to catch their regressions.
As for bug reports, I do agree detailed problems are better for developers, but I just don't understand how an average user could manage to write bug reports. Someone might realize after some browsing their browser gets slow. After a while they just find that the memory consumption is unbearable. I had an IE user switch to FF circa 1.0 days. He said that after leaving the browser open for 3 straight days, it had some 200MB+ taken. IE in the same browsing did not exhibit that behaviour. Tell me how does that guy file a bug report??? He concluded that FF takes up too much memory and that his fair assessment. I don't consider that "FUD" or whatever.
I'm sure no developer responded to a memory leak bug report by suggesting that you buy more memory. That's pure FUD again.
I never said developer, but its the truth evangelists, advocates, whatever have countered with the "buy more memory" recommendation. Not just on Mozilla but other OSS projects as well. Sadly even on devel mailing lists. Oh well. Now that I can afford it, that's what I do. I just feel bad for those who can't.
Even if I accepted there were no memory leaks, then I would have to accept that Mozilla products just take up way too much memory.
Yes, there are memory leaks, but they are being fixed. There's no need to keep complaining.
The nature of Mozilla development in the past has induced regressions. Correct me if I am wrong but I don't think much has changed, so I don't see why future memory leaks (new features, etc) cannot pop up.
And I disagree, unless users complain about something it typically will not get attention. Many bugs in software are there because there are acceptable work arounds. And after a while, known bugs, become "features". The workaround I've been told, and I quote: "buy more memory, memory is cheap".
So complain, just be reasonable in your expectations and try to be helpful. "The squeaky well gets the grease"
Apple controls its stuff, so the customer can control his own stuff.
So tell me is the iPod I buy "my stuff" or is it "apple's stuff"?
All the freedom in the world is useless if one cannot make use of it. Linux, as great as it is, as much as I like it, as much as I use it, STILL isn't ready for grandma or uncle bob. WHY? Because it takes a guy like me to download (or buy) it, install it, set it all up so that it looks like something they are familiar with.
You have merged different topics. To the best of my understanding there is no freedom provided in buying Apple (or Microsoft) because you must comply with the EULA. Additionally familiarity and usability for "grandma" has to do with the developers or packagers it has nothing to do with vendors controlling hardware or software. I've seen people in their 40's confused with OS X.
And between the two consumer OSes out there, I much rather trust Apple to actually get things right, which means allowing the user to do what the user wants without digging for heinous workarounds and without restrictions.
What is the definition of "trust"? Any company that sells a product to a consumer should be selling what the consumer wants, NOT what the company wants it's consumers to want. How can anybody be so foolish to think that a corporation thinks more about consumers than their profit margin?
In two years, when the choice is VISTA and Mac OS X Tabby I know which one I'll be forced to buy because it will be the ONLY one that will "just work".
This makes no sense. "forced"? You choose to buy.
This is why Linux Fanboys didn't get the iPod, why many don't get the iPhone. It is exactly the same thing, people will pay for something if it makes it easier to use, or more convenient. Think McDonalds vs homemade burgers. I can make my family better tasting burgers if I cook them myself, but sometimes Daddy doesn't want to cook and just wants to feed the kids. The phrase, "I can make a better burger for less cost" doesn't really matter.
I think you need to do some research. Many linux users have iPods. I was impressed at how well (and easy) it was to setup an iPod in Gnome. Plug'n'play with an iPod icon on the desktop. I wonder who programmed that code? Apple developers?
However you are right, people will gravitate towards easy to use. However the concept of "easy to use" becomes worthless if you are restricted to only one environment or circumstances. Granted MP3's are played on the iPod which is okay since most downloaded music is MP3, but what about downloaded video? When everybody wants to a video player that is "easier to use, or more convenient" how will they convert their DivX, Xvid, WMV, etc. to play on their Video iPod? Will that be easy?
The point you seem to be missing is that Apple DOES indeed control "their stuff" however for only their benefit - NOT yours. DRM does not make things easier for you, but easier for Apple. However this is capitalism, it is your choice what you buy, but remember when buying things that implement restrictions you in some ways support those restrictions. I hate thinking that there is a superficial reason why something does not work on some hardware that I personally bought and own.
The simplest solution already used by hundreds of users of newly purchased laptops and desktops with Windows XP is the PC Decrapifier. Originally named the Dell De-Crapifier the name was soon changed to support other machines "with crap" (I'm sure Dell didn't like their name dragged through the crap).
Although this is only for XP, I'll bet a version for Vista will come along soon. Or at least one could hope.
"Also Fedora 7 (dure April) intends to include the nouveau drivers - which is great as out-of-the-box Fedora can't include the binary nVidia driver necessary to have AIGLX working."
I'm not sure what level of confidence can be placed in a very incomplete driver that (AFAIK, please correct me) still has not had any releases. As if the driver will be mature enough by April for even a notable minority of users who probably will be better off with the binary driver.
"And to anyone who thinks this is unnecessary as there is the binary driver - just wait until you card is dropped from the official support and the old driver stops working with some future kernel."
The nature of this comment implies old and outdated hardware. So I fail to see how this is any different from developers discontinueing support on hardware they no longer use or find too difficult to support. Merely implying that because a driver is fully open source, it will be supported indefinitely is not realistic.
Although to be fair, I personally would love to see every single driver for all hardware 100% open source. Pipe dreams, I know.
Aaah, but there is NO lockin to Linux whatsoever. If the license permits, then you CAN port it. I think you are missing the "freedom" implied by the open source. Developers don't like their (typically GPL) code to be used to further the (often unfair) practices of closed source/proprietary OS's (ex: Windows).
What is "true open source"? The license of the code dictates how you can port and how you can run, etc. That is why the authors have to be vocal on their intentions with their code. They are merely opinions - NOT obligatory like a EULA which closed source/proprietary OS's typically enforce on end users.
However, giving people a way to work around bugs in Windows makes them stay longer with Windows. That's why I consider porting software to Windows sabotage. It does not help people under Windows, in the contrary. It makes them stay longer with Windows. And while they stay, they will put pressure on others to also use Windows. It only helps Microsoft.
Its an argument (that often get personal) but does raise some very valid points.
I totally agree. I use gvim along side visual studio when I want to reference some code or just pull up a single file to edit. If gvim had some mechanism to open/close the braces, comments, functions, etc (+/- like in vs.net) -- then I would be complete.
"I think what is going on is simply supply-and-demand curves at work."... I am not saying it is the case, but do you suppose it is possible that an entity with strong market share and corporate power can meddle with "simple" supply and demand? Isn't that the point (speculation, rather) here?
On the other note, lots of things in this country (US) have inflated prices. It's a fact. Offshoring is a good stimulus to the economic position of the US. If you consider the economy of the "planet" as a whole, the only things I see that can seriously counter the disparity of wealth are either a massive flow of jobs elsewhere (offshoring) or strong competition from foreign engineers, innovators, scientists, etc. More power to India.
Disclaimer (I didn't read the article), but I imagine they refer to the inflated market value of a software engineer and the retention costs of good talent. (Which may or may not translate to added costs for the end user.)... I do imagine that the best talent may not thrive in every aspect if compacted in only one company. I would think some competitive nature is required. People will still need to "break the mold" - even if that mold eventually becomes the Google way of doing things.
To answer your question on "whom": the business who must support multiple options.... Not sure how, but if it wasn't clear, the OP was referring to the administrative side, not the end user side. Specifically on this case, the IT cost associated with supporting 2 office solutions.
"In all business cases, supporting two competing, parallel systems is going to cost you more, regardless of what they are."
why is this? Please explain.
Quote simply when you have 2 systems that are meant to do the same thing, there is a high probability that one system will get marginalized. This is not intentional but a common tendancy. It can be seen virtually everywhere.... I downloaded a driver that was supported in 2.4 and 2.6 kernel, the 2.6 worked fine but the 2.4 had a simple bug which I was able to fix - point was that it wasn't tested.... Sometimes we see features implemented first (or better) in a Windows version but not in the Mac (or Linux if lucky) version.... This is only software. Similar problems exist in other aspects of technology. Broadband users supported better than dial-up. Etc.
From the main site: "Rails is a full-stack, open-source web framework in Ruby for writing real-world applications with joy and less code than most frameworks spend doing XML sit-ups".... Curious: so how does this compete with other web frameworks in use (LAMP, J2EE,.Net, etc)? Pros, Cons? Any users/developers here?
"No, I look at the site of the code before I download the code."... Although I do the same, that doesn't necessarily gaurantee security or authenticity. Open source project sites have been hacked in the past. Sometimes there are so many mirrors - no gaurantees.... But in all seriousness, you have to give up somewhere. I'm not sure how many people verify MD5 of SHA (outside of massive downloads) or doublecheck their mirror. As for verifying source code? I don't know if that's practical. I recall a past kernel release which had a filesystem corruption bug. Big oops there. Even if the code/binary is not malicious, doesn't mean you're free from risk. Again, no gaurantees.
I don't understand why people use terms like "love". Its only code executed on some hardware. If people would put personal arguments aside, I think a lot more work could be done.... Is it wrong to have a preference towards Microsoft? Not really. But it could imply a level of close-mindedness in that I doubt most Windows users have used ANYTHING else. But that is not the user's fault.... And for some people who cannot stand "open source" because of the attitudes and philosophy of some of the developers or communities, well that doesn't make much sense either.
A simple formula would be that people should use whatever gets the job done given their personal needs, finances and philosophies. If your philosophical view on software or ability to pay comes before your need for certain functionality, then that is okay. It is your PERSONAL choice. It doesn't need to be imposed on others.
In the end it isn't logical to assail on personal preference.
Out of curiousity... Do you really want a rather simple front end interface to be patented? Do you really think google's was so inventive or unique in the first place? (Assuming you were serious)
I don't pretend to know compilers. I know you can compare machine code output, but runtime of code on target platform is also something to consider. In any event, he should either compare same compiler on different platforms or same platform using different compilers.
"you cannot code C99 in VS C++, even 6.0"... I have compiled C99 code that was originally designed for Unix using an extension I installed in VS6.
You can agree with whoever you want, all code needs to be maintained. I just found maintenance easier in VS. I never made generalizations out my own observations.
"You're saying from one version to the next things break all the time."... I never said nor implied that. What I do say is that things can and sometimes do break. Furthermore I was clearly talking about compilers and source code. Things may compile and work for the most part but there are many times when linking to a newer library can cause something to not work correctly (it may compile but there may be a functionality change). I see this in things like MPlayer all the time.
Not sure why you took it personally. I know many linux users who get their stuff to work as well as myself. However just because it works for one or some, does not imply it can and will work for everyone else. I was merely making observations, NOT generalizations.
As for your comparison, use cygwin/gcc, mingw or something in Windows and compare that. Seems like a more realistic comparison
"you can't expect code to work if you leave it stagnent."... Are you somehow interpretting my expectations? I NEVER EXPECT old stuff coded in GCC to work today. My point about Visual Studio was that my stagnant code worked rather well.... It's not as if I was making a general statement about all code or all compilers. Sheesh.
"I have roughly a 1000 packages installed [some trivial, some large like Gimp]."... What does installed apps have to do with code maintenance?
"I have NO IDEA WHATSOEVER what version of GCC they used to develop/test their software with."... What's your point? Does this means you compiled it? Does this means that NO ONE tests against different versions of GCC?
Aaah, GCC on linux and VS on windows. Unless you're a compiler/OS architect (or know a great on the subject matter), then you are comparing apples and oranges. Try compiling stuff with gcc on windows, that would be better.
"But I believe my point was valid and I stand by it."... No offense, but in all fairness your "point" was more opinion than objective (I didn't see much of an argument made). Which is fine. I'm not one to judge the "validity" of opinions.
"ABI changes are not really an issue in the long run with open source as you can just recompile."... Not true, often you will have to do some recoding/maintenance with gcc changes.
"As for your code from 97 not working now. Shame on you. I have code [even with assembler blocks] that will buil with 2.95.2, 3.4.4 and 4.0.1... Just because you can't code properly doesn't mean others can't."... Why is everybody so immaturely presumptuous? It has nothing to do with proper coding but rather having to do the EXTRA work of maintaining the code. And my point stated was that maintenance was easier with VS. Just because your project worked fine over compilers doesn't mean the thousands of other OSS developers will adhere to the best coding practices.
"VS hasn't really changed that much fundamentally. It's optimizer is weak"... What is your basis for comparison? GCC on Windows vs. VS on Windows?.
"This is because it was improperly coded back then.." Most coding is done to a point where it works exactly as you want. Furthermore I know for a fact that I did not code the Windows stuff correctly and it works okay today.
"and / or you have not upgraded to newer versions.. function calls have changed and progressed.."... Newer versions of what? Even things such as the usage of "main()" have changed. Upgrades in compilers and libraries is what presents these problems.
" where as (as you have explained) Visual Studio has not changed... grown.. or made better any of its internal functions."... Are you speaking out of experience? I know VS has changed significantly. It supports many new things to the point where I can easily extend rather ancient code. But my point was that what it supported in 1997 or 2000 seems to work just as well now. Same cannot be said for linux/gcc, sometimes rewriting is the best option there.
Slashdot Mirror
I'm pretty sure you were in the minority. Was the fighting worth the time to handwrite it?
I am curious how you can gauge the ignorance or intelligence of that many people around you (in your case 1million)?
That sort of thinking to me seems like common human tendancy. More realistic it is "uncertain" until proven, which is also pretty basic. And since you mention agendas, how can you be certain all the information/education you have acquired is free of any bias?
I have to disagree with your generalizations, not all software induces so many regressions. I'm not saying Mozilla is the worst, but by no means are they the best. ... My comment regarding the regressions was based on watching the early development of Mozilla from circa 2000 to about 2004 or so, as of late I do not know if they have taken better step to catch their regressions.
As for bug reports, I do agree detailed problems are better for developers, but I just don't understand how an average user could manage to write bug reports. Someone might realize after some browsing their browser gets slow. After a while they just find that the memory consumption is unbearable. I had an IE user switch to FF circa 1.0 days. He said that after leaving the browser open for 3 straight days, it had some 200MB+ taken. IE in the same browsing did not exhibit that behaviour. Tell me how does that guy file a bug report??? He concluded that FF takes up too much memory and that his fair assessment. I don't consider that "FUD" or whatever.
I never said developer, but its the truth evangelists, advocates, whatever have countered with the "buy more memory" recommendation. Not just on Mozilla but other OSS projects as well. Sadly even on devel mailing lists. Oh well. Now that I can afford it, that's what I do. I just feel bad for those who can't.
Even if I accepted there were no memory leaks, then I would have to accept that Mozilla products just take up way too much memory.
The nature of Mozilla development in the past has induced regressions. Correct me if I am wrong but I don't think much has changed, so I don't see why future memory leaks (new features, etc) cannot pop up.
And I disagree, unless users complain about something it typically will not get attention. Many bugs in software are there because there are acceptable work arounds. And after a while, known bugs, become "features". The workaround I've been told, and I quote: "buy more memory, memory is cheap".
So complain, just be reasonable in your expectations and try to be helpful. "The squeaky well gets the grease"
So tell me is the iPod I buy "my stuff" or is it "apple's stuff"?
You have merged different topics. To the best of my understanding there is no freedom provided in buying Apple (or Microsoft) because you must comply with the EULA. Additionally familiarity and usability for "grandma" has to do with the developers or packagers it has nothing to do with vendors controlling hardware or software. I've seen people in their 40's confused with OS X.
What is the definition of "trust"? Any company that sells a product to a consumer should be selling what the consumer wants, NOT what the company wants it's consumers to want. How can anybody be so foolish to think that a corporation thinks more about consumers than their profit margin?
This makes no sense. "forced"? You choose to buy.
I think you need to do some research. Many linux users have iPods. I was impressed at how well (and easy) it was to setup an iPod in Gnome. Plug'n'play with an iPod icon on the desktop. I wonder who programmed that code? Apple developers?
However you are right, people will gravitate towards easy to use. However the concept of "easy to use" becomes worthless if you are restricted to only one environment or circumstances. Granted MP3's are played on the iPod which is okay since most downloaded music is MP3, but what about downloaded video? When everybody wants to a video player that is "easier to use, or more convenient" how will they convert their DivX, Xvid, WMV, etc. to play on their Video iPod? Will that be easy?
The point you seem to be missing is that Apple DOES indeed control "their stuff" however for only their benefit - NOT yours. DRM does not make things easier for you, but easier for Apple. However this is capitalism, it is your choice what you buy, but remember when buying things that implement restrictions you in some ways support those restrictions. I hate thinking that there is a superficial reason why something does not work on some hardware that I personally bought and own.
The simplest solution already used by hundreds of users of newly purchased laptops and desktops with Windows XP is the PC Decrapifier. Originally named the Dell De-Crapifier the name was soon changed to support other machines "with crap" (I'm sure Dell didn't like their name dragged through the crap).
Although this is only for XP, I'll bet a version for Vista will come along soon. Or at least one could hope.
"Also Fedora 7 (dure April) intends to include the nouveau drivers - which is great as out-of-the-box Fedora can't include the binary nVidia driver necessary to have AIGLX working."
I'm not sure what level of confidence can be placed in a very incomplete driver that (AFAIK, please correct me) still has not had any releases. As if the driver will be mature enough by April for even a notable minority of users who probably will be better off with the binary driver.
"And to anyone who thinks this is unnecessary as there is the binary driver - just wait until you card is dropped from the official support and the old driver stops working with some future kernel."
The nature of this comment implies old and outdated hardware. So I fail to see how this is any different from developers discontinueing support on hardware they no longer use or find too difficult to support. Merely implying that because a driver is fully open source, it will be supported indefinitely is not realistic.
Although to be fair, I personally would love to see every single driver for all hardware 100% open source. Pipe dreams, I know.
Aaah, but there is NO lockin to Linux whatsoever. If the license permits, then you CAN port it. I think you are missing the "freedom" implied by the open source. Developers don't like their (typically GPL) code to be used to further the (often unfair) practices of closed source/proprietary OS's (ex: Windows).
What is "true open source"? The license of the code dictates how you can port and how you can run, etc. That is why the authors have to be vocal on their intentions with their code. They are merely opinions - NOT obligatory like a EULA which closed source/proprietary OS's typically enforce on end users.
Its an argument (that often get personal) but does raise some very valid points.
Thank you for your candor. I agree completely.
I totally agree. I use gvim along side visual studio when I want to reference some code or just pull up a single file to edit. If gvim had some mechanism to open/close the braces, comments, functions, etc (+/- like in vs.net) -- then I would be complete.
"I think what is going on is simply supply-and-demand curves at work." ... I am not saying it is the case, but do you suppose it is possible that an entity with strong market share and corporate power can meddle with "simple" supply and demand? Isn't that the point (speculation, rather) here?
On the other note, lots of things in this country (US) have inflated prices. It's a fact. Offshoring is a good stimulus to the economic position of the US. If you consider the economy of the "planet" as a whole, the only things I see that can seriously counter the disparity of wealth are either a massive flow of jobs elsewhere (offshoring) or strong competition from foreign engineers, innovators, scientists, etc. More power to India.
Disclaimer (I didn't read the article), but I imagine they refer to the inflated market value of a software engineer and the retention costs of good talent. (Which may or may not translate to added costs for the end user.) ... I do imagine that the best talent may not thrive in every aspect if compacted in only one company. I would think some competitive nature is required. People will still need to "break the mold" - even if that mold eventually becomes the Google way of doing things.
To answer your question on "whom": the business who must support multiple options. ... Not sure how, but if it wasn't clear, the OP was referring to the administrative side, not the end user side. Specifically on this case, the IT cost associated with supporting 2 office solutions.
From the main site: "Rails is a full-stack, open-source web framework in Ruby for writing real-world applications with joy and less code than most frameworks spend doing XML sit-ups". ... Curious: so how does this compete with other web frameworks in use (LAMP, J2EE, .Net, etc)? Pros, Cons? Any users/developers here?
"No, I look at the site of the code before I download the code." ... Although I do the same, that doesn't necessarily gaurantee security or authenticity. Open source project sites have been hacked in the past. Sometimes there are so many mirrors - no gaurantees. ... But in all seriousness, you have to give up somewhere. I'm not sure how many people verify MD5 of SHA (outside of massive downloads) or doublecheck their mirror. As for verifying source code? I don't know if that's practical. I recall a past kernel release which had a filesystem corruption bug. Big oops there. Even if the code/binary is not malicious, doesn't mean you're free from risk. Again, no gaurantees.
I don't understand why people use terms like "love". Its only code executed on some hardware. If people would put personal arguments aside, I think a lot more work could be done. ... Is it wrong to have a preference towards Microsoft? Not really. But it could imply a level of close-mindedness in that I doubt most Windows users have used ANYTHING else. But that is not the user's fault. ... And for some people who cannot stand "open source" because of the attitudes and philosophy of some of the developers or communities, well that doesn't make much sense either.
A simple formula would be that people should use whatever gets the job done given their personal needs, finances and philosophies. If your philosophical view on software or ability to pay comes before your need for certain functionality, then that is okay. It is your PERSONAL choice. It doesn't need to be imposed on others.
In the end it isn't logical to assail on personal preference.
Out of curiousity ... Do you really want a rather simple front end interface to be patented? Do you really think google's was so inventive or unique in the first place? (Assuming you were serious)
I don't pretend to know compilers. I know you can compare machine code output, but runtime of code on target platform is also something to consider. In any event, he should either compare same compiler on different platforms or same platform using different compilers.
... I have compiled C99 code that was originally designed for Unix using an extension I installed in VS6.
"you cannot code C99 in VS C++, even 6.0"
You can agree with whoever you want, all code needs to be maintained. I just found maintenance easier in VS. I never made generalizations out my own observations.
"You're saying from one version to the next things break all the time." ... I never said nor implied that. What I do say is that things can and sometimes do break. Furthermore I was clearly talking about compilers and source code. Things may compile and work for the most part but there are many times when linking to a newer library can cause something to not work correctly (it may compile but there may be a functionality change). I see this in things like MPlayer all the time.
Not sure why you took it personally. I know many linux users who get their stuff to work as well as myself. However just because it works for one or some, does not imply it can and will work for everyone else. I was merely making observations, NOT generalizations.
As for your comparison, use cygwin/gcc, mingw or something in Windows and compare that. Seems like a more realistic comparison
"you can't expect code to work if you leave it stagnent." ... Are you somehow interpretting my expectations? I NEVER EXPECT old stuff coded in GCC to work today. My point about Visual Studio was that my stagnant code worked rather well. ... It's not as if I was making a general statement about all code or all compilers. Sheesh.
"I have roughly a 1000 packages installed [some trivial, some large like Gimp]." ... What does installed apps have to do with code maintenance?
... What's your point? Does this means you compiled it? Does this means that NO ONE tests against different versions of GCC?
"I have NO IDEA WHATSOEVER what version of GCC they used to develop/test their software with."
Aaah, GCC on linux and VS on windows. Unless you're a compiler/OS architect (or know a great on the subject matter), then you are comparing apples and oranges. Try compiling stuff with gcc on windows, that would be better.
"But I believe my point was valid and I stand by it." ... No offense, but in all fairness your "point" was more opinion than objective (I didn't see much of an argument made). Which is fine. I'm not one to judge the "validity" of opinions.
"ABI changes are not really an issue in the long run with open source as you can just recompile." ... Not true, often you will have to do some recoding/maintenance with gcc changes.
... Just because you can't code properly doesn't mean others can't." ... Why is everybody so immaturely presumptuous? It has nothing to do with proper coding but rather having to do the EXTRA work of maintaining the code. And my point stated was that maintenance was easier with VS. Just because your project worked fine over compilers doesn't mean the thousands of other OSS developers will adhere to the best coding practices.
... What is your basis for comparison? GCC on Windows vs. VS on Windows?.
"As for your code from 97 not working now. Shame on you. I have code [even with assembler blocks] that will buil with 2.95.2, 3.4.4 and 4.0.1
"VS hasn't really changed that much fundamentally. It's optimizer is weak"
"This is because it was improperly coded back then.." Most coding is done to a point where it works exactly as you want. Furthermore I know for a fact that I did not code the Windows stuff correctly and it works okay today.
... Newer versions of what? Even things such as the usage of "main()" have changed. Upgrades in compilers and libraries is what presents these problems.
... Are you speaking out of experience? I know VS has changed significantly. It supports many new things to the point where I can easily extend rather ancient code. But my point was that what it supported in 1997 or 2000 seems to work just as well now. Same cannot be said for linux/gcc, sometimes rewriting is the best option there.
"and / or you have not upgraded to newer versions.. function calls have changed and progressed.."
" where as (as you have explained) Visual Studio has not changed... grown.. or made better any of its internal functions."