Follow the links. Most of the yes, and some of the no are links to explanations. In particular, he is comparing languages, not compilers (which often compensate for shortcommings in the languages). And he is comparing with a freestanding C++ implementation, and argue why.
In the US, very few find science or technology interesting career choices. The only way the US can compete is through import of skilled workers.
It might be best for the world economy if US stopped the import. It would mean development of new technology would move outside the US, first through outsourcing.
Of course, outsoucing could be outlawed too. Then the US would end up as technological backwater. Which would mean non-technological industries would be unable to compete as well.
Back before I became a law-abiding citizen, my experience was that the more fine grained the privilege system is, the easier it is to crack. The key is that you get first some small, insignificant privileges, and then use these to gain some slightly larger privileges, and so on. On Unix systems, the key to breakin tended to be sgid, nor suid.
If administrators and users are both compentent and careful, breaking in is hard. But it is easier to be incompetent for a complex system, and it is more common to be lazy if less apparently is at stake. "Ok, if someone work around this, they will get access to the shared printer. Not worth worrying about".
With one GB of free network accessible storage, it would actually be useful to be able to access it as a file system. Either as module directly in the kernel, as a nfs-on-gmail user space server, or as one of the virtual file systems supported by the gnome / kde / emacs file system abstraction layer.
If MS has to disclose their API's *now*, it won't help them much if they win the case in five years, and get told that they don't need to disclose them anyway.
In a working market economy, profits will be minimal. If anyone is selling a product with a high margin, some competitor will take makretshare by selling the same product with a lower margin. Either way, profits will be small.
Since the players in the market are motivated by maximizing profits, they will always try to circumvent the market forces, mostly by obstructing their competitors. For a company that holds a monopoly in one area, one way to do this is to bundle products from other areas. This is basically how Microsoft works.
Ensuring a working market is, in my view, the primary responsibility of a government.
The reason you hear otherwise often on/., is that the Libertarians are very visible here. Libetarianism is an ideology founded on the axiom that goivernment is evil, and nothing good can come from it. Thus, libertarians will never be able to understand how a market works.
The quatifications is just one way to organize your "adress book". There are others. And you can ignore it. The default is "friend", stick to that and everything works just fine. I do that, I have less than a screenful of Orkut friends, no need quantifice them.
On the other hand, I have a zillion names in my cell-phone. Here I find it very useful that there is two levels, I use the "vip" level for those I call often, and the default level for everyone else. Which mean I can find the vip numbers very fast.
You know, develop and own a whole new market, and only in the end lose it to a ruthless company with unlimited funds, which none-the-less had to use illegal measures to overtake the market.
And, as a bonus, come out of the situation independently rich.
That is the kind of failure I easily could live with.
Linux is old in the hobbyist market. Linux is the player to beat in the server market. And in the scientific computing market. It is now well-established in the embedded field. It is getting a foothold in the corporate desktop market.
A small company can handle large customers by subcontracting. In fact, this is what outsourcing is about. Rasing profit margins by shrinking the company, by outsourcing specific tasks to companies specialising in those.
We need to get the apps devels to participate too
on
YaST to Become Open Source
·
· Score: 3, Insightful
The whole idea of having the GUI config tools work "on top" of independtly developed applications, written with no thought of GUI, or even non-nerd, configurations, is a loss-loss situation.
What we need is for a standardized way for the application developers to communicate the possible configuration choices and their legal values to the config tools, and for the tools to communicate these choices to the applications.
The interface must be extremely simple to use and light weight in order to be acccepted by the application developers. And it must be stand alone, not depend on any particular framwork or other libraries. The primary interface should be to the application developers, because it is their accept we need first. Our ultimate goal, to serve the users, will have to come next. We won't serve users by having a cool interface that no applications support.
I believe it can be done, though. I got such an interface accepted among Emacs developers, and I suspect similar tools are accepted in the limited domain of KDE and Gnome. That such a tool can exist in the whole domain of free software, is shown by the acceptance of the gettext interface. Those free software projects that do localization, tend to use the gettext interface. Because it is so simple, non-intrusive, and toolkit independent.
If you can consider 626 parlimentarians a "single point of failure". Also, most of the 626 are elected indirectly on party list votes, not directly by personal votes, because most EU nations use proportional voting.
Finally, all the parliament can do is to reject the commision as a whole, which is unlikely to happen based on a single issue.
I'd expect most consumer PC's to be sold with Minimal Windows plus a number of other application bundled, just like many of them are sold with XP Home plus MS Works or another "productivity package" bundled.
Many would bundle MS products, but many would not, thus creating a situation where MS packages would actually have to compete on merit and mindshare, not just by being the default.
That would demonstrate the danger of depending on a single vendor in a way everybody inside and outside EU could understand. Microsoft would have much more to lose on such an action than the EU, who could simply declare that MS by its illegal actions has lost its copyright, or maybe temporarily suspend it while transition to other systems.
The EU commision is composed by people suggested by the national governments, and approved as a whole by the EU parliament. Most of the national governments are elected by the national parliaments, which is mostly composed by people elected on party lists in a propertional system, rather than directly by the voters.
It is an extremely indirect form for democracy, at best, and it is easy to influence by lobbyism and somewhat prone to corruptions. And it is damn slow. However, it is much harder to influence by manipulating the elections. There is no "single point of failure" like with the US president.
My own current project, a 75k lines of code scientific program, has 9 casts. Three are "static_cast", which does not mean "think of this data as something else", but "convert this data to something else". Two are "dynamic_cast", which is a run-time checked cast. Four are "const_cast" which are problematic. Two of them are for interfacing with an extrenal library, and two actually indicate an internal design problems. Anyway most of the other languages you mentioned doesn't even have "const", so any such design problems will be hidden.
There isn't a single "reinterp_cast", which is the kind of cast that you are talking about. They are not needed or common in "modern" C++ code.
There are no old (C) style casts, I compile with a flag that warns about them.
A lot of these patents sound like what we cs nerds discussed at evening in the terminal rooms in the 80'ties. "If the trend of smaller disks continue, then one day you can put them everywhere, like in cameras. Compresses, there could be thousands of images. I don't want to carry around a hard-disc, we could do it today with static ram. In a few year it will be compact enough to store at least as many pictures as film." There was a lot of dicussion about what the net would mean, if everyone had access.
We did not think of ourselves as visionaries, we just average geeks who listed the obvious applications of the improved technology as idle speculation. Applications that was certain to come given the technology. The kind of application that should not be patentable, but has been because the patent system has failed on its purpose, namely to encourage people to disclose inventions that would otherwise be lost to society as a whole.
I believe the "hindsight" argument is overdone. If something is obvious in hindsight, in 99.81% of the cases it is also something many people would come up with once the technology was ready, and thus something that should not be patentable.
> He also made the claim that if you worked at a > BioTech company used Linux to create a new formula > for a drug, you have to GPL the drug. The heck?
I guess this is just a hint that Darl is going after bio-tech companies next. Most of these use Unix in one variant or another, and with Darl logic, their research products must be tainted by SCO IP as well.
Given that the execs gave Peter Jackson a lot of rope with LotR, and he delivered a product that made the company, I suspect they will make Jackson manage the much smaller budget of the Hobbit as he pretty much want to.
There is a huge difference between running binaries under Wine that were never intended to run under wine on one hand, and modifying the source in order to compile binaries that run smoothly under both MS-Windows and Wine.
Of course, if they are actrually going to the trouble of testing their programs under wine, they might as well take the extra step of compiling them under Linux with libwine.
New mail notification: Yes. Encryption: Yes Follow-ups: Probably not. I have ever used the build-in calendar. Forward attached/Inline: Yes Write HTML mail: No Multiple accounts: Yes Customizable keybindings: Yes, extremely:-) Full index search: No, requires an add-on (nnir) Advanced searching: Yes IMAP search: Don't know, I don't use IMAP. Search folders: Yes Spam filter: No build in spam filter. Good support for external spam filters, and good general filtering ability. Handle mailing lists: Yes, if I understand it correctly. Do not download mail rules: Don't know. Labels for e-mail: No, not if they are talking about RMAIL style labels. Create filter from message: No Emoticons: Yes LDAP: No Message threading: Yes Mail storage format: mbox, babyl, mh, usenet, and more...
Slashdot Mirror
The -Os flag has become dramatically better during the 3.x series.
Follow the links. Most of the yes, and some of the no are links to explanations. In particular, he is comparing languages, not compilers (which often compensate for shortcommings in the languages). And he is comparing with a freestanding C++ implementation, and argue why.
There are plenty of fictional universes with clear ownership.
which is "almost over 100% faster".
Where there is a will, there is a way!
In the US, very few find science or technology interesting career choices. The only way the US can compete is through import of skilled workers.
It might be best for the world economy if US stopped the import. It would mean development of new technology would move outside the US, first through outsourcing.
Of course, outsoucing could be outlawed too. Then the US would end up as technological backwater. Which would mean non-technological industries would be unable to compete as well.
You can have one of the two, but not both.
Back before I became a law-abiding citizen, my experience was that the more fine grained the privilege system is, the easier it is to crack. The key is that you get first some small, insignificant privileges, and then use these to gain some slightly larger privileges, and so on. On Unix systems, the key to breakin tended to be sgid, nor suid.
If administrators and users are both compentent and careful, breaking in is hard. But it is easier to be incompetent for a complex system, and it is more common to be lazy if less apparently is at stake. "Ok, if someone work around this, they will get access to the shared printer. Not worth worrying about".
With one GB of free network accessible storage, it would actually be useful to be able to access it as a file system. Either as module directly in the kernel, as a nfs-on-gmail user space server, or as one of the virtual file systems supported by the gnome / kde / emacs file system abstraction layer.
I guess we will see all three (or five) of those.
If MS has to disclose their API's *now*, it won't help them much if they win the case in five years, and get told that they don't need to disclose them anyway.
In a working market economy, profits will be minimal. If anyone is selling a product with a high margin, some competitor will take makretshare by selling the same product with a lower margin. Either way, profits will be small.
/., is that the Libertarians are very visible here. Libetarianism is an ideology founded on the axiom that goivernment is evil, and nothing good can come from it. Thus, libertarians will never be able to understand how a market works.
Since the players in the market are motivated by maximizing profits, they will always try to circumvent the market forces, mostly by obstructing their competitors. For a company that holds a monopoly in one area, one way to do this is to bundle products from other areas. This is basically how Microsoft works.
Ensuring a working market is, in my view, the primary responsibility of a government.
The reason you hear otherwise often on
The quatifications is just one way to organize your "adress book". There are others. And you can ignore it. The default is "friend", stick to that and everything works just fine. I do that, I have less than a screenful of Orkut friends, no need quantifice them.
On the other hand, I have a zillion names in my cell-phone. Here I find it very useful that there is two levels, I use the "vip" level for those I call often, and the default level for everyone else. Which mean I can find the vip numbers very fast.
You know, develop and own a whole new market, and only in the end lose it to a ruthless company with unlimited funds, which none-the-less had to use illegal measures to overtake the market.
And, as a bonus, come out of the situation independently rich.
That is the kind of failure I easily could live with.
And so on.
Linux is old in the hobbyist market. Linux is the player to beat in the server market. And in the scientific computing market. It is now well-established in the embedded field. It is getting a foothold in the corporate desktop market.
The home desktop market is still missing.
A small company can handle large customers by subcontracting. In fact, this is what outsourcing is about. Rasing profit margins by shrinking the company, by outsourcing specific tasks to companies specialising in those.
The whole idea of having the GUI config tools work "on top" of independtly developed applications, written with no thought of GUI, or even non-nerd, configurations, is a loss-loss situation.
What we need is for a standardized way for the application developers to communicate the possible configuration choices and their legal values to the config tools, and for the tools to communicate these choices to the applications.
The interface must be extremely simple to use and light weight in order to be acccepted by the application developers. And it must be stand alone, not depend on any particular framwork or other libraries. The primary interface should be to the application developers, because it is their accept we need first. Our ultimate goal, to serve the users, will have to come next. We won't serve users by having a cool interface that no applications support.
I believe it can be done, though. I got such an interface accepted among Emacs developers, and I suspect similar tools are accepted in the limited domain of KDE and Gnome. That such a tool can exist in the whole domain of free software, is shown by the acceptance of the gettext interface. Those free software projects that do localization, tend to use the gettext interface. Because it is so simple, non-intrusive, and toolkit independent.
If you can consider 626 parlimentarians a "single point of failure". Also, most of the 626 are elected indirectly on party list votes, not directly by personal votes, because most EU nations use proportional voting.
Finally, all the parliament can do is to reject the commision as a whole, which is unlikely to happen based on a single issue.
and thus create true competition.
I'd expect most consumer PC's to be sold with Minimal Windows plus a number of other application bundled, just like many of them are sold with XP Home plus MS Works or another "productivity package" bundled.
Many would bundle MS products, but many would not, thus creating a situation where MS packages would actually have to compete on merit and mindshare, not just by being the default.
That would demonstrate the danger of depending on a single vendor in a way everybody inside and outside EU could understand. Microsoft would have much more to lose on such an action than the EU, who could simply declare that MS by its illegal actions has lost its copyright, or maybe temporarily suspend it while transition to other systems.
The EU commision is composed by people suggested by the national governments, and approved as a whole by the EU parliament. Most of the national governments are elected by the national parliaments, which is mostly composed by people elected on party lists in a propertional system, rather than directly by the voters.
It is an extremely indirect form for democracy, at best, and it is easy to influence by lobbyism and somewhat prone to corruptions. And it is damn slow. However, it is much harder to influence by manipulating the elections. There is no "single point of failure" like with the US president.
My own current project, a 75k lines of code scientific program, has 9 casts. Three are "static_cast", which does not mean "think of this data as something else", but "convert this data to something else". Two are "dynamic_cast", which is a run-time checked cast. Four are "const_cast" which are problematic. Two of them are for interfacing with an extrenal library, and two actually indicate an internal design problems. Anyway most of the other languages you mentioned doesn't even have "const", so any such design problems will be hidden.
There isn't a single "reinterp_cast", which is the kind of cast that you are talking about. They are not needed or common in "modern" C++ code.
There are no old (C) style casts, I compile with a flag that warns about them.
A lot of these patents sound like what we cs nerds discussed at evening in the terminal rooms in the 80'ties. "If the trend of smaller disks continue, then one day you can put them everywhere, like in cameras. Compresses, there could be thousands of images. I don't want to carry around a hard-disc, we could do it today with static ram. In a few year it will be compact enough to store at least as many pictures as film." There was a lot of dicussion about what the net would mean, if everyone had access.
We did not think of ourselves as visionaries, we just average geeks who listed the obvious applications of the improved technology as idle speculation. Applications that was certain to come given the technology. The kind of application that should not be patentable, but has been because the patent system has failed on its purpose, namely to encourage people to disclose inventions that would otherwise be lost to society as a whole.
I believe the "hindsight" argument is overdone. If something is obvious in hindsight, in 99.81% of the cases it is also something many people would come up with once the technology was ready, and thus something that should not be patentable.
> He also made the claim that if you worked at a
> BioTech company used Linux to create a new formula
> for a drug, you have to GPL the drug. The heck?
I guess this is just a hint that Darl is going after bio-tech companies next. Most of these use Unix in one variant or another, and with Darl logic, their research products must be tainted by SCO IP as well.
Bilbo wrote the story after all. He could either read it for a "Frodo-child", or read it for some elven children on the other side of the pond.
I.e. not read it up all the way, but having Bilbo read it as a frame for the movie.
It would also help give the hobbit movie the fairy tale feel of the book.
Given that the execs gave Peter Jackson a lot of rope with LotR, and he delivered a product that made the company, I suspect they will make Jackson manage the much smaller budget of the Hobbit as he pretty much want to.
There is a huge difference between running binaries under Wine that were never intended to run under wine on one hand, and modifying the source in order to compile binaries that run smoothly under both MS-Windows and Wine.
Of course, if they are actrually going to the trouble of testing their programs under wine, they might as well take the extra step of compiling them under Linux with libwine.
Let's see how the old generation compares:
:-)
New mail notification: Yes.
Encryption: Yes
Follow-ups: Probably not. I have ever used the build-in calendar.
Forward attached/Inline: Yes
Write HTML mail: No
Multiple accounts: Yes
Customizable keybindings: Yes, extremely
Full index search: No, requires an add-on (nnir)
Advanced searching: Yes
IMAP search: Don't know, I don't use IMAP.
Search folders: Yes
Spam filter: No build in spam filter. Good support for external spam filters, and good general filtering ability.
Handle mailing lists: Yes, if I understand it correctly.
Do not download mail rules: Don't know.
Labels for e-mail: No, not if they are talking about RMAIL style labels.
Create filter from message: No
Emoticons: Yes
LDAP: No
Message threading: Yes
Mail storage format: mbox, babyl, mh, usenet, and more...