All the GCC 3.x releases attempt to adhere to the same C++ ABI standard. It was developed for IA64, but GCC use it for all platforms. Intel and HP both have compilers (for IA64) that attempts to implement the same standard.
However, bugs were found in the implementation of the ABI standard in GCC 3.0 and 3.1. They had two choices, declare the bugs "the new GCC standard" and screw interoperabilitty with other vendors, or fix the bugs. They choose the later.
So yes, it is the GCC developers "fault" for writting bug in the first place. However, the bugs were somewhat esoteric (the Intel test suite didn't find them), and the standard complex. As a programmer, I won't blame anyone for writting buggy code. Only for not fixing the bugs.
Also note that "most" C++ code won't be affected, however the GCC developers can't really say "link, and hope for the best" when they know some legal constructs will be incompatible.
The organization should properly be named as the above.
Laws that require government offices to use free software does not limit the choice of software, it limits the choice of licenses. Specificaly, it limits the ability to chose a license that bind the purchaser to a single vendor for support and updates.
Obviously, most vendor would like to be the single choice for support and updates, since being the single source allows you to set the price as you see fit (the alternate would be for the pruchaser to switch to another platform, which requires retraining and is extremely expensive).
While a single vendor license may offer some immidiate benfits, for example the vendor may offer a price below development cost in anticipation of later income on support and updates, a license that allow a free choice of vendor may be cheaper on the long run, as competition between vendors will make sure that the need of the purchaser is not ignored and the price is close to cost. Appart from the price, there is alwaus the very real risk that a single vendor will "change its focus" to something that is not you.
The initiative this does not think government offices should be allowed to make license requirements that is in their own long time best interest, but instead alone should choose based on short term tecnical merrit, ignoring the long time benefits of free vendor choice.
A law requiring "you" (singular) or "me" or any other individual or private corporation to only purchase software under free licenses would be bad.
However, any individual, company or organization should be allowed to formulate a policy for which license restrictions they will accept. If an organization believe it is in their best long term interest to only purchase software under licenses that avoid being dependent on a single (or fix set) of suppliers, they should be free to formulate and enforce such a policy.
Anyone who believe that "free vendor choise" will be a benefit in the long term, should try to make organizations they have a stake in adopt such a policy through whatever mechanisms are appropriate for that organization.
If the organization is a government, such a policy take the form of a law, even when the only affects issues internal to the government.
He got an honorary doctorate at Aalborg University in Denmark while I was there are. His acceptence speach was all about why Norway should not join the EU. While it was entertaining, especially since Denmark had joined the EU long before, it was somewhat disappointing for us CS students, who had hoped to hear something about OOP or his latest language, beta.
Thank you for labeling the content of your message clearly.
> Differences:
You mean similarities.
> Faxes are entirely passive: to receive on you > must be ready and prone, and there is no > effective way to screen faxes.
Same is true for email.
> Faxes cost quite a bit more than nothing to > receieve.
Same is true for email, especially at the border of the net. Not everybody is a mp3 downloading teenager in the US with boradband like you.
> E-mail is totally different. It can be actively > and usefully filtered. It costs little to > accept a pre-lim connection, and then block a > message.
The bst blacklist filters get 50% of the spam, and have a positive number of false positives (i.e. real mail accidentially junked) as well.
So using them is a question of whether the time saved in manual filtering is worth the cost of a lost order non and then. Sadly, spam is so numerious and expensive that the trade off of using blacklist usually pay of.
Whitelists like TMDA get close to 100% of the spam, with the risk of loosing valuable messages for people who can't or won't confirm. They aren't easily installed though, and it is unclear whether they will keep working when they become popular.
> I have an account which I never get spam in.
And I have a phone number that never get junk faxes. So what. If you keep your contact info (be it email or fax) a secret, you greatly reduce the risk of getting spammed (some spammers try random numbers or email adresses). You also greatly reduce the chance of getting contacted, which kind of remove the reason for having an email or fax at the first place.
I suspect the "state" is Sterlings analogi is Microsoft.
But in any case:
Last I wrote to a member of the Danish government (an argument against an "opt-in" spam law), the answer came in form of a MS Word document. So the state requires that I posses MS software in order to participate in the democracy.
Jon is currently being prosecuted in Norway for developing software that would weaken the MS monopoly. While that is not the formal charge, that is the formal effect.
Small isolated fragments of the government of a few nations are trying to help liften the Microsoft monopoly, while the vast majority of the governments everywhere are actively or passively helping maintain it.
And no, Sterling basically flamed everyone in his rant. You are just being blinded by your own prejudices to see it.
> Still, it's hardly "contrarian". The sucking up to the audience would make Jon Katz cringe.
Quotes like: "You keep feebly hoping that something will actually work right out of the box, and maybe even look nice. But then you get stuff like Gnome, KDE and Eazel..." hardly seem like sucking up to an audience where where developers of these systems attend.
As far as I can see, Bruce Sterling basically flames everything, wishing himself back to 80'ties.
But he did make the mistake of flaming something from Apple, thus giving the Apple zeolots a change to prove that their fire burn brighter than anyone elses, even on/.
In fact, this spring I was sailing peacefully in my boat at the North Sea, when I was approached by a wooden ship sailing under a skull and bones flag. When it reached my, my boat was boarded by a wild looking one eyed man with a large beard and a hook instead of one hand. Before I could react, he demanded all my possesions, and threatened to play a Spice World DVD for me on a portable computer running Linux unless I complied immediately. When I pointed out that the DVD was encoded and wouldn't be playable on his computer, he just laughed and showed me the DeCSS source. At that point, I had no other options than to comply.
So DeCSS is obviously a piracy tool.
Some claim it can be used for unauthorized copying as well.
It is much easier to convince someone to use a specific flag, than to install a third party patch. The later takes more time, and require you to trust more people.
This is actual a common critisism of bsing a society on the free market alone, made by true conservatives. In your example, the social network is showned to be more important than the market forces. For a conservative, networking and cultural values should take precedense over market forces.
and you will still find some of the most intelligent discussions on moderated Usenet groups, like comp.lang.c++.moderated. The very fact that the vast majority of Usenet groups are crap and increasingly ignored by new users, also makes it possible to create a few oases of intelligent umoderated discussions with a rather select user clientel.
In contrast to earlier, the GCC 3.x ABI is based on a written, cross-vendor standard. This means the implementation can have bugs. The bugs was subtle enough not to be caught before 3.0, but given the diversity of C++ coding styles, it is really hard to say how common the affected code is.
Before, the GCC C++ ABI was whatever GCC produced, so the ABI couldn't really have bugs.
They could of course declare the GCC ABI to be the ABI descrbed in the standard, except for the bugs not caught by 3.0. However, I'm glad they are going with the written standard, rather than trying to Microsoft their bugs into being a de-facto standard.
SUSE was one of the parties which asked for an early 3.2 release, so they could base their distribution on that (instead of 3.1). Red Hat, Debian and FreeBSD was three other names I can remember from the discussion.
After the initial press conference, there quickly came a lot of early confirmations from various respected laboratories. There were also a large number of "early drafts" of papers from teoreticians for teories that should explain the finding.
However, as far as I knew none of it made it through the review process. I guess most of it was withdrawn as more well-planed experiments failed to reproduce the results.
I suspect the real lesson is not the peer-reviewed system itself, but the problems that come when you go around it by publishing through the press, instead of waiting for the system.
I used valgrind for a heavy numeric program, and while it took hours to run even a simple test case, I don't consider it much of a problem given the alternative. Tracking down memory errors can take days and is extremely frustrating.
Valgrind is a program I run if I suspect a memory error, and maybe every half year just to feel safe. It is used a lot less than the debugger, as memory errors are far less common than logical errors, at least for me. But when I need valgrind, it is godsend.
And of course, the really cool thing about valgrind is that it takes no human time to use. You just type
valgrind yourprogram -your arguments
in some terminal window, and go back to working on something else. No special compile or build options, or any manual intervention needed. When it is finished, maybe next day, you read the problem repport.
There are a couple of small ABI bugs in 3.1. After advice from the Red Hat, SUSE, Debian, and FreeBSD teams, the GCC steering committe has decided to release a 3.2 with just the ABI bug fixes, but no new features. The hope is that vendors will standardize on 3.2 for cross-distribution compatibility, and ignore 3.1.
The GCC developers are obviously embarased that ABI bugs was found after 3.0 was releases.
The term "open source" as a replacement for "free software" was coined at a conference sponsored by Tim O'Reilly. There was a lot of free software pings, especially those interested in its commercial value. RMS wasn't invited. Cygnus Support was there, and offered their own trademarked term, SourceWare, but it was rejected. I don't know exactly who came up with the "Open Source", it may have been ESR.
The term "open sources" has obviously been used by information gathering paople, such as journalists, for publically available sources, such as libraries. But a quick look through google confirms that it hasn't been used before the conference for software, except in messages such as "error: could not open source file".
Unfortunately, the term has lost a lot of its value recently, since many paople think it just mean "has source code available", rather than the precise meaning coined at the conference.
Slashdot Mirror
All the GCC 3.x releases attempt to adhere to the same C++ ABI standard. It was developed for IA64, but GCC use it for all platforms. Intel and HP both have compilers (for IA64) that attempts to implement the same standard.
However, bugs were found in the implementation of the ABI standard in GCC 3.0 and 3.1. They had two choices, declare the bugs "the new GCC standard" and screw interoperabilitty with other vendors, or fix the bugs. They choose the later.
So yes, it is the GCC developers "fault" for writting bug in the first place. However, the bugs were somewhat esoteric (the Intel test suite didn't find them), and the standard complex. As a programmer, I won't blame anyone for writting buggy code. Only for not fixing the bugs.
Also note that "most" C++ code won't be affected, however the GCC developers can't really say "link, and hope for the best" when they know some legal constructs will be incompatible.
The organization should properly be named as the above.
Laws that require government offices to use free software does not limit the choice of software, it limits the choice of licenses. Specificaly, it limits the ability to chose a license that bind the purchaser to a single vendor for support and updates.
Obviously, most vendor would like to be the single choice for support and updates, since being the single source allows you to set the price as you see fit (the alternate would be for the pruchaser to switch to another platform, which requires retraining and is extremely expensive).
While a single vendor license may offer some immidiate benfits, for example the vendor may offer a price below development cost in anticipation of later income on support and updates, a license that allow a free choice of vendor may be cheaper on the long run, as competition between vendors will make sure that the need of the purchaser is not ignored and the price is close to cost. Appart from the price, there is alwaus the very real risk that a single vendor will "change its focus" to something that is not you.
The initiative this does not think government offices should be allowed to make license requirements that is in their own long time best interest, but instead alone should choose based on short term tecnical merrit, ignoring the long time benefits of free vendor choice.
A law requiring "you" (singular) or "me" or any other individual or private corporation to only purchase software under free licenses would be bad.
However, any individual, company or organization should be allowed to formulate a policy for which license restrictions they will accept. If an organization believe it is in their best long term interest to only purchase software under licenses that avoid being dependent on a single (or fix set) of suppliers, they should be free to formulate and enforce such a policy.
Anyone who believe that "free vendor choise" will be a benefit in the long term, should try to make organizations they have a stake in adopt such a policy through whatever mechanisms are appropriate for that organization.
If the organization is a government, such a policy take the form of a law, even when the only affects issues internal to the government.
I really doubt that, have you any statistics that shows that the latest versions of XP and IE are already more popular than older versions?
the target market of AOL is people who care deeply what rendering engine their browser use.
There have been several attempts based on GhostScript to produce a real postscript API for X11.
He got an honorary doctorate at Aalborg University in Denmark while I was there are. His acceptence speach was all about why Norway should not join the EU. While it was entertaining, especially since Denmark had joined the EU long before, it was somewhat disappointing for us CS students, who had hoped to hear something about OOP or his latest language, beta.
> What a load of BS.
Thank you for labeling the content of your message clearly.
> Differences:
You mean similarities.
> Faxes are entirely passive: to receive on you
> must be ready and prone, and there is no
> effective way to screen faxes.
Same is true for email.
> Faxes cost quite a bit more than nothing to
> receieve.
Same is true for email, especially at the border of the net. Not everybody is a mp3 downloading teenager in the US with boradband like you.
> E-mail is totally different. It can be actively
> and usefully filtered. It costs little to
> accept a pre-lim connection, and then block a
> message.
The bst blacklist filters get 50% of the spam, and have a positive number of false positives (i.e. real mail accidentially junked) as well.
So using them is a question of whether the time saved in manual filtering is worth the cost of a lost order non and then. Sadly, spam is so numerious and expensive that the trade off of using blacklist usually pay of.
Whitelists like TMDA get close to 100% of the spam, with the risk of loosing valuable messages for people who can't or won't confirm. They aren't easily installed though, and it is unclear whether they will keep working when they become popular.
> I have an account which I never get spam in.
And I have a phone number that never get junk faxes. So what. If you keep your contact info (be it email or fax) a secret, you greatly reduce the risk of getting spammed (some spammers try random numbers or email adresses). You also greatly reduce the chance of getting contacted, which kind of remove the reason for having an email or fax at the first place.
Actually, I suspect it is just because most of the /. crowd are to young to appreciate a good rant or to recognize intelligent humor when they see it.
Instead they get upset when he make fun of their pet peeve, be it Microsoft, OS/X, user friendliness of free software, or the geek culture.
I suspect the "state" is Sterlings analogi is Microsoft.
But in any case:
Last I wrote to a member of the Danish government (an argument against an "opt-in" spam law), the answer came in form of a MS Word document. So the state requires that I posses MS software in order to participate in the democracy.
Jon is currently being prosecuted in Norway for developing software that would weaken the MS monopoly. While that is not the formal charge, that is the formal effect.
Small isolated fragments of the government of a few nations are trying to help liften the Microsoft monopoly, while the vast majority of the governments everywhere are actively or passively helping maintain it.
And no, Sterling basically flamed everyone in his rant. You are just being blinded by your own prejudices to see it.
> Still, it's hardly "contrarian". The sucking up to the audience would make Jon Katz cringe.
/.
Quotes like: "You keep feebly hoping that something will actually work right out of the box, and maybe even look nice. But then you get stuff like Gnome, KDE and Eazel..." hardly seem like sucking up to an audience where where developers of these systems attend.
As far as I can see, Bruce Sterling basically flames everything, wishing himself back to 80'ties.
But he did make the mistake of flaming something from Apple, thus giving the Apple zeolots a change to prove that their fire burn brighter than anyone elses, even on
In fact, this spring I was sailing peacefully in my boat at the North Sea, when I was approached by a wooden ship sailing under a skull and bones flag. When it reached my, my boat was boarded by a wild looking one eyed man with a large beard and a hook instead of one hand. Before I could react, he demanded all my possesions, and threatened to play a Spice World DVD for me on a portable computer running Linux unless I complied immediately. When I pointed out that the DVD was encoded and wouldn't be playable on his computer, he just laughed and showed me the DeCSS source. At that point, I had no other options than to comply.
So DeCSS is obviously a piracy tool.
Some claim it can be used for unauthorized copying as well.
It is much easier to convince someone to use a specific flag, than to install a third party patch. The later takes more time, and require you to trust more people.
It was thought to be the most efficient way to implement some C++ constructs (I don't remember which).
Some other (Lisp?) compilers use similar tricks.
You are reading the wrong threads.
.
This is actual a common critisism of bsing a society on the free market alone, made by true conservatives. In your example, the social network is showned to be more important than the market forces. For a conservative, networking and cultural values should take precedense over market forces.
and you will still find some of the most intelligent discussions on moderated Usenet groups, like comp.lang.c++.moderated. The very fact that the vast majority of Usenet groups are crap and increasingly ignored by new users, also makes it possible to create a few oases of intelligent umoderated discussions with a rather select user clientel.
since the A constructor and destructor will be called fewer times, and they may have side effects.
However, the optimization is explicitly allowed by the standard, thus code that depend on the side effects in the example is broken.
In contrast to earlier, the GCC 3.x ABI is based on a written, cross-vendor standard. This means the implementation can have bugs. The bugs was subtle enough not to be caught before 3.0, but given the diversity of C++ coding styles, it is really hard to say how common the affected code is.
Before, the GCC C++ ABI was whatever GCC produced, so the ABI couldn't really have bugs.
They could of course declare the GCC ABI to be the ABI descrbed in the standard, except for the bugs not caught by 3.0. However, I'm glad they are going with the written standard, rather than trying to Microsoft their bugs into being a de-facto standard.
SUSE was one of the parties which asked for an early 3.2 release, so they could base their distribution on that (instead of 3.1). Red Hat, Debian and FreeBSD was three other names I can remember from the discussion.
After the initial press conference, there quickly came a lot of early confirmations from various respected laboratories. There were also a large number of "early drafts" of papers from teoreticians for teories that should explain the finding.
However, as far as I knew none of it made it through the review process. I guess most of it was withdrawn as more well-planed experiments failed to reproduce the results.
I suspect the real lesson is not the peer-reviewed system itself, but the problems that come when you go around it by publishing through the press, instead of waiting for the system.
Valgrind is a program I run if I suspect a memory error, and maybe every half year just to feel safe. It is used a lot less than the debugger, as memory errors are far less common than logical errors, at least for me. But when I need valgrind, it is godsend.
And of course, the really cool thing about valgrind is that it takes no human time to use. You just type
in some terminal window, and go back to working on something else. No special compile or build options, or any manual intervention needed. When it is finished, maybe next day, you read the problem repport.The GCC developers are obviously embarased that ABI bugs was found after 3.0 was releases.
The term "open source" as a replacement for "free software" was coined at a conference sponsored by Tim O'Reilly. There was a lot of free software pings, especially those interested in its commercial value. RMS wasn't invited. Cygnus Support was there, and offered their own trademarked term, SourceWare, but it was rejected. I don't know exactly who came up with the "Open Source", it may have been ESR.
The term "open sources" has obviously been used by information gathering paople, such as journalists, for publically available sources, such as libraries. But a quick look through google confirms that it hasn't been used before the conference for software, except in messages such as "error: could not open source file".
Unfortunately, the term has lost a lot of its value recently, since many paople think it just mean "has source code available", rather than the precise meaning coined at the conference.
I don't think Hurd is incomplete, just irrelevant.
If Linux hadn't been there, the GNU system would probably be based on a BSD kernel.