Far the largest standard conformance problem in 2.95 was the C++ standard library (v2), which was developed long before the standard. A standard conforming library (v3) was under development, but was not included in 2.96.
So while gcc 2.96 was close to 3.0 from a C point of view, it was much closer to 2.95 from a C++ point of view.
...was actually the largest problem. While the new inliner is much better from a conceptual point of view, the heuristics wasn't changed appropriately. The result was that far too much was inlined, since the new inliner was able to inline far more. Making compilation extremely slow, and even making the code slower (due to cache misses).
And much C++ code depend on a good inliner, i.e. if you want vector to be an alternative to C arrays, the access calls must be inlined.
In fact, the GCC steering commite has decided to release a GCC 3.2 which is just 3.1 plus the ABI bug fixes. The hope is that GNU/Linux/BSD distributors will standardize on the 3.2 ABI, and that no more ABI bugs will pop up.
They should have called it "GNUPro" or something, and made it clear from the version string that this was not based on an FSF release. There never was an official version 2.96 release of gcc
...StarOffice and Apache for the XP box as well, if the purpose is to compare similar equiped boxes?
I don't know about MySQL on win32, and I would guess the Gtk applications would be clearly inferior on XP than on Mandrake, since they provide a native interface on the later.
Everyone loves pointing out that anyone can get their hands on the tools necessary to modify open-source software, but they tend to conveniently ignore the fact that not everyone has the programming skills necessary to do so.
True, which is why I believe that is poor advocacy. Most users will not find the argument impressive, as they know they can't change the code themselves.
The real advantage (from the non-programmers point of view) is that free software gives you the a much larger choice of suppliers. As long as the market exists, someone will be their to support it. With non-free software, you are depending on a single supplier, who may at any time refocus their interrest away from you.
Of course, even with free software the market can become so small that the cost of finding supplier becomes too large. But at least it is your wallet, and not the strategic geniuses in some board room, that decides when that point has been reached.
Good user feedback is essential to a non-commercial free software project, but bad user feedback can kill it. It is the difference between writing
"I love your software, but wouldn't it be cool if it could do XYZZY?"
and
"Your software sucks because it can't do XYZZY!"
The first kind of feedback makes the developers feel appreciated, the second make them think if this is really how they want to spend their free time.
So users essentially have the choice of whether they will be part of the solution, or part of the problem.
Some other user advice:
- Never make demands. It is increadible aggrevating when someone think they have a right to your free time. This also includes formulations like "your project must do XYZZY, otherwise it looks unprofessionel".
- Never make threats, even if you think of them as facts. This includes "unless you implement XYZZY, I'll have to switch to ". If you want to switch, just do it, don't advertise it.
- Never, ever try to take the user community hostage. E.g. "The developer isn't listening to the users, because he doesn't implement XYZZY."
Always remember, it is the developer who (perhaps) do you a favor by releasing his code. You are not doing the developer a favor by using the code. If you feel that relation emotionally stressful, gratis software is probably not for you. Find someone you can pay for the software (whether it is open source or not), in that case it becomes an ordinary economic transaction, where the two parties are equals.
How much are fractions (i.e. numbers of the form x/y) usd in the US? In Denmark, most non-mathematically inclined young people prefer decimals (i.e. numbers of the form xx.yy) because that is what calculators use. In fact, calculation with fractions has recently been dropped from primary school, so the new generations will not even know how to calculate with fractions.
Why do you think we need a new definition of spam?
> Any e-mail I don't want = spam
Wrong, although it is undertsandable why many new users believe so. Most unwanted email they get is spam, so when more experienced users call that spam, they believe it cover all unwanted email.
> I guess Mandrake is sending their newsletter to *@*.* now, right?
Not as far as I know.
> Spam is unsolicited bulk e-mail
Yes. Of course, spam is more than that. We used the term for mass-posted messages to Usenet before junk email became a problem on email. And before the term reached Usenet, it was used for the practice of flooding online text games with messages, preventing other players their turn.
> with a specific commercial pitch or advertisement including a price.
No such requirement have ever been part of the definition of spam. While most email spam have a commercial twist, there are a low that doesn't (Jesus loves you, learn about the supresseed ethnics, help this worthy cause).
The "includes a price" clause is plain stupid, the majority of spam doesn't include spam, it just refers to a web site.
> Introductory e-mails (especially sent to a > specific address), newsletters and business > correspondence is not spam.
Depend on whether it is bulk and unsolicited. It need to be both to be spam.
> In regards to email, the original poster is correct.
He is correct that his definition of spam is new.
> Feel free to look at the definition of SPAM in any anti-spam law.
No law I have heard of define the term SPAM. Some laws outlow specific types of spam, and sometimes email that are not spam, but without defining a specific term for what they outlaw.
Just up from 6 month. It is paid by the state similar to unemployment, allthough some "family friendly" employers like IBM compensate up to full salary.
The same has been true for some time here for "bits and pieces" of endangered species. You have to prove that you got it legally if it is sufficiently new. Such restrictions are likely to be the only reason some of the species are still on the endangered list, rather than the extinct list.
While I suspect you are right that most free software developers have unrelated day jobs (or are students), I suspect most free software is created by people whose day job are related to the software. Most of the really big free software projects (gcc, gdb, gnome, kde, apache, samba, linux) have many full time developers.
... was not started with the intention of replacing a proprietary product with a free one. Rather, the intention was to learn i386 assembler.
XFree86 was (I believe) started as an alternative to the non-free PC X servers, but X itself was started as a project to research networked windowing systems.
GNU and Lindows was intended to replace Unix and MS Windows.
...was bought by Red Hat. As far as I know, they were never unprofitable, allthough in the early days they did get a few friendly development contracts with the FSF to stay in business. These days, it seem like most of Red Hats "wins" (and profit) come from the part of the company that used to be Cygnus.
Cygnus never seriously tried to sell GNU products, instead they sold support. In fact, their original name was Cygnus Support.
At one time they did have some "boxed products", GNUpro and even a Cygwin 1.0 box (which I'm the proud owner of). However, the real money came from support and development contracts.
It was pre-installed on the Mac, and it didn't run on older machines (well, it kind of ran under emulation on the Lisa). It may even have been stored in ROM on the Mac.
It wasn't particularly light hearted either. I have read the book and seen the movie, and the movie catches the mood of the book perfectly.
I agree that it will have to be dumped down a lot to reach to Hollywood idea of a sci-fi audience, i.e. the people who think _The Matrix_ was a "deep" movie.
when they are trying to establish a new free format in competition with existing formats, they use licenses with fewer restrictions.
Examples: libz, the compression format use by gzip and png, which were up against compress and gif. Guile, the GNU scripting and extension language, which were up against TCL. .
it wasn't as much his work on Mozilla, but he had lots of tutorials and kept the "new sites" list, which back then was the primary source of information about the web (it was before indexes and search engines).
Basically, he did a lot of competent "advocacy" work.
...there is no obvious solution on ressource starvation attacks. You can add an arbitrary limit, but arbitrary limitrs are annoying. Why should a person who want an enormous "A" for a poster in Gimp, and who have plenty of virtual memory suffer because of an arbitrary limit? And if we set the arbitrary limit to high, the "bug" will still affect small memory machines, and thus not really be solved.
The software can try to "guess" the limit from information about system memory size, and some heuristics (i.e. guesswork) of how much memory other applications are likely to need. That would obviously be very unreliable.
The least bad "solution" on the server side would probably be a soft limit covering "common uses", with an option to increase or disable the limit using "xset" for the occational Gimp artist who need a huge letter.
However, whether this should be doen depend on the design of X11. X11 is generally designed to be a relatively "thin" server, pushing the UI to the client side. I don't know if X11 is designed to be robust in the case of unreasonable demands from the clients. If not, it might be silly to add checks for font size on the server side, if it doesn't make checks for e.g. pixmap sizes or other client requests. In that case, the check belong on the client side.
The "News for nerds. Stuff that matters." tagline sounds good, but doesn't really say much. So the whiners of course whine that the news isn't for them, and the stuff doesn't matter for them, because that is how whiners reacts.
Unlike mature people who would just go away and find some place targeted towards them, they purile insist that the whole world must revolve around their needs, and thus use the tag line as an excuse to whine when they see articles about the fight for freedoms or for GNU/Linus, which has always been the core of slashdot.
I can understand that a OS/2 advocate would feel homeless these dayes, but the/. focus has never been about one mammut company making a slightly better product than another mammut company, and attracting a horde of fanboys because of that. GNU/Linux is something quite different, it is about freedom, not technology. You would most likely find yourself more at home in the countless technology oriented sites on the net.
Most applications will attemnpt to allocate sufficient memory to handle the task the user assign to it, and depend on the system to refuse the request if there are not enough memory. They then handle the refusal with warying amount of grace. It should not crash the OS, unless the OS itself is broken.
For example, if you feed GCC with ridiculous large input, GCC will (attempt) to allocate ridiculous amount of memory. Which is how it should be, the applications should not try to second guess the user.
Applications that take data from untrusted sources, like web browsers, should course make sanity checks. So the error is in Mozilla, not X11.
Nonetheless, one can expect more from a desktop server like X11 than from more traditional applications, since if the desktop crash all the user visible applications will go with it. So it would be a reasonable feature for X11 to make more sanity checks on its input than other local programs do.
Slashdot is and has always been an advocacy site, and has never prentended to be anything else.
It presents the GNU/Linux and free software side, which is a small step towards bringing balance, as we do not have the big advertisement budgets to buy editorial good will, or money to order favorable rewievs from "the customer is always right" analysis companies.
What I am getting tired of is the the people who whine that slashdot is not Ars Technica or kuro5hin, both excellent web places with a different focus than slahsdot.
What about the fact that we STILL don't really take advantage of gfx hardware for 2D presentation?
What do you mean "we", white man? I have "taken advantage of" 2D gfx hardware under Unix for longer than slashdot (or Linux) has existed.
or the fact that fonts still look like ass?
They fonts don't look "like ass" on my screen. I guess what you want is anti-aliasing. The free technology for that is awailable, it is just a question of installing it. Maybe your OS distributor have done it for you in a sufficiently recent version.
They can't use native widgets for forms because no native widget prowide all the functionality one can specify with style sheets. And given they have to use their own widget set (XUL) for forms, it makes sense to use it for the entire application. Especially since there are other browsers (Galean, K-meleon) that provides the native widget application around the Mozilla (Gecko) rendering engine.
However, Mozilla does the next best thing. It speaks with the native theme manager to draw the XUL widgets, for systems that have a theme manager. So the widgets will look native.
Slashdot Mirror
Far the largest standard conformance problem in 2.95 was the C++ standard library (v2), which was developed long before the standard. A standard conforming library (v3) was under development, but was not included in 2.96.
So while gcc 2.96 was close to 3.0 from a C point of view, it was much closer to 2.95 from a C++ point of view.
...was actually the largest problem. While the new inliner is much better from a conceptual point of view, the heuristics wasn't changed appropriately. The result was that far too much was inlined, since the new inliner was able to inline far more. Making compilation extremely slow, and even making the code slower (due to cache misses).
And much C++ code depend on a good inliner, i.e. if you want vector to be an alternative to C arrays, the access calls must be inlined.
In fact, the GCC steering commite has decided to release a GCC 3.2 which is just 3.1 plus the ABI bug fixes. The hope is that GNU/Linux/BSD distributors will standardize on the 3.2 ABI, and that no more ABI bugs will pop up.
They should have called it "GNUPro" or something, and made it clear from the version string that this was not based on an FSF release. There never was an official version 2.96 release of gcc
...StarOffice and Apache for the XP box as well, if the purpose is to compare similar equiped boxes?
I don't know about MySQL on win32, and I would guess the Gtk applications would be clearly inferior on XP than on Mandrake, since they provide a native interface on the later.
The real advantage (from the non-programmers point of view) is that free software gives you the a much larger choice of suppliers. As long as the market exists, someone will be their to support it. With non-free software, you are depending on a single supplier, who may at any time refocus their interrest away from you.
Of course, even with free software the market can become so small that the cost of finding supplier becomes too large. But at least it is your wallet, and not the strategic geniuses in some board room, that decides when that point has been reached.
Good user feedback is essential to a non-commercial free software project, but bad user feedback can kill it. It is the difference between writing
"I love your software, but wouldn't it be cool if it could do XYZZY?"
and
"Your software sucks because it can't do XYZZY!"
The first kind of feedback makes the developers feel appreciated, the second make them think if this is really how they want to spend their free time.
So users essentially have the choice of whether they will be part of the solution, or part of the problem.
Some other user advice:
- Never make demands. It is increadible aggrevating when someone think they have a right to your free time. This also includes formulations like "your project must do XYZZY, otherwise it looks unprofessionel".
- Never make threats, even if you think of them as facts. This includes "unless you implement XYZZY, I'll have to switch to ". If you want to switch, just do it, don't advertise it.
- Never, ever try to take the user community hostage. E.g. "The developer isn't listening to the users, because he doesn't implement XYZZY."
Always remember, it is the developer who (perhaps) do you a favor by releasing his code. You are not doing the developer a favor by using the code. If you feel that relation emotionally stressful, gratis software is probably not for you. Find someone you can pay for the software (whether it is open source or not), in that case it becomes an ordinary economic transaction, where the two parties are equals.
How much are fractions (i.e. numbers of the form x/y) usd in the US? In Denmark, most non-mathematically inclined young people prefer decimals (i.e. numbers of the form xx.yy) because that is what calculators use. In fact, calculation with fractions has recently been dropped from primary school, so the new generations will not even know how to calculate with fractions.
Nope, in front of.
And the sign will have the subtitle "And for a low, low fee, you can use it."
Why do you think we need a new definition of spam?
> Any e-mail I don't want = spam
Wrong, although it is undertsandable why many new users believe so. Most unwanted email they get is spam, so when more experienced users call that spam, they believe it cover all unwanted email.
> I guess Mandrake is sending their newsletter to *@*.* now, right?
Not as far as I know.
> Spam is unsolicited bulk e-mail
Yes. Of course, spam is more than that. We used the term for mass-posted messages to Usenet before junk email became a problem on email. And before the term reached Usenet, it was used for the practice of flooding online text games with messages, preventing other players their turn.
> with a specific commercial pitch or advertisement including a price.
No such requirement have ever been part of the definition of spam. While most email spam have a commercial twist, there are a low that doesn't (Jesus loves you, learn about the supresseed ethnics, help this worthy cause).
The "includes a price" clause is plain stupid, the majority of spam doesn't include spam, it just refers to a web site.
> Introductory e-mails (especially sent to a
> specific address), newsletters and business
> correspondence is not spam.
Depend on whether it is bulk and unsolicited. It need to be both to be spam.
> SPAM is Unsolicited Bulk Email. Period.
Correct.
> In regards to email, the original poster is correct.
He is correct that his definition of spam is new.
> Feel free to look at the definition of SPAM in any anti-spam law.
No law I have heard of define the term SPAM. Some laws outlow specific types of spam, and sometimes email that are not spam, but without defining a specific term for what they outlaw.
Just up from 6 month. It is paid by the state similar to unemployment, allthough some "family friendly" employers like IBM compensate up to full salary.
The same has been true for some time here for "bits and pieces" of endangered species. You have to prove that you got it legally if it is sufficiently new. Such restrictions are likely to be the only reason some of the species are still on the endangered list, rather than the extinct list.
I don't see any problems with that.
While I suspect you are right that most free software developers have unrelated day jobs (or are students), I suspect most free software is created by people whose day job are related to the software. Most of the really big free software projects (gcc, gdb, gnome, kde, apache, samba, linux) have many full time developers.
... was not started with the intention of replacing a proprietary product with a free one. Rather, the intention was to learn i386 assembler.
XFree86 was (I believe) started as an alternative to the non-free PC X servers, but X itself was started as a project to research networked windowing systems.
GNU and Lindows was intended to replace Unix and MS Windows.
[ Yes, I know I'm picking nits ]
...was bought by Red Hat. As far as I know, they were never unprofitable, allthough in the early days they did get a few friendly development contracts with the FSF to stay in business. These days, it seem like most of Red Hats "wins" (and profit) come from the part of the company that used to be Cygnus.
Cygnus never seriously tried to sell GNU products, instead they sold support. In fact, their original name was Cygnus Support.
At one time they did have some "boxed products", GNUpro and even a Cygwin 1.0 box (which I'm the proud owner of). However, the real money came from support and development contracts.
It was pre-installed on the Mac, and it didn't run on older machines (well, it kind of ran under emulation on the Lisa). It may even have been stored in ROM on the Mac.
It wasn't particularly light hearted either. I have read the book and seen the movie, and the movie catches the mood of the book perfectly.
I agree that it will have to be dumped down a lot to reach to Hollywood idea of a sci-fi audience, i.e. the people who think _The Matrix_ was a "deep" movie.
when they are trying to establish a new free format in competition with existing formats, they use licenses with fewer restrictions.
Examples: libz, the compression format use by gzip and png, which were up against compress and gif. Guile, the GNU scripting and extension language, which were up against TCL.
.
it wasn't as much his work on Mozilla, but he had lots of tutorials and kept the "new sites" list, which back then was the primary source of information about the web (it was before indexes and search engines).
Basically, he did a lot of competent "advocacy" work.
...there is no obvious solution on ressource starvation attacks. You can add an arbitrary limit, but arbitrary limitrs are annoying. Why should a person who want an enormous "A" for a poster in Gimp, and who have plenty of virtual memory suffer because of an arbitrary limit? And if we set the arbitrary limit to high, the "bug" will still affect small memory machines, and thus not really be solved.
The software can try to "guess" the limit from information about system memory size, and some heuristics (i.e. guesswork) of how much memory other applications are likely to need. That would obviously be very unreliable.
The least bad "solution" on the server side would probably be a soft limit covering "common uses", with an option to increase or disable the limit using "xset" for the occational Gimp artist who need a huge letter.
However, whether this should be doen depend on the design of X11. X11 is generally designed to be a relatively "thin" server, pushing the UI to the client side. I don't know if X11 is designed to be robust in the case of unreasonable demands from the clients. If not, it might be silly to add checks for font size on the server side, if it doesn't make checks for e.g. pixmap sizes or other client requests. In that case, the check belong on the client side.
Unlike mature people who would just go away and find some place targeted towards them, they purile insist that the whole world must revolve around their needs, and thus use the tag line as an excuse to whine when they see articles about the fight for freedoms or for GNU/Linus, which has always been the core of slashdot.
I can understand that a OS/2 advocate would feel homeless these dayes, but the /. focus has never been about one mammut company making a slightly better product than another mammut company, and attracting a horde of fanboys because of that. GNU/Linux is something quite different, it is about freedom, not technology. You would most likely find yourself more at home in the countless technology oriented sites on the net.
Most applications will attemnpt to allocate sufficient memory to handle the task the user assign to it, and depend on the system to refuse the request if there are not enough memory. They then handle the refusal with warying amount of grace. It should not crash the OS, unless the OS itself is broken.
For example, if you feed GCC with ridiculous large input, GCC will (attempt) to allocate ridiculous amount of memory. Which is how it should be, the applications should not try to second guess the user.
Applications that take data from untrusted sources, like web browsers, should course make sanity checks. So the error is in Mozilla, not X11.
Nonetheless, one can expect more from a desktop server like X11 than from more traditional applications, since if the desktop crash all the user visible applications will go with it. So it would be a reasonable feature for X11 to make more sanity checks on its input than other local programs do.
It presents the GNU/Linux and free software side, which is a small step towards bringing balance, as we do not have the big advertisement budgets to buy editorial good will, or money to order favorable rewievs from "the customer is always right" analysis companies.
What I am getting tired of is the the people who whine that slashdot is not Ars Technica or kuro5hin, both excellent web places with a different focus than slahsdot.
What do you mean "we", white man? I have "taken advantage of" 2D gfx hardware under Unix for longer than slashdot (or Linux) has existed. They fonts don't look "like ass" on my screen. I guess what you want is anti-aliasing. The free technology for that is awailable, it is just a question of installing it. Maybe your OS distributor have done it for you in a sufficiently recent version.They can't use native widgets for forms because no native widget prowide all the functionality one can specify with style sheets. And given they have to use their own widget set (XUL) for forms, it makes sense to use it for the entire application. Especially since there are other browsers (Galean, K-meleon) that provides the native widget application around the Mozilla (Gecko) rendering engine.
However, Mozilla does the next best thing. It speaks with the native theme manager to draw the XUL widgets, for systems that have a theme manager. So the widgets will look native.