Computer science is a part of discrete math (actually it is the part of discrete math dealing with computation, things derived from lambda- and combinator-calculus). How could this be the same as, say physics, which (most of the time) deals with algebra?
Oh, I see, he means computer sciense, not computing sciense!
Setting up a VPN currently, I got an idea for kind of a solution to this: Make the Air network a separate, masqueraded one, have an evil firewall only allowing ssh out. In addition, you may want to look at the ssh-connection and only allow ones that uses ssh-identity authentification, not password. Then let everyone tunnel his/her traffic to his/her home-box via a ssh-channel. This way, the only thing coming out of your box is an encrypted channel to some other box, and all possibly illegal, unencrypted stuff comes out of that box... You still have some problems, but not at all as much as before (for example, you can prove you can not know the content of the packets, and thus can not be required to filter them for RIAA-member-copyrighte-songs, pr0n or whatnot. The only problem you have left is if someone mannages to get into a box which is not his/her own, using ssh with ssh-identity... Which basicly means they have had to craxor this box from somewhere else before, and that's not your problem...
We have teh choice to do that now, too. BSD is out there. However, we have _shoosen_ to use GPLed code.
Are you, a non-employee of MandrakeSoft actually trying to tell _me_ how unprofitable and unexistent the buisinessmodel of the company I work for, and the work I do is???
OK, this is not a nice replay, but such a blatant flame doesn't deserve anything less:
Shut up you bastard and get yourself a real work at some company that creates real software, provides something to real customers and makes real money!
Unfourtunately, the numbers for MandrakeSoft aren't public, so I can not give you them, but I can assure you we won't eat our investors up an leave!
We use software with quite some different, all free, licenses. We do make changes to said software, and releases them under those licenses. All software we create ourselves is GPLed.
Do you really want us to create BSD-licensed software??? _That_ would be to give up our livelyhood, as then SUN or whoever would just steal our code!
It indeed is a small price.
I'm working as an employee of MandrakeSoft. I create Free Software for a living. MandrakeSoft would not exist if there hadn't been any Free Software. At this point, where Operating Systems have become this complex, writing one up from scratch (not only a kernel, but a whole OS including a set of usefull utilities) as a commercial project is just impossible - it would take too much time and money. It wouldn't be competitive (against allready existing OSes).
Thus, we haven't given up anything. Because without Free Software, there wouldn't be any MandrakeSoft not to give something up.
> The GPL is, by the way, a key element in the
> recent woes of Red Hat and other "Linux
> companies." It prevents them from adding unique
> value to their products while at the same time
> undercutting their sales and destroying their
> markets.
As if we would be able to build our own OS from the ground up, all code by our selves. There's just no way we could do this. There's simply no time. The GPL allows us, for the small price of giving our code away too to everyone, put together, and sell, what others have created.
This is not undercutting our sales and destroying our market - this is creating amarket and undercutting old-fashioned software companies' market....
I, for one, know that 50% or so of all _my_ bugs are related to implicit casts. I hate them. I rather do explicit casts allways than that mess. Memory management is much less of a problem.
> The only way to make a profit is to wriggle
> through loopholes in the GPL, and even then it's
> a struggle.
> They can in their spare time, sure. But open
> source rarely pays the bills, which is why most
> programmers work in closed source environments
> where there is enough money to pay their
> paychecks.
This is utterly wrong. I work for MandrakeSoft. I have been working for a closed-source company. Both are startups (about the same age, actually). I earn more than the double here from what I did at that closed source company. MandrakeSoft is going well. And we are NOT making money by trying to go through some loophole of the GPL/LGPL/GFDL! Some companies does, yes, but you don't have to. At all.
What the GPL does, is to require you to innovate (that word sounds a bit bad to me, it have been so missused), to develop, to create new, better versions. If you don't people are just going to copy their versions among frioends. But a new version from the author, you have to get directly from him/her, at least if you want it early. This is not making it next to impossible to make money, just leveling the field and requireing a fair amount of work to compensate for the income, which is not the case in traditional licensing models, where you can sit on your back and get money for something you hacked 10 years ago.
These statements are not _nessesarily_ the opinion of my emploeyr, but my own, and as far as I know, they match those of MandrakeSoft pretty well.
No, it's not free, but it is as free as anything can get. And as fair. Why?
Either, I set a price for my work, my software, and MS have to pay if they want to use it. My price happens to be a license to all of their modifications. This way, the buy my software for theirs.
Or, I give the software away for free, they use it, and then asks me for money if I want to use their modification.
Why do they have the right to ask me for a return, if I do not have the some right?
So no, GPL is not completely free, but it is completely fair, which BSD license is not.
Why is there no penalty for politicians who vote for a non-constitutional law? A nice, economic penalty for those who vote for a law which is later ruled as unconstitutional, would probably remove quite some stupid law-making...
I'm astonished you mentioned the problem, without realizing it: Only the most successfull will be able to clone. But what definition of successfull serves mankind best? Do we really want more buisseness execs? Do we really want more shareholders? Is that what would benefit mankind?
I would rather get more hackers, street artists, theater actrices, scientists and whatever, just not marketing execs, please!
And, who if humans are to decide what genes to duplicate - we defenitely are back at 1945.
No, this is a difference between american and european/nordic law (I only know the law of Sweden, so I can not really tell for the rest of aurope, but looking at the Bill of Rights and the Declaration of Human Rights, I assume this is the case all over aurope):
In the american system, you lose all your rights when convicted.
In sweden, for example, you do not loose most of your rights; e.g. you are still entitled to vote (this right can never be withdrawn).
First of all, a free system is not aimed primarily at making binary aplications work, but at making free aplications, which comes with source, work.
Of course binary compatibility is nuice - it means you, or your software vendor, doesn't have to recompile everything now and then. But it comes at a high price - unexpandability. You can not add a field to a datastructure, since that makes the struct bigger, and breaks compatibility. In source, adding a field is never a problem, and compatibility amounts to preserving old fields that someone might expect, and put values that they won't dislike, into these fields.
Of course, you can do uggly tricks like a hash-table of the extra fields for all objects of one type, that you index with the pointer to the original object. This is for example supported in glib. But it's terribly uggly, and is to beg for problems (like mem. management problems).
I agree however that glibc have had some problems - it hasn't allways been 100% source-compatible...
And - try to search for 100% binary compatibility between say Windows 95 and Windows NT 4.0. Have fuN!
Dah, the slashdot-system ate my second link 'cause I put Href instaed of href (I think) (Hey slashcode coders!). It should have been http://www.aracnet.com/~seagull/NJB/protocol/...
Oups, but, as you had only learned how to memorize, you forgot to think and realize that thinking and memoriztion are two different things.
Kids are today thought how to think. Some of them get rather bright on it. They are not thought how to memorize, or tricks for speedy head-calculation of rather trivial formulas.
And, we need not to teach them what to think, but the mistakes we made, so they kan think of new ones.
FInally someone who admires OS/2's PM/WorkspaceShell! Someone should make a clone for Linux using Gtk/GNOME, instaed of that uggly Explorerish Nautilus!
There is one, and one, single usage for the mouse - selection. Mouse selection of text-areas and mouse-copy-paste is _much_ faster than keyboard. Example: An emacs with two windows (those amacs-windows, not x-windows ones). You want to copy from one of them and paste in the other one. Keyboard: move to start of source CTRL+SPACE, move to end of source, META+w, CTRL+x,CTRL+o, move to insertion point, CTRL+y. Mouse: Klick on source start, move mouse to source end, releasebutton, move to insertion point, klick middle-button.
In addition, I would like to argue that keyboard support is an area where most widgetsets for X sucks - most X programs are unusable from the keyboard! As a _small_ example: There is no key in Gtk+-apps for reaching the menubar. You can reach an item in it if that item has a shortcut, and then move among them with the arrow keys, but there is no general shortcut that allways brings you to the first menubar item (Normally the File menu, so ALT+F should work in most places, but not _allways_)... And there are worse examples.
M$ Winblows _is_ usable without a mouse. A graphical Linux workstation is not (How to you start a shell if you get a GDM and then a stupid window-manager with too little keyboard support? How do you click links in Netscape/Mozilla from the keyboard??).
That's the problem with this scheme: Either you take money for _any_ MB, whatever it contains, or you'l have to check whetever the data is free or money should be collected. In the former case, you can not give anything away for free (this was the scheme proposed). In the latter, the user can allways gzip/gpg/whatever-obscurity-thing-he-wants the data so the tag-searcher thing won't find it.
There are only two possibilities:
Either freedom wins, and you won't have any profits, or profit wins and you won't have an y freedom.
And if I'm downloading Free Software? Or by the way, I set up a web-page with things I want to give away for free? I don't want any ISP to collect per-MB-charges on my free stuff and hand it to Madonna or whoever, and I don't want to recieve any paychecks myself either! This system requires that there is no such thing as gratis information. But there is... And btw, not only does it inhibit free information - it enfoirces the price of all information to be the same.
Because you won't be able to view their content usinga free viewer under a free OS, since a viewer must ensure the data is not copied once extracted from the drive, just shown on the screen. Thus it can not be Open Source - if it where, anyone could easily change to source to allow that. This is where the problem lies.
I have no problem with you wanting to getting enslaved - but I will fight with my teeths if you want me to go the same way.
Slashdot Mirror
Computer science is a part of discrete math (actually it is the part of discrete math dealing with computation, things derived from lambda- and combinator-calculus). How could this be the same as, say physics, which (most of the time) deals with algebra?
Oh, I see, he means computer sciense, not computing sciense!
Setting up a VPN currently, I got an idea for kind of a solution to this: Make the Air network a separate, masqueraded one, have an evil firewall only allowing ssh out. In addition, you may want to look at the ssh-connection and only allow ones that uses ssh-identity authentification, not password. Then let everyone tunnel his/her traffic to his/her home-box via a ssh-channel. This way, the only thing coming out of your box is an encrypted channel to some other box, and all possibly illegal, unencrypted stuff comes out of that box... You still have some problems, but not at all as much as before (for example, you can prove you can not know the content of the packets, and thus can not be required to filter them for RIAA-member-copyrighte-songs, pr0n or whatnot. The only problem you have left is if someone mannages to get into a box which is not his/her own, using ssh with ssh-identity... Which basicly means they have had to craxor this box from somewhere else before, and that's not your problem...
We have teh choice to do that now, too. BSD is out there. However, we have _shoosen_ to use GPLed code.
Are you, a non-employee of MandrakeSoft actually trying to tell _me_ how unprofitable and unexistent the buisinessmodel of the company I work for, and the work I do is???
OK, this is not a nice replay, but such a blatant flame doesn't deserve anything less:
Shut up you bastard and get yourself a real work at some company that creates real software, provides something to real customers and makes real money!
Unfourtunately, the numbers for MandrakeSoft aren't public, so I can not give you them, but I can assure you we won't eat our investors up an leave!
We use software with quite some different, all free, licenses. We do make changes to said software, and releases them under those licenses. All software we create ourselves is GPLed.
Do you really want us to create BSD-licensed software??? _That_ would be to give up our livelyhood, as then SUN or whoever would just steal our code!
It indeed is a small price.
I'm working as an employee of MandrakeSoft. I create Free Software for a living. MandrakeSoft would not exist if there hadn't been any Free Software. At this point, where Operating Systems have become this complex, writing one up from scratch (not only a kernel, but a whole OS including a set of usefull utilities) as a commercial project is just impossible - it would take too much time and money. It wouldn't be competitive (against allready existing OSes).
Thus, we haven't given up anything. Because without Free Software, there wouldn't be any MandrakeSoft not to give something up.
> The GPL is, by the way, a key element in the
> recent woes of Red Hat and other "Linux
> companies." It prevents them from adding unique
> value to their products while at the same time
> undercutting their sales and destroying their
> markets.
As if we would be able to build our own OS from the ground up, all code by our selves. There's just no way we could do this. There's simply no time. The GPL allows us, for the small price of giving our code away too to everyone, put together, and sell, what others have created.
This is not undercutting our sales and destroying our market - this is creating amarket and undercutting old-fashioned software companies' market....
I, for one, know that 50% or so of all _my_ bugs are related to implicit casts. I hate them. I rather do explicit casts allways than that mess. Memory management is much less of a problem.
> The only way to make a profit is to wriggle
> through loopholes in the GPL, and even then it's
> a struggle.
> They can in their spare time, sure. But open
> source rarely pays the bills, which is why most
> programmers work in closed source environments
> where there is enough money to pay their
> paychecks.
This is utterly wrong. I work for MandrakeSoft. I have been working for a closed-source company. Both are startups (about the same age, actually). I earn more than the double here from what I did at that closed source company. MandrakeSoft is going well. And we are NOT making money by trying to go through some loophole of the GPL/LGPL/GFDL! Some companies does, yes, but you don't have to. At all.
What the GPL does, is to require you to innovate (that word sounds a bit bad to me, it have been so missused), to develop, to create new, better versions. If you don't people are just going to copy their versions among frioends. But a new version from the author, you have to get directly from him/her, at least if you want it early. This is not making it next to impossible to make money, just leveling the field and requireing a fair amount of work to compensate for the income, which is not the case in traditional licensing models, where you can sit on your back and get money for something you hacked 10 years ago.
These statements are not _nessesarily_ the opinion of my emploeyr, but my own, and as far as I know, they match those of MandrakeSoft pretty well.
Hm. Won't this upset these orgs. I mean, this is hindering of free trade...
Btw, someone will probably _quickly_ find out a standard way of bridging over these chips if they aren't integrated into some other chips ine one dye.
No, it's not free, but it is as free as anything can get. And as fair. Why?
Either, I set a price for my work, my software, and MS have to pay if they want to use it. My price happens to be a license to all of their modifications. This way, the buy my software for theirs.
Or, I give the software away for free, they use it, and then asks me for money if I want to use their modification.
Why do they have the right to ask me for a return, if I do not have the some right?
So no, GPL is not completely free, but it is completely fair, which BSD license is not.
Why is there no penalty for politicians who vote for a non-constitutional law? A nice, economic penalty for those who vote for a law which is later ruled as unconstitutional, would probably remove quite some stupid law-making...
I'm astonished you mentioned the problem, without realizing it: Only the most successfull will be able to clone. But what definition of successfull serves mankind best? Do we really want more buisseness execs? Do we really want more shareholders? Is that what would benefit mankind?
I would rather get more hackers, street artists, theater actrices, scientists and whatever, just not marketing execs, please!
And, who if humans are to decide what genes to duplicate - we defenitely are back at 1945.
No, this is a difference between american and european/nordic law (I only know the law of Sweden, so I can not really tell for the rest of aurope, but looking at the Bill of Rights and the Declaration of Human Rights, I assume this is the case all over aurope):
In the american system, you lose all your rights when convicted.
In sweden, for example, you do not loose most of your rights; e.g. you are still entitled to vote (this right can never be withdrawn).
First of all, a free system is not aimed primarily at making binary aplications work, but at making free aplications, which comes with source, work.
Of course binary compatibility is nuice - it means you, or your software vendor, doesn't have to recompile everything now and then. But it comes at a high price - unexpandability. You can not add a field to a datastructure, since that makes the struct bigger, and breaks compatibility. In source, adding a field is never a problem, and compatibility amounts to preserving old fields that someone might expect, and put values that they won't dislike, into these fields.
Of course, you can do uggly tricks like a hash-table of the extra fields for all objects of one type, that you index with the pointer to the original object. This is for example supported in glib. But it's terribly uggly, and is to beg for problems (like mem. management problems).
I agree however that glibc have had some problems - it hasn't allways been 100% source-compatible...
And - try to search for 100% binary compatibility between say Windows 95 and Windows NT 4.0. Have fuN!
Dah, the slashdot-system ate my second link 'cause I put Href instaed of href (I think) (Hey slashcode coders!). It should have been http://www.aracnet.com/~seagull/NJB/protocol/...
Check your URLs, crew!
A more working link to the pdf version, and a link to the homepage which includes a MS-word version(!)...
More like rsync probably...
Oups, but, as you had only learned how to memorize, you forgot to think and realize that thinking and memoriztion are two different things.
Kids are today thought how to think. Some of them get rather bright on it. They are not thought how to memorize, or tricks for speedy head-calculation of rather trivial formulas.
And, we need not to teach them what to think, but the mistakes we made, so they kan think of new ones.
FInally someone who admires OS/2's PM/WorkspaceShell! Someone should make a clone for Linux using Gtk/GNOME, instaed of that uggly Explorerish Nautilus!
There is one, and one, single usage for the mouse - selection. Mouse selection of text-areas and mouse-copy-paste is _much_ faster than keyboard. Example: An emacs with two windows (those amacs-windows, not x-windows ones). You want to copy from one of them and paste in the other one. Keyboard: move to start of source CTRL+SPACE, move to end of source, META+w, CTRL+x,CTRL+o, move to insertion point, CTRL+y. Mouse: Klick on source start, move mouse to source end, releasebutton, move to insertion point, klick middle-button.
In addition, I would like to argue that keyboard support is an area where most widgetsets for X sucks - most X programs are unusable from the keyboard! As a _small_ example: There is no key in Gtk+-apps for reaching the menubar. You can reach an item in it if that item has a shortcut, and then move among them with the arrow keys, but there is no general shortcut that allways brings you to the first menubar item (Normally the File menu, so ALT+F should work in most places, but not _allways_)... And there are worse examples.
M$ Winblows _is_ usable without a mouse. A graphical Linux workstation is not (How to you start a shell if you get a GDM and then a stupid window-manager with too little keyboard support? How do you click links in Netscape/Mozilla from the keyboard??).
pre8:
- ReiserFS merge
Nice, nice, nice. Finally they agree.
Could someone please convert that trailer into MPEG? Everyone does not have a QT viewer...
That's the problem with this scheme: Either you take money for _any_ MB, whatever it contains, or you'l have to check whetever the data is free or money should be collected. In the former case, you can not give anything away for free (this was the scheme proposed). In the latter, the user can allways gzip/gpg/whatever-obscurity-thing-he-wants the data so the tag-searcher thing won't find it.
There are only two possibilities:
Either freedom wins, and you won't have any profits, or profit wins and you won't have an y freedom.
And if I'm downloading Free Software? Or by the way, I set up a web-page with things I want to give away for free? I don't want any ISP to collect per-MB-charges on my free stuff and hand it to Madonna or whoever, and I don't want to recieve any paychecks myself either! This system requires that there is no such thing as gratis information. But there is... And btw, not only does it inhibit free information - it enfoirces the price of all information to be the same.
Because you won't be able to view their content usinga free viewer under a free OS, since a viewer must ensure the data is not copied once extracted from the drive, just shown on the screen. Thus it can not be Open Source - if it where, anyone could easily change to source to allow that. This is where the problem lies.
I have no problem with you wanting to getting enslaved - but I will fight with my teeths if you want me to go the same way.
Remembers me of that we have this new movie "chickenrun" out. Is it a coincidence?