Linux developer sought
My employer is looking to hire a Linux software developer with the following requirements: - 3+ years of experience - University degree (BA or Bsc.) - Windows or Linux knowledge (or Expert in Linux) - Networking (TCP/IP, drivers) - Independent - Security understanding/knowledge is a plus - System understanding and good analysis capabilities - Database knowledge is a plus - C++ is a plus Please respond to me directly rather than to the list. Thanks, Adi Stav [EMAIL PROTECTED] +972-52-3514449 = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Religous War Bait (sic) (was: Re: Perl Meeting)
On Tue, Feb 26, 2002 at 10:19:34PM +0200, Omer Zak wrote: > > Give us a break, let's hold the meeting in a classical Sushi place, in a > more pleasant city, and a decent time of the week such as Saturday 12:00. > Give us one more break, and make the meeting one of Python and Scheme > fans, rather than of Perl addicts. > > Anything else about which one can start a silly argument? Yes. Let's hold it in English -- Hebrew is totally unsuseable in discussion of technical matters :) = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Gnome2beta
On Sat, Feb 23, 2002 at 12:56:39AM +0200, Oren Held wrote: > > I even had > to compile python2 with xml/expat modules in order for libglade to get > compiled (I think it's needed only for the compilation, though. it'll be > sad if it depends on python). I know that libglade has explicit support for Python (it's a very language-dependent library, afterall). I guess that this support is the reason for the Python dependency, so it might be worth checking if it can be compiled without it. Unless you want to run Python programs that use libglade, that is. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Fwd: Re: Basic Compiler
On Mon, Feb 11, 2002 at 04:18:03AM -0800, Boaz wrote: > Moreover, Logo is not dead as you might think. > MIT developed StarLogo which is similar to Logo but has some turtles > and not only one. It is useful for Artificial Life researches. > > See: http://agents.www.media.mit.edu/groups/el/Projects/starlogo/ Ye Gawds... First object-oriented FORTRAN, and now multi-threaded Logo... What will they invent next!? Functional assembly? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Linux filenames with definite encoding (Was: FTP server with intl support)
On Sun, Feb 10, 2002 at 08:48:36PM +0200, Ilya Konstantinov wrote: > > Before Linux FTP daemons could offer filenames in a definite encoding, > Linux needs some way to get a definite of a filename, and AFAIK the > kernel offers no such standard way (via an extended version of readdir > etc.). Moreover, ext2 doesn't even define a standard encoding for > filenames (unlike NTFS filesystems, which define Unicode as the internal > filename storage form, removing the encoding ambiguity). > > Are there any functions for this on other Unixes? (I guess such > functions weren't defined by POSIX) Or will Linux / glibc need to make > up its own function names for this functionality? (if we'd ever apply > with this problem to the Linux kernel maintainers and they'd agree to > fix it) UNIX treats all user data as opaque in principle, so explicitly admitting that file names may have different encodings in an API would be very ugly (but see the newly-accepted "extended attributes" API in 2.5.x). I cannot see any benefit in having a single filesystem allow different files to have different encodings for their names, so the only issue is encoding conversion done in each filesystem's code between the on-disk representation and some standard Linux encoding (UTF8?). I assume that such conversion is done because Linux can be configured to support certain Windows charset codepages explicitly. So this is a pure userspace issue -- if all the files you create on your system are named with a standard encoding then there's no problem for an FTP server to convert to it any encoding the client requests. And if you /don't/ create all the files on your system with the same encoding, no kernel feature can help you. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: The Great Kernel CVS Mutiny
On Sun, Jan 20, 2002 at 08:33:33AM +0200, Shlomi Fish wrote: > > I disagree with your claims. And by raising the flag of mutiny I do not > intend to demand them. I intend to implement a system that will make them > a reality. > > This is a productive mutiny, in which people do something instead of just > whine, attack, or spread FUD. I know you disagree with our clainms. Again: why would your suggestion not lead to a reduction in kernel coherence and quality as we described? > Like I said, I believe it's impossible for Linus to pass his decision for > each and every patch that goes in. OK - whether the VM should be replaced > - should be ultimately his decision. However, if there's a patch to the > VM, then the VM maintainer or its architect should be able to apply it > without consulting LT. Alright. Right now it is Linus who is the maintainer of the VM. Can you suggest someone who would do the job better? > As of now, from what Moshe Bar said, there were many patches, which fix > important bugs, and were rejected by Linus, since only he has an ultimate > repsonsibility on what goes into linux-2.[45].x-vanilla. Moshe said that > the S.u.s.e kernel had 200 patches to apply. By definition (according to > the Joel test which I referenced) the kernel should be at any point > bug-free. This is why using a CVS and delegating responsibility for > applying patches is a must. I don't think you completely understand the nature of VM debugging. More often than not, it's not a straightforward business like "Oops, NULL pointer dereference -- I forgot to initialize", where it's very clear both where the bug is and what the fix does. VM debugging can, perhaps, more often be described as "tuning". No VM can behave perfectly, or even well, in all circumenstances, so the simplistic goal of "bug free" does not apply. Instead, the VM hackers try to have the system general enough and yet respond well to all reasonable cases. That's very hard to do because you can't really know in advance which tuning directions truely solve the problems and which only appear to do so due to the by-definition limited range of behavior models that the developer's initial testing covered. That was exactly the problem with Rik's original 2.4.0 VM -- it appeared to be perfect, but the larger user base introduced by the even version number uncovered serious problems with it. So you can't really separate the global policy making with taking in VM patches when it comes to VM tuning. Thinking, in advance, which VM patches would improve the system in the long run and which would not require such qualities as complete understanding of the entire system, complete understanding of the usage patterns by userspace code and by users, consulting the contradicting suggested ways to solve the problem, ability to weigh the existing arguments and testing results for and against a patch, and, of course, a great deal of intuition. I doubt that in the last months before he passed the kernel on to Marcelo Linus did much for the kernel other than just this. Now, you can say his judgement was mistaken in some choices, perhaps he should have been more aggressive in accepting patches or less aggressive at other times. But if Linus didn't do this job, someone else would still have to. You can't evade that. > > > to see that I'm not the only person who thinks so. And give me another > > > example for a project with the scope and number of developers of the Linux > > > kernel that does not use a source control system. > > > > Give me another example for an open source operating system that > > captured a significant market share. > > > > Give me another example for a successful major operating system in > > the market that is not controlled by a single vendor. > > > > Give me another example of successful operating system developed in > > the last 10 years that uses monolothic kernels. > > > > Give me another example of a successful operating system that does > > not include any management utilities or libraries. > > > > These are all non-sequitors. I don't see how they are derived from the > original request for an example. They intend to show that the existence or the lack of an example of other projects working similarly to Linux has no relevancy to the issue at hand, unless one is willing to consider every uniqueness of Linux a weakness. > My article referenced the Joel Test, which Joel Spolsky claims is a better > alternative than SEMA. I don't know if the Joel Test is about top-down or > bottom-up software. In fact, I think it applies to software in general. It can't be used with the bottom-up development model as it requires writing code according to a specification and a schedule. > My model can be used whether for bottom-up or top-down software > development. No, it cannot. Your model assumes that the role of the top developer is to direct the project (since that is his only way to control it), while in a bottom-up approach the to
Re: [OT] Linux Kernel Mentality
On Mon, Jul 09, 2001 at 12:27:43AM +0300, Hetz Ben Hamo wrote: > H? David, > > Yes, it's a well known story about the "fuck" word on the kernel - people > were very frustrated while developing the kernel for various platforms > (specially Sparc 64) and they added this "fuck" thing (you'll find it mostly > there). Linus refuse to accept any patch that will clear this - and he's > right - it's people work, free expression ;) IIRC one of the more compelling arguments in favor of removing the words was that it made the kernel source code illegal to distribute in a surprisingly large number of courtries... I'm glad it wasn't enough :) According to the graph, most of the rise in the popularity of "fuck" is concentrated in the beginning of the 2.1 series. Now what happened there... Oh yeah, the introduction of fine-grained SMP :) = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: LInux modem problem
On Fri, Jul 06, 2001 at 03:41:46PM -0700, oren epshtein wrote: > i have fast rs pci 56k modem and i don know how to > install it... > i read all manuals i can- but nothing is helping > when i use the kppp with Query Modem- it says the > modem is busy (my is in ttys2) > > > i did minicom -s and configured all well- and when i > write > minicom it says that i have a input/output error or > something like this... > > > i have tried to download the last stable Kernel but it > didnt help... > > what can i do? Most probably, you have a winmodem that has no or non-trivial support in Linux. If you search for your model and Linux you should probably find out whether it is supported or not. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: samba
On Fri, Jul 06, 2001 at 04:14:20PM -, [EMAIL PROTECTED] wrote: > > Right now you have 2 samba installed - one in /usr/bin and one in > > /usr/local/bin > All source projects are built by default to prefix=/usr/local, while rpm are > built with prefix=/usr, is there a good reason for that? Indeed there is. You might want to read the FHS (Filesystem Hierarchy Standard) sometimes, I personally find it a very interesting and informative document about the filesystem design ideas. To sum it up though, the idea is to avoid clashes between the system's software and the administrator's. For example, the system might require Apache 1.3 in /usr while the admin installs 2.0 in /usr/local without having to worry about hitting some name in RPM. They can then also find all of their installed software and remove it easily, or export the entire /usr/local hierarchy through NFS or equivalent to the entire local network. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: (main_buf = *main_buf_p) is not syntacticly like (i = *j) ?
On Tue, Jul 03, 2001 at 03:10:05PM +0200, Shachar Shemesh wrote:
> short C++ code:
>
> void A()
> {
> try
> {
> class someclass var1(constructor_arguments);
>
> B();
> some_more_activities();
> } catch( ... )
> {
> some_exception_code();
> }
> }
>
> void B()
> {
> class someotherclass var2( constructor_arguments );
>
> ...
> C();
> }
>
>
> void C()
> {
> blah();
>
> if( adi_stav_was_here==true )
> throw NOOO(5);
> }
>
> If we try to (as CFront used to do) compile this code into C, we see
> that the long jump triggered by the throw in function C must stop inside
> function B, in order to run the var2 destructor. This catch handler can
> then long jump to function A to handle the explicit catch handler.
I've never used CFront. But why would an implicit exception handler
be different from an explicit one? That could easily be implemented
by having each exception handler keep a pointer to the previous jmp_buf,
restore it once it finishes (either naturally or through exception).
> Conclusion 1 - if you enable exception handling in C++, all your
> functions probably carry exception handlers (there are good things to
> say for Java's "you must declare exceptions" syntax).
If it makes you happy, I've always considered those kind of safe
destructors the most useful thing in C++, and the lack of useable
exceptions the most annoying feature-lack of C.
> Conclusion 2 - exceptions are not global, far from it.
You always have only one active exception handler at a time. That's
what I mean by "global". Can you imagine a situation where you'd want to
pass one of two jmp_buf pointers, say, based on the value of a flag?
> For those of you still only programming in C, replace "destructor" with
> "cleanup code".
Actually I find that a lot of people nowdays start with C++ or Java and
move to C later. Like I did, even though I switched very early
experience-wise.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: Backup files with ordinary user.
On Tue, Jul 03, 2001 at 01:26:08PM +0300, Ben-Nes Michael wrote: > Hi All > > I want to backup my servers files using tar from a central Location. > To do this im using ssh with RSAauth to log from my backup server to all the > other servers. > > I log using ordinary user to be on the safer side but then I can't read some > directories ( for instance: /var/spool/croon - rwx__ ) > I thought of few options: > > 1. adding the ordinary user to root Group > 2. add access to some directive like chgrp or chmod > 3. tar as root - I don't like this option > > What is the list suggestions ? It seems to me that the first and second options can cause or hide more security risks than the third. Why not just tar as root? If what you are worried about is root accessing the network, you could write a script that creates the tar file as root, and then drops priveleges for the actual transmission. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: (main_buf = *main_buf_p) is not syntacticly like (i = *j) ?
On Tue, Jul 03, 2001 at 12:13:47AM +0300, Shaul Karl wrote: > Thank you for the responses. > What I have done wrong is already pointed out. > If you are curious what I am trying to do: > > 1. I wanted to have jmp_buf* passed as a function parameter in order to avoid > a global variable. Therefore, in the function body I had to use a local > variable and initialize it with this pointer: > jmp_buf local_variable = *function_paramter; You're using jmp_buf as an exception throwing mechanism. I'd say it WOULD make sense to have jmp_buf a global (or per-thread, yuck) variable, since exceptions are global by nature. It would be also more portable, unless the standard mandates that jmp_buf is an array -- anyone happens to know if it does? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: (main_buf = *main_buf_p) is not syntacticly like (i = *j) ?
On Mon, Jul 02, 2001 at 03:32:24PM +0300, Nadav Har'El wrote:
> On Mon, Jul 02, 2001, Shaul Karl wrote about "(main_buf = *main_buf_p) is not
>syntacticly like (i = *j) ?":
> > #include
> >
> > int main()
> > {
> > jmp_buf test_env;
> > jmp_buf *test_env_p = &test_env;
> >
> > test_env = *test_env_p;
> > return 0;
> > }
> > jmp_test.c:9: incompatible types in assignment
> >
> > What did I miss?
>
> Well, the problem is the definition of jmp_buf: in Linux,
> typedef struct ... jmp_buf[1];
>
> Which means the jmp_buf type is an array. In C you can't normally assign
> arrays like you did (because C thinks you're trying to assign pointers, rather
> than the content of the array), so either do
> *test_env=**test_env_p;
>
> (to assign the first element, but this is highly dependent on the linux
> implementation),
>
> or the more portable (and thus better) approach is to use memcpy() to copy a
> jmp_buf.
>
> BTW, don't forget the following "NOTES" from the Linux setjmp() manual:
> "setjmp() and sigsetjmp make programs hard to understand
>and maintain. If possible an alternative should be used."
BTW: if we're already talking about setjmp(), longjmp() and difficulty to
understand, has anyone read the article from the GNU Pth source
distribution? It deals with applications of setjmp() and similar calls in
thread implementations, and I found it fascinatingly mindboggling.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: (main_buf = *main_buf_p) is not syntacticly like (i = *j) ?
On Mon, Jul 02, 2001 at 03:00:21PM +0300, Shaul Karl wrote:
> I have two 11 lines C programs which are supposed to be syntacticly similar.
> Yet int_test.c get compiled while the other does not:
>
> Script started on Mon Jul 2 14:51:10 2001
> [14:51:10 tmp]$ more *_test.c
> ::
> int_test.c
> ::
>
> #include
>
> int main()
> {
> int i;
> int *j = &i;
>
> i = *j;
> return 0;
> }
> ::
> jmp_test.c
> ::
>
> #include
>
> int main()
> {
> jmp_buf test_env;
> jmp_buf *test_env_p = &test_env;
>
> test_env = *test_env_p;
> return 0;
> }
> [14:51:25 tmp]$ cc -Wall -c int_test.c
> [14:51:47 tmp]$ cc -Wall -c jmp_test.c
> jmp_test.c: In function `main':
> jmp_test.c:9: incompatible types in assignment
> [14:51:56 tmp]$ exit
> exit
>
> Script done on Mon Jul 2 14:52:02 2001
>
>
> What did I miss?
I don't think you are supposed to assign jmp_buf, only to use it with
setjmp() and longjmp(). It is not a basic compiler type -- it is a
system type defined by your libc and/or your OS and there are no
guarantees about its definition IIRC. Specifically glibc defines
jmpbuf as a 1-size array of a struct (so that calls to functions will
only pass a pointer, I suppose), and the "incompatible type" error is
what gcc gives you when you try to assign to an array.
If you really want to assign to an array or to any other type, you can
always wrap them with a struct, or simply memcpy(). But if you are trying
to do that you are probably doing something wrong. What are you after?
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: Linux-friendly CDR
On Wed, Jun 27, 2001 at 10:05:06PM +0300, Gal Goldschmidt wrote: > On Wed, 27 Jun 2001, Dan Kenigsberg wrote: > > > I would like to buy a Linux-friendly CDR. > > > > What would you recommend? > I can tell you what not to buy: HP USB. > If it's IDE or SCSI ( internal or external), it's probably > Linux friendly. > > With IDE you might need a bit of a kernel recompile for > SCSI-IDE. I wouldn't be so melodramatic. The ide-scsi.o is a standard module; it certainly exists in the standard installations of both Red Hat and Debian. An additional boot parameter to LILO would suffice. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: MS prohibits licensees using open source as development tools?
On Thu, Jun 28, 2001 at 11:26:42AM -0700, Efraim Yawitz wrote:
>
>
> On 27 Jun 2001, Oleg Goldshmidt wrote:
>
> > Matan Ziv-Av <[EMAIL PROTECTED]> writes:
> >
> > > Do we care about this?
> >
> > Yes, if you work in a mixed Open Source / Microsoft environment.
> >
> > > Does this have anything to do with linux?
> >
> > Yes. From now on you will have to read M$ and other licenses with
> > extra care to figure out whether or not you can use Linux (which is
> > specifically listed in the MIT EULA as a "tool" you cannot use)
> > *during development*, not just as a part of your final product.
> How can anyone not care about this? What is to prevent M$ from selling
>('licensing') software with a condition that requires you to buy a certain brand of
>car, read a certain newspaper, or vote for a certain political party? Does this kind
>of stuff really hold up in court? I hope this is getting a lot of publicity in the
>"non-technical" media.
They can bundle their SDK with a web browser, a car and a newspaper
subscription if they want to (antitrust laws notwithstanding), but it
is not legally possible for you or anyone to trade certain rights away,
such as your voting right and your first-born :) Such a contract won't
be kept by a court even if you sign it with blood.
By the way, I remember reading in the newspaper that a certain judge
ruled that there CAN be a contractual obligation to marriage, according
to present law, for which he criticized the lawmakers. I don't know what
ever came out of it, but if it holds, William Gates III can certainly
consider the continuation of his dynasty guaranteed.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: MS prohibits licensees using open source as development tools?
On Wed, Jun 27, 2001 at 02:56:36PM +0300, guy keren wrote: > > On Tue, 26 Jun 2001, Omer Zak wrote: > > > The situation was blown up, due to Microsoft's own FUD tactics.They > > claim that GPL is more "infectious" than it really is.They claim that if > > you edit and compile code using GPLed tools, your code must, too, be > > GPLed.This is untrue. > > more then this - they put 'GPL' and 'LGPL' in the same basket - there is > no legal problem in binding code that uses microsoft's code together with > LGPL-ed libraries. And the GPL has a special section EXPLICITLY disclaiming any copyright of the original software's copyright holders over its produced output. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: aol
On Mon, Jun 25, 2001 at 12:31:45PM +0200, mike ray wrote: > does any one have the ip # of AOL so i can block them in the firewall > > Michael W Ray > > IT Manager Mike, such a question does not belong on linux-il. You can be thankful to Ommer Mussaev who talked me out of giving 192.114.47.57 as an answer. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Linux Redhat installation and networking
On Mon, Jun 25, 2001 at 12:44:07PM +0200, Eran Levy wrote: > At 22:27 25/06/01 +0300, you wrote: > >On Mon, Jun 25, 2001 at 11:20:55AM +0200, Eran Levy wrote: > > > windows NT have NT-LOADER as I know. When you are installing windows NT > > > with his loader Linux will appear in the menu so every time you want to > > run > > > Linux you can just choose linux from the menu and press ENTER. OR you dont > > > have to install the NT loader and LILO will stay but you will have to add > > > to LILO a few options to make NT partition aviable in your LILO. > > > >Do you mean Microsoft actually went out of their way to ease Linux/NT > >dual-booting by detecting Linux and inserting it into the menu? Or > >simply that such a configuration is possible? > The NT loader is auto detecting your Linux partition, and to run Linux you > just have to choose Linux from the menu and then press ENTER. That's what I understood -- I wanted to make sure my eyes do not deceive me :) This changes my opinion of Microsoft quite a bit. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Linux Redhat installation and networking
On Mon, Jun 25, 2001 at 11:20:55AM +0200, Eran Levy wrote: > windows NT have NT-LOADER as I know. When you are installing windows NT > with his loader Linux will appear in the menu so every time you want to run > Linux you can just choose linux from the menu and press ENTER. OR you dont > have to install the NT loader and LILO will stay but you will have to add > to LILO a few options to make NT partition aviable in your LILO. Do you mean Microsoft actually went out of their way to ease Linux/NT dual-booting by detecting Linux and inserting it into the menu? Or simply that such a configuration is possible? By the way, I always found loadlin.exe very convenient in such situations -- check it out. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: compiling kde2 with gcc 3.0
On Wed, Jun 20, 2001 at 02:16:35PM +0300, mulix wrote: > On Wed, 20 Jun 2001, Adi Stav wrote: > > > I haven't tried, but if I understood correctly the frontend was hardly > > changed, while the C++ frontend was almost rewritten from scratch. Probably > > Gnome compilation will be affected less strongly than KDE. At work I tried > > to recompile some of our C libraries (works fine) and C++ librariess > > (tons of errors, fails). I don't know whether the C++ problems in KDE's case > > and in my own expriments' case were GCC's fauilt or that the code was wrong, > > though. > > much more likely that it's the code's fault. g++ (and libstdc++v3) > conform much better to the c++ standard, which means it will not accept > code it might've accepted in the past. > > on behalf of the gcc developers, if you think you found a bug in gcc, > reconsider. one of the critical release criteria of gcc was that there > will be _no_ regressions from gcc-2.95.2. code that compiles with > gcc-2.95.2 and no longer compiles with gcc-3.0 is probably not standards > conforming (enough). consult with someone who really knows the language. > try other (standards conforming!) compilers. only then, if you are sure > you found a bug, use the 'gccbug' shell script to report it. So I thought. Afterall, the C standard is a much more forgiving (ok ok, unsafe) than the C++ standard, plus there's less to conform to. I wonder how long it'd take for GCC 3.0 to be considered a safe C++ compiler to ship alone for stable distributions, though. > re compiling the kernel with gcc-3.0: you can do it, but it's not > recommended. even if the compilation succeeds (i dont think it does, at > the moment, some problem with builtin functions) there is a whole slew > of arcane magic and black wizardry that can go wrong, if the compiler > does something differently than what it used to do. you might get a > kernel that runs, but silently trashes your data. in other words, try > it, but if it breaks, you get to keep both pieces... Considering that the GCC 3.0 i386 backend was rewritten, I am genuinely surprised to hear that it compiled at all. Again, I'm wondering how long it'd take for GCC 3.0 to be considered safe for the kernel. I'll place my conservative bet of 18 months, at least. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: compiling kde2 with gcc 3.0
On Wed, Jun 20, 2001 at 11:00:50AM +0300, Hetz Ben Hamo wrote: > Hi, > > Don't try to compile KDE on GCC 3.0 - it will give you tons of error and many > applications will not work... > > As soon as KDE 2.2 will be out (probably in a month or so) - the cvs tree > will be "unfrozen" and people would be able to commit code. There are several > peopel already who are working on making KDE work with GCC 3.0 > > How is it on the Gnome side? does it compile and run perfectly? I haven't tried, but if I understood correctly the frontend was hardly changed, while the C++ frontend was almost rewritten from scratch. Probably Gnome compilation will be affected less strongly than KDE. At work I tried to recompile some of our C libraries (works fine) and C++ librariess (tons of errors, fails). I don't know whether the C++ problems in KDE's case and in my own expriments' case were GCC's fauilt or that the code was wrong, though. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: remote graphical usage
On Wed, Jun 13, 2001 at 10:59:42AM +0300, Erez Doron wrote: > Hi > > I am looking of a way to have a local display/keyboard, which is > connected on a slow channel ( i.e. modem ) to unix servers. > > I need a way to run X applictions on the remote server, and see them as > they were local. > > I tried using X - was too slow > I tried to use X protocol compression called 'dxpc' - still too slow > I tried to use Xvnc - better but still not fast enough. > > anyone knows of a way to do so except using Citrix or X-win32-server > over microsoft's rdesktop ? >From what I've seen a big hit on networked X performance is sophisticated toolkits. GTK+ applications for example take ages to load and to run, while old Athena programs are (relatively) quick. I suspect that this might have to do with applications loading images onto the X server, and that playing around with GTK themes and engines might help, but I've never verified this. So I'd suggest trying using simpler X applications if this is an option. By the way, why do you expect that using a different X server (X-win32-server) would make it faster? I am not familiar with this server. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Intresting Strange Address - Not Linux Q
On Tue, Jun 05, 2001 at 02:17:26PM +0300, Haim Gelfenbeyn wrote: > As you know IP address in just 32-bit number. Many browsers (including > NS and IE4+ as far as I know) support both the regilar > "number.dot.number..." form and specifying IP as-is (e.g. as 32-bit > decimal number). To convert from 32-bit numeric form to the regular > one simply convert the number to hex and then convert all 4 bytes to > decimal separately. > > Note: using such form won't really hide the IP address, it will only > make it harder to read. Also, some proxy servers (particularely squid) > don't support it. And for a good reason. Such nasties are often used by spammers to know which of the people they sent spam to actually click spam links, so that they can put them on their "extra preferred customers" list. Sorry. Well, you can always change your email address... Or ask Nadav for his uberscripts :) = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Syntax highlighting in Emacs
On Sun, May 27, 2001 at 03:19:21AM +0300, [EMAIL PROTECTED] wrote: > > Howdy all. > > A surface search of the "FE", as you refer to it brought up nothing, so > I'm asking you people, how do I enable syntax highlighting in *console* > Emacs? (X version is trivial, it's enabled by default..) There's no color in GNU Emacs 20 on the console. But there will be such a feature in GNU Emacs 21. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Mandrake 2.4.4 kernel
On Sun, May 20, 2001 at 08:15:21PM +0300, Alex Shnitman wrote: > > > > 1. Much faster, thanks to a major optimization of the way fork-exec is > > >implemented. > > > > IIRC, that feature's effects on the scheduler were not considered properly, > > and you could get all kinds of undesired scheduling behavior under 2.4.3, > > such as unresponsive processes and the like, and Linus reverted it in 2.4.4. > > Actually 2.4.4 introduced this change, and it's been reverted in > 2.4.5-pre1. I stand corrected. > I really don't understand how come it created problems though. Can > anyone explain it? Why would user-level processes care about > scheduling intricacies in the kernel? How is fork implemented in other > Unix variants -- is it done the slow way in all of them? The change was for the parent to give its child all of what remained of its timeslice immediately after creating it. This way, when the scheduler gets called as the syscall exits, it figures out that the child ought to run before the parent gets the fork() return value. The problem is that although no userspace should make assumptions about scheduling, the scheduler itself is really a delicate system and has to make lots of its own assumptions about userspace behaviour and about its own. If you have a lot of forks running, you've got a lot of worth of timeslices being handed around and concentrated in the hands of children, and that is a skew wasn't considered in the scheduler's original design. IIRC neither FreeBSD nor OpenBSD do these child gifts. It's not necessarily "the slow way", it just has advanteges in fork/exec situations. Anyhow, from the reactions to the idea I guess it may be pretty novel in Unix in general -- otherwise someone would HAVE to mention that FooOS already does this with considerable performance advanteges, and Linus would respond that FooOS is a piece of crap he wouldn't touch with a ten foot pole, and someone else would point out that the Foo approach is much more scalable, which is why it's preferred to Linux in all the serious companies HE knows, and we'd end up with the usual flamewar, not unlike the ones kuro5hin.org gets every time someone mentions that some story was already posted on Slashdot six weeks ago. In any case, KT says one proposed solution was to give the child only two thirds of the parnet's remaining timeslice, or something. I wonder why they don't simply force a context switch to the child just after it's created instead of messing around with priorities... Any kernel hackers who care to comment? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Mandrake 2.4.4 kernel
On Fri, May 18, 2001 at 10:19:49AM +0300, Eli Marmor wrote: > As a service for the confused reader, who may wonder what's the big deal, > after all it was not a long time ago when Mandrake released their 8.0 and > Linus released 2.4.3 (the version included with 8.0), let me add the > following notes, about the large improvements of the latest AC of 2.4.4 > over 2.4.3: > > 1. Much faster, thanks to a major optimization of the way fork-exec is >implemented. IIRC, that feature's effects on the scheduler were not considered properly, and you could get all kinds of undesired scheduling behavior under 2.4.3, such as unresponsive processes and the like, and Linus reverted it in 2.4.4. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: linux in the goverment
On Sun, May 13, 2001 at 10:26:51PM +0300, Yotam Rubin wrote: > Just out of curiosity, do you know of a single language that is actually bidi? > As far as I know, bi-directional languages do not exist. > All countries are "ready" for Linux, most simply don't know it yet. Modern hebrew IS bidirectional. You don't write "In 2001 I sent this email", or "this email was sent using SMTP" without bidi. Theoretically it is purely right to left, but in practice important parts of our Hebrew communication is written left to right. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Bidi support for Linux
On Wed, May 09, 2001 at 02:45:38PM +0300, Ilya Konstantinov wrote: > On Wed, May 09, 2001 at 12:37:08PM +0300, Nadav Har'El wrote: > > But note that modern (read Microsoft-influenced) keyboards also have other > > useless keys which we could "confiscate" for Hebrew use, such as the right > > Alt key (which is commonly used for multiple language support), the "Windows" > > key and the "Menu" key. > > I'd prefer the "Menu" key to actually rise the context menu. KDE2 even > offers it as a default shortcut, though for some reason it doesn't > work. The "Windows" key is really pretty useless though. One of the keys I (and others I know) use the most, to change consoles. And it would be silly to have a different set of bidi keys for console and for X. Ctrl-Shift by itself is free anywhere AFAIK. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Bidi support for Linux
On Wed, May 09, 2001 at 02:39:34PM +0300, Nadav Har'El wrote: > On Wed, May 09, 2001, Oleg Goldshmidt wrote about "Re: Bidi support for Linux": > > "Nadav Har'El" <[EMAIL PROTECTED]> writes: > > > Some time ago I was also using the Caps-Lock (which I find a totally useless > > > key ever since I stopped programming in Fortran ;), > > [this is really off-topic, but...] > > Fortran is case-insensitive, but its users aren't - many of them are used to > seeing all code in uppercase, and consider lowercase an abomination. I've > seen several Fortran programmers use the Caps-Lock key to type whole sections > of code in uppercase. I usually rebelled and entered stuff in lower-case - it > was also an interesting way to see which part of the code I wrote, and which > part was written by other people :) > > Anyway, my point is that I don't know *anybody*, except Fortran programmers, > who ever use the Caps-Lock key. What good is it? FOR SHOUTING IN EMAILS? > It only gets in the way of my left pinkie, once in a while a while causing > me to think I'm insane and don't remember my password. I vote for the > abolition of the Caps-Lock key. All in favor, type > > xmodmap -e 'clear lock' Nay. I really, really need my Caps-Lock, because I use it for a Control key. As an Emacs user, this mapping has probably saved me from torn hand tendons more often than I'd like to think about it, and it is also much more convenient with vi. Shift-Numlock is my current Caps-Lock key - you can take this one :) = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Bidi support for Linux
On Wed, May 09, 2001 at 02:22:19PM +0300, Shai Berger wrote: > Tzafrir Cohen wrote: > > > > On Wed, 9 May 2001, Ilya Konstantinov wrote: > > > > > > > > Using Scroll-Lock or Caps-Lock is interesting from the practical point > > > of view, but from the human interfaces design point of view, it > > > obfuscates the meaning of keys whose meaning is originally clear, > > > > Clear = you got used to it. > > > > I second that, Personally, I find ctrl-shift on both sides easy to use and remember, regardless of what I'm used to, and very clear that they imply certain direction. But maybe that's because I played way too much computer pinball as a child :) ..lost of good arguments snipped... > This is the important point: Using modifiers as keys is bad for you. > AFAIK, you cannot set keyboard shortcuts to ctrl-shift in KDE nor GNOME, > because for both of them these are not key presses. It would be better > to have no default bindings, with a Win3.1-like language-icon system, > than to try to force ctrl-shift to be a legitimate key. Why not keep ctrl-shift-key passing the event on to whoever was looking for it, but keep ctrl-shift alone as a direction changer? I think that's what Windows does, too. I feel that direction change is a very Meta thing, Meta enough to justify this. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: trace TCP session
On Wed, May 09, 2001 at 12:37:50PM +0300, Alon Altman wrote: > Hi, > Is there any way to be able to easily trace the data transferred in a TCP > session that was collected via tcpdump or strace? > Is there an existing tool to split tcpdump data into TCP connections and > create a user-readable output of the connection flow? > > I want to use this to check what do some closed-source apps are doing... Ethereal has this feature, at least in its GUI - it also has CLI interface, but I don't know if the CLI is feature-complete. If you only need to use it on a few TCP connections, the GUI will suffice. strace/truss might also prove useful, if it is a Unix program. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: no linux GUI in 97?
On Mon, Apr 23, 2001 at 02:47:02PM +0300, Yotam Rubin wrote: > Most revered sir. > > That does not imply cluelessness, it's a highly inflammatory statement. > X window is not a GUI. Besides, what's a paradigm for a GUI? I find the > Windows GUI horribly illogical and user-hostile. > KDE and GNOME are well documented, unlike their Windows counterpart, which aims > at obscuring very important details. Microsoft has simply convinced the general > public that it has created a "user-friendly" OS. He does not deserve to be > flamed, we should simply make him aware of his error via conventional means. > > Regards, > > On Mon, Apr 23, 2001 at 11:50:22AM +0200, Alon Barzilai wrote: > > not clueless? > > look what he wrote me this morning: > > > > "I did not write that Linux started in 1997. As for GUI, it still > > doesn't > > have one in the conservative sense. The GUIs offered are more like > > elaborated skins than real GUIs." > > > > X windows is elaborated skins? Nah. From the authors's statement it looks that he is quite clueful, but is strongly and unconventionally opinionated (or at least, quoting the opinion of someone else who is strongly and unconventionally opinionated). If he were truely clueless he'd say something like "I wouldn't call it a GUI while you still have to use the command line for most things". His original statemet is too different from the usual media conception of Linux for me to believe that the difference between his opinion and ours results from lack of knowledge. It implies that he knows Gnome and KDE have many flashy themes, for example. But then again, I have never spoken to him... I wonder what he means by "GUI in the conservative sense". = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Redhat Investor-relations run by Microsoft?
On Sun, Apr 22, 2001 at 10:02:30PM +0300, Yotam Rubin wrote: > Each OSS based company defines its own "Open Source business model" > Redhat's business model cannot be compared with Valinux's because > they focus on different market shares. Please note that I merely gave > an example: I tried to show that even the OSS based company with the highest > potential did not obtain sufficient revenue. Companies which produce Linux > have lower revenue potential I don't understand what you are trying to say: If they can't be compared, why should VA's success (or lack of, or whatever) have anything to do with Red Hat's? The way I see it, the only thing in common between Red Hat and VA is that they both relied on GNU/Linux success and acceptance -- and that, of all things, they both guessed right. Other than that, Red Hat broke even last quarter while VA still has (I think) long to go... = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Redhat Investor-relations run by Microsoft?
I wonder if we could even come up with an agreed upon definition of "open source business model". I suspect it is no less bogus than "Internet business model". People and companies make money the way they always have, which either is by offering products and services to others who wish to pay for them, or by taking it by force in some way. This is not going to change. Opensource/Free software might be involved to an extent, or not at all. I can't, in any way, see how Red Hat's business model can be even COMPARED with, for example, VA Linux's original business model, much less referred to with the same term. VA Linux's model was selling computer hardware! This is the farthest from unproven as you can get in the IT industry. Way too proven, in fact, because it forces competition with several players with as many as 50 years of experience. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Redhat Investor-relations run by Microsoft?
On Sun, Apr 22, 2001 at 04:26:59PM +0300, Nadav Har'El wrote: > I was looking at Redhat's annual report to the SEC (April 19. See > http://biz.yahoo.com/e/010419/rhat.html). The "Factors Affecting Future > Results" section is amazing, and full of depressing statements (actually, > the entire report is full of depressing statements). It might be literally > true, but it looks like something taken directly from Microsoft propaganda > (like the Halloween documents). > Here are a few select paragraphs (edited slightly), for your enjoyment :) ..snip... I don't know. Aren't Red Hat, as a publicly-traded company, required BY LAW to provide in their SEC reports such warnings about weaknesses? By the way: I actually considered the Halloween Documents pretty good Linux propaganda at the time... = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: FTP hangs, taking 100% CPU
On Mon, Mar 12, 2001 at 11:29:57PM +0200, Aviram Jenik wrote: > There's a problem that's been driving us crazy for a while here. > > One of our users connects using explorer as an FTP client. I'm not sure > whether or not Explorer does an explicit disconnect when the window is > closed, but in any case, after the FTP session is done, the process in.ftpd > remains 'alive' until it's killed manually. > It can even continue to 'run' for days, even though the FTP client is long > gone. What's worse, it takes 100% CPU, and if a second connection is > initiated, a second process remains with an additional 100% CPU (and this > really takes the fun out of our 2 CPU machine). > > The whole thing is very strange, since there's no problem with any FTP > clients besides this one, and there's nothing special about our FTP daemon > (wu-ftpd-2.6.0). > Does anybody have a clue about what might be causing this? First of all Wu-FTPD is so full of security holes it's not even funny. Now, what does an strace of these processes show? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: which distro for newbie?
On Sat, Mar 03, 2001 at 04:02:18PM +0200, Christoph Bugel wrote: > A collegue at work (a very good C++ programmer, by the way) asked me how to > install Linux, and where to get it.. I find this a hard question to answer. > Personally I have been using slackware for some years now, and I have no need > to try any other distro. But I don't know if this is what I should advice to > other people who explicitly ask for something 'windows-like' and 'easy to > install'. > > I have suse 6.3 cdroms lying around, but things move so fast here that this > seems like way too old. I should probably advise him to try redhat, but I never > installed that. Does anyone know how redhat can be installed from the web (I > don't have a cdrom). Can I download an iso image and burn it? Are there up to > date images on linux.org.il? Is it easy to install redhat from the web or did > they make it extra hard, trying to get people to buy a cd? And should I use the > latest version or not? By the way, has anyone tried installing Progeny? Would you recommend it for beginners? = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Hebrew Keyboard
On Thu, Mar 01, 2001 at 09:50:43PM +0200, Oleg Goldshmidt wrote: > Ely Levy <[EMAIL PROTECTED]> writes: > > > On 1 Mar 2001, Oleg Goldshmidt wrote: > > | Yemeni, Saudi, Iraqi, Pakistani. Are we going to make geographical > > | or ethnic speculations? > > | > > that the hebrew names? > > in english pakistanian iraqian . > > Yes, it's Saudi, Iraqi, etc in English. Note that all of these contries speak Semitic languages. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: root shell changed.
On Sun, Feb 25, 2001 at 05:35:11PM +0200, Sagi Bashari wrote: > boot in single user mode (ie. linux 1), edit your /etc/passwd file, and > change it. But you can't login without a valid shell, not even in single user mode... = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: root shell changed.
On Sun, Feb 25, 2001 at 05:18:12PM +0200, Maxim Kryachko wrote: > Hi list. > Some guy here changed by mistake the shell of root user to something > nonexistent. > Now every time he tries to su to root, he is trown out with the massage "no > shell" > Is there a way to correct the problem? Check out "su -s". I haven't tried it at home (who needs trouble?). When all else fails, I usually reboot and specify "linux init=/bin/sh" at the lilo prompt. Does wonders on all kinds of catch 22 type goof ups. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SUSE 7.1
On Sun, Feb 25, 2001 at 08:08:27AM +0200, Stanislav Malyshev a.k.a Frodo wrote: > AS>> Your opinions are yours, of course, but I think you are > AS>> misrepresenting RMS here. He has always objected to limiting the use > AS>> of software. Consider his recent review of the new APSL (Apple Public > > Well, so he's a hypocrite here. Because when it is regarding his software > (specifically readline) he is actively objecting making programs that are > interoperable with (link with) this library but not GPL - thus limiting > use, not only distribution. I already quoted RMS on that on the list. That's "work based on the program", NOT use. Were there a limit on use, you wouldn't have been able to use readline in a proprietary in-house project, for example. > Also, if you had read discussion on /. some couple of months ago on GPL > 3, RMS said specifically that he wants to plug "appserver loophole". Means > limiting use again. You haven't yet seen the GPLv3, but you have already decided that it will limit running programs and that RMS is a therefore a hyprocrite. > Actually, nowdays it's pretty much impossible to distinguish between > separately distributing two parts of the application and making two > applications interoperable. New technology is giving the legal system new challenges every day. That is not new, and it doesn't mean that copyright laws are no longer relevant to technology, as much as I wish they weren't. > Term "derived work" is not defined and never > was, and yet less there's any hope it will be. What makes you think "derived work" (or "work based on the program", as the GPL refers to it) is undefined? I am not lawyer, but neither are you, and everybody else seems to think otherwise. > A good lawyer probably will > tear in shreds any "derived work" claim but simply copying the code, and, > alternatively, make two pieces of software that came once in touch > "derived work" of each other - depending on who pays him. Using such words > in a (wannabe) legal document is very unfortunate move. It held fine for the past 10 years. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SUSE 7.1
On Sat, Feb 24, 2001 at 02:45:37PM +0200, Stanislav Malyshev a.k.a Frodo wrote: > OH>> I heard from some people that the yast license is pretty 'sly' and some > OH>> bad things about it. I prefer GPL not because I want to see its source or > OH>> modify it (Although maybe it could be nice), but because I want to be free > OH>> to use it. > > GPL has nothing to do with usage, and can not, because copyright law, on > which GPL bases, controls distribution, not use. Nothing in the copyright > law can prevent you to do anything with the text of the book, but it can > prevent you from redistributing the text. > BTW, RMS is very eager to change this, and make GPL control the use too Your opinions are yours, of course, but I think you are misrepresenting RMS here. He has always objected to limiting the use of software. Consider his recent review of the new APSL (Apple Public Source License) at http://www.fsf.org/philosophy/apsl.html : At a fundamental level, the APSL makes a claim that, if it became accepted, would stretch copyright powers in a dangerous way: it claims to be able to set conditions for simply *running* the software. As I understand it, copyright law in the US does not permit this, except when encryption or a license manager is used to enforce the conditions. It would be terribly ironic if a failed attempt at making a free software license resulted in an extension of the effective range of copyright power. > - to prevent "loopholes" such as separating GPLed code inside > proprietary application and running it as independant module, without > redistributing it or even including it in the proprietary product > distribution. However nice it seems for "free software", it's very very > dangerous move. But this is another topic... > > OH>> Sure. There are more nice licenses such as the BSD license.. But > OH>> most of the other licenses are not as free as GPL is, so I > > GPL is much less free that BSD. In fact, GPL is the most restrictive open > source license known to me - it not only mandates you to opensource all > your changes or additions to the code, it specifically prescribes under > what license you should do it. And, as I said, RMS wants to make it even > more restrictive. That's "beneficial for specific group of software > developers" - to which you may or may not to belong, but not "free". > > OH>> Besides, you talk like the whole open source community is > OH>> sponsored by these companies. I know very few open source > OH>> projects which are sponsored. > > Oh, well, so you know not much. That's nice you realize it. Next step is > to try and know more. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SUSE 7.1
On Sat, Feb 24, 2001 at 12:43:23AM +0200, Noam Meltzer wrote: > It's wrong because that's not the idea of linux, it ment to be a free > software. > I usually get confused here, but whatever license it is on (the kernel), > GPL or GNU (I think GPL) says that u have to put it free of charge for > d/l or something like that. Hope i didn't talk bullshit now, but > still... "Free Software" in the sense of "tochna hofshit", not "tochna hinam". Not only you are allowed and encouraged to sell it, but you cannot demand people you give the software NOT to sell it. See http://gnu.org/philosophy/selling.html for more details. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: SUSE 7.1
On Sat, Feb 24, 2001 at 03:55:21AM +0200, Oren Held wrote: > Hello Hetz > > The wrong thing is that they are getting money for programs that > volunteers did _FOR FREE_, on their own time. What is wrong with it? If the volunteers did not want their software to be used and profited from, they would not have released their code as free software. The way I see it, and implicitly from their license, the way these volunteers are seeing it, is that SuSE is actually helping them in getting their software used more. > instead of giving the money > for the volunteers I'm giving it to the company who packed this and made > some nice setup program ? Then why won't you send the volunteers some money, too? Anyhow, a large part of the CD money does go to the projects, through sponsorships. SuSE is one of the larger sponsors of free software projects -- off the top of my head, Andreas Archangelli (the Linux VM hacker), and XFree86 (they sponsored the first Rage128 driver, which I've used for quite a while, even though I didn't pay SuSE a cent). > no thanks. they should release the d/l version > while selling it, imo. You can buy the CD and place its contents on a web site if you like. > That's why I prefer to use a non-commercial distro (debian). Which is also being sold, for various prices. Debian is mostly distributed by commercial companies for completely commercial reasons. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Programming problem
On Sun, Feb 04, 2001 at 09:44:43AM -0500, jteich wrote: > I wanted to run a shell script program that needs to run with root permissions, and >I wanted to > be able to start this script without su'ing. I tried to set the suid bit on the >program, but I > found out that the suid bit only works for binary executables. Then I wrote a simple >c program > that starts up the shell program through the system() call, and I chown the program >to root and > set the suid bit on it, but for some reason the shell still wouldn't run as root. >What am I > doing wrong and is there any way aroud this? This is not very secure, because there are many tricks that a user can play on a suid shell. A better approach would be to see which commands in the script need to run suid and only privelege these commands to relevant users (by setting their suid bit or by writing a wrapper like you did). - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Debian mirror in Palestine
On Fri, Jan 26, 2001 at 10:25:03AM +0200, Ben-Nes Michael wrote: > I don't lugh at them, but I joined .il mailing list and not pl, sk , ru or what > ever. > and il = mean Israel. > > If there are Palestinian users they should open their own list or behave by the > moral code which are held in israel. > > Marc can open a new pglu to our neighbors, and stop upsetting people. > > any why currently im recycling what other have said. Marc was offering you in his mail the use of his local Debian mirror. You don't have to use it if you don't want, but attacking him for this mail seems to me like an inappropriate response. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Debian mirror in Palestine
On Thu, Jan 25, 2001 at 05:17:27PM +0200, Gilad Ben-Yossef wrote: > > > Ben-Nes Michael wrote: > > > This mirror will be a big use for Gaza and Jericho :P > > Don't laugh too hard. I have a nagging suspicion (but no real proofs) > that the residents of the palestine authority and their ilks make much > more use of Free Software, for obvious economical reasons, from the > a typical Israeli user. As much as the "free software helps the Third World" posts and stories on kuro5hin warm my heart, I never thought they had much base in reality. I think that when most poor people find MS-Windows too expensive, they simply share it, caring about neither ideology nor copyright law :) - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: advanced linux desktop book
On Sun, Jan 21, 2001 at 05:02:14PM +0200, [EMAIL PROTECTED] wrote: > Think of a microsoft oriented person, who used to work with MS desktops and > servers, and decides to learn linux. > He picks up a book about using linux as a server (web, ftp, mail, > firewall,etc.) which teaches him the basic work > with linux and how to specifically configure the servers. > Now he wants to use linux as his desktop, but he runs into many problems, > which are not always technical. > something like that? Then I think you will find a lot of information in the LDP, and in many Linux books too. It's just that the seperation between server and desktop is not that clearly defined. You can just skip the irrelevant chapters, like you would anyhow :) If you are looking for a book that teaches only working with a desktop machine's graphical environment and application, that's probably a different story. But you said you are looking for more advanced information so I suppose this isn't it. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: advanced linux desktop book
On Sun, Jan 21, 2001 at 02:10:36PM +0200, [EMAIL PROTECTED] wrote: > shalom lekolam, > > does anyone know a book for the intermediate and advanced user on how > to use the linux system as your desktop? All I could find were books who > teach > you how to do ls, and install linux, nothing more serious... (on the desktop > side) What exactly are you looking for that is desktop-specific? I can't think of many "advanced" functions that are only relevant to desktop machines but not to servers, unless you are looking specifically for Quake, video editing and the like. With both servers and desktops you need to know how manage a filesystem, configure networking and the such -- they are mostly the same for servers and desktops. So I suppose the guides and the HOWTOs from the LDP at http://www.linuxdoc.org could be very useful. Some companies put them in print from time to time. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: distribution comparison
On Thu, Jan 11, 2001 at 03:44:38PM +0200, [EMAIL PROTECTED] wrote: > hey all, > > is there a ready made comparison table somewere, or let's just make one of > our own, > to compare difference between linux distributions? (like mandrake is the > easiest to use, red hat is the easiest to install, etc...) > or how about you just throw in ideas? I think it would be hard to make one because you cannot really see all the advantages and disadvantages of a distribution until you have been using it for a while. Most of the distribution reviews are made by technical reporters who can only afford to get a very rough idea of what the distribution is all about -- often only the installation process and a quick pass over the filesystem layout. I like your idea of comparing different issues in a distribution rather than looking for a single "best one", because I see distibutions as providing several totally sepearate services. Technical things I'm looking for in a distribution are: 1. Good installation. 2. Automatic hardware configuration. 3. Upgrade and manage the installation easily. 4. Integration: the packages should share services and configuration options among them. For example, when you install a Debian package it automatically gets added to all the menus of all the desktop environments and window managers (I think Mandrake is using this system too?). Or, installing a package should make sure it already has a working /etc/init.d script, etc. 5. Quality: the packages should be installed with good defaults and work the way a very good system administrator would have recommended them to. 6. Scope: how much software (and which software, depending on your needs) does the distribution support? 7. Configurability: how easy it is to continue using the distribution's services even when you want to change the packages' default configuration. I don't really think "ease of use" is an issue for distributions. Usually it's the desktop environments' job to provide easy tools, and you can find all of these tools in several distributions, regardless of which distribution they originated in. Now I've only worked with Red Hat and Debian (and it's been several years since I last touched Slackware). I'd say that Debian (2.2) truelly excells in upgradability, managebality, integration, configurability, quality and scope, does ok or less with installation but totally fails hardware. Red Hat (6.2) by my experience FWIW does very badly in upgradeability and quality, but has an okay scope, hardware, configurability and integration, and has an excellent installer. > also a comparison between kde and gnome would be nice.. why use this or the > other? I don't think that would say much. You don't really have to choose between running Gnome or KDE, you can run whichever components of the two you prefer whenever you need them. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: RH7
On Wed, Jan 10, 2001 at 07:16:49PM +0200, mike ray wrote:
> where is the inetd.conf in RedHat 7 ???
It is a well-followed standard for inetd.conf to be at /etc, but then
again I've never with RedHat 7. But I do remember that RedHat 6.2,
under default installation (Gnome workstation) does not install inetd
at all. So if you cannot find the file, check out whether all the RPMs
you need are actually installed.
- Adi Stav
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: hebrew filenames..
On Thu, Jan 04, 2001 at 10:41:06AM +0200, Tal Amir wrote: > hi, > > mounting as vfat is not relevant since i can access other hebrew files > that are not inside the forser in question.that also rules out the > codepage option. i'm beginging to think that maybe that spesiphic block on > the disk is bad. > > as far as dosfsck, it shows no problems. > this is wierd! my mandrake just hates those songs, i guess... ;) Only one thing left to do: post a story to Slashdot and blame the RIAA :) - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: KDE 2
On Wed, Jan 03, 2001 at 10:45:24PM +0200, David Hananel wrote: > And then, After I try to install the libs I get a message telling that 2 > files are missing: > libssl.so.0 and libcrypto.so. > > So I installed OPENSSL, but still these 2 files are missing... On Thu, Jan 04, 2001 at 12:34:02AM +0200, Hetz Ben Hamo wrote: > Hi, > > This might be a shameless plug - but I would suggest to you to install our > free product - the Aduva Manager - and install from there KDE 2.0 - > http://www.aduva.com > > The libs you're talking about are openSSL indeed - but it seems your .so > files are not in your /etc/ld.so.conf config files.. I think (correct me if I'm wrong) that RPM finds its per-file dependencies directly on the filesystem, regardless of /etc/ld.so.conf? If this is the case, then the problem is most likely that David's RPM installed its files in some unreasonable place such as /usr/local, or maybe that the RPM did not install correctly. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: hebrew filenames..
On Wed, Jan 03, 2001 at 09:20:15PM +0200, Tal Amir wrote: > hi list, > > i have a folder containing mp3, sitting in my linux box and shared to the > rest of the lan via samba. > this folder contains a bus foler with hebrew files (that i was foolish > enough to name in hebrew fonts.) > > the situation is that linux shows the files as empty directory's (even if > it is an mp3 file) and cannot access, rename or delete it in any way. > > i cannot access those files from windows or linux in any way at all. > is there any way to restore the files, or even to delete them ? > because when an application from windows tried to read the directory in > question, it gets a fatal error and crushes. (how NOT ms tipical..) ;) > > idea's ? Just a check: you did remember to mount the filesystem as vfat rather than msdos, right? Also, try msdos anyhow. You might not see the pretty names, but at least you'll be able to access the files. Also, run dosfsck (or scandisk) on your partition; the problem might be unrelated to names. And lastly, try to see if you have the Microsoft filesystem Hebrew codepage compiled into your kernel (or as a module). - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: bootdisk
On Wed, Jan 03, 2001 at 05:29:49PM +0200, Erez Doron wrote: > hi > > I've installed a new computer with raid and lvm-root > > I'm trying to create a backup boot methos ( i.e. boot via floppy/ies or > cd ) > > my root ( compressed ) and the kernel image together are about > 1700kbytes > so i can not make it a single-boot diskete > ( funny, but if i compile my utils as static, so i do not need libc.so , > it takes less > disk space befor compression, but more after ... ) > > i tried making it 1 root + 1 boot but: > > if i write root=/dev/vg0/root ( lvm root ) it will not prmbpt/load the > ramdisk ... > if i write root=/dev/fd0 it will not know that it needs to load > /dev/vg0/root as root > ( either write it under lilo, or use rdev ) > > i tried making a single 2880 image and boot via cd. but because i do not > really have > a 2880 floppy, i made it as ram disk but lilo does not want to install > itself on ramdisk. > and working with rdev, gave me the same problems as above > > any idea ? Several things you can do: 1. There are many, many bootdisks projects on Freshmeat. Take one, adjust to your needs, post the result on Freshmeat, rince, repeat :) 2. Check out "busybox" to save space on a boot disk (Debian use it on their installation and rescue disks). It combines simplified versions many necessary utilities into one, saving a lot of space at the price of convenience. 3. Consider using initrd. ObRTFM: Check out the Linux documentation for kernel parameters in the Linux source documentation directory. I found it very helpful. Especially the ramdisk and initrd parameters. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Ride from Jerusalem at 10:00?
Is there leaving Jerusalem late, at around 10:00? - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: error
On Wed, Jan 03, 2001 at 10:43:34AM +0200, Tal Amir wrote: > hi, > > when trying to create a group called "all" for all of the mail accounts on > my server,(using linuxconf) i get : > > > > Executing: /usr/sbin/usermod -l all -c 'all mail accounts' -d > /home/all -s '/bin/false' -G '' -u 511 -g 231 > Executing: /usr/bin/chage -m -1 -M 9 -W -1 all > Archiving /etc/quota.conf, version Home-Office > Archiving /etc/aliases, version Home-Office > Segmentation fault > > > > any idea anyone ? Come on, it's a segfault. A crash. A bug. Something That Is Not Supposed To Happen. What is there to have an idea about? If at least you had written what distribution you are running, and what version, and what version is Linuxconf, I could at least have searched some bug databases to see if it's a known bug -- not that I would, because you can do that just as easily yourself. Update your Linuxconf or use something else. And file a bug report with your distribution, if one has not already been filed. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Some more console questions.
On Tue, Jan 02, 2001 at 08:07:09PM +0200, Oded Arbel wrote: > Hi list > > I botched my lilo config and forgot to setup vga=788 when I upgraded my > kernel. now I'm stuck in 25x80, and I really don't want to reboot again. I > tried probing the man pages for a way to change my console to the extended > vga mode, but to no avail. can you give me some pointers please ? IIRC, the kernel changes text modes using the BIOS VESA calls, which can only be read in protected mode (read: during boot). So I think you can't. The situation is probably different if you are using frame buffer. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Boot
On Sun, Dec 31, 2000 at 06:43:40PM +, Manor G. wrote: > Hi, > Boot using your bootdisk, login as root and reinstall lilo, > you can do that by using the command 'lilo' > if you get an error, or if you feel you need reconfiguration, > i suggest using `liloconfig`, which is easy to use, or, do it manually. > > this problem occurs sometimes when you don't reinstall lilo after you rebuilt or > upgraded your kernel, but thats not your case i guess. Also possible is the 1024 cylinder limit in old LILOs. If nothing else helps, try getting a new LILO that does not have that limit (sorry, I don't remember which version solved the problem). = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ipchains style
On Sun, Dec 31, 2000 at 10:01:07AM +0200, guy keren wrote: > btw, in the new kernel (2.4), where netfilter is used, there is a new > notion of tables. unlike usage of multiple chains, usage of multiple > tables does add extra functionality, in that it allows you to have one set > of rules perform a complete "computation", and then the packets coming out > of it are re-processed by the rules in a second table. i already so a case > where this allowed for more functionality then s possible using kernel > 2.2's chains. Hmm. How is that different from from creating custom chains in ipchains and sending packets from one chain to another? - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ipchains
On Sat, Dec 30, 2000 at 10:25:31PM +0200, Nadav Har'El wrote: > > I'd say that as soon as a company releases software, it doesn't matter > > whether the company's core business is hardware or not. The software > > is governed by the same laws. > > > > I'm not sure I understand: do you mean the GPL should or should not be > > applicable to hardware? Or firmware (there was a discussion on > > debian-devel about firmware not long ago, unfortunately without > > conclusive findings). > > I'm not talking about applying the GPL to the hardware itself, or about > hardware companies making software (a la Sun's Java) - I was talking > about how the GPL applies to software that comes prepackaged with some > hardware. > > I think that the computer-VCR-appliance makers I described above should be > allowed to put GPL software on their machines and sell them. And here is > my rationale: > > Consider a PC maker, say Compaq (just an example!), that sells computers > with Linux (say, a Redhat distribution) preinstalled in them (with a manual > that describes how to get the sources of the programs from the Internet). Is > this legal? I think it would be very hard to find somebody who thinks it isn't. > Now, what if our PC maker adds a proprietary piece of software to the > preinstalled system, e.g., some cool game that runs on Linux? Is this ok? > I have a very hard time to understand why not. The sources to that game are > not available, but the sources to all the GPLed programs are still available. > So, if you consider what our hypothetical PC maker did as being legal and > acceptable, then how different is this from the computer-VCR case, where > somebody is selling a PC with a bit of extra hardware (TV card, remote > control, etc.) and some selection of proprietary software (the software that > does the video recording and provides the user interface)? I don't think it contradicts the GPL, unless GPLed programs were modified with proprietary extensions in the process. Of course, I would much prefer free software on such Linux VCR. Proprietary software is not as well accepted as free software in the Linux world, and is considered "taking advantage without giving back". > Because I consider these two cases to be impossible to differentiate, and > because the first case (PC maker selling preinstalled linux-based machines) > is so acceptable and even desirable to the free-software community, I would > have to conclude that making a hardware package that includes a GPLed program > should be considered legal. And indeed is. But why not walk the extra mile? I think Tivo did. > By the way, what is RMS's thoughts about selling a CD containing some GPLed > programs (complete with full source code and GPL license) and some other > commercial programs (that don't have source code, and the license prohibit > copying them)? This is very similar to the hardware case above, and I would > consider such a distribution ok too, but maybe other people disagree with > me here. RMS has stated his opinion loud and clear in the GPL: "In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License." ;) > > And by the way: has anyone in Israel tried any of these machines? I > > suppose the schedule planning features cannot be used, because they > > are tried to a service that is only given in the US, but it still > > sounds cool enough. > > I don't have any personal experience with them, unfortunately. > > Watch out: if you buy such a machine in the U.S. (or Japan), then most likely > it will only work for NTSC... Before you buy one, check if they can also > record PAL. Ouch. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ipchains
On Sat, Dec 30, 2000 at 10:14:33PM +0200, Stanislav Malyshev a.k.a Frodo wrote: > AS>> That is necessary for copyleft. If you could take Linux and release it > > Sure. So be aware that any time you read "proprietary" in FSF texts, you > should read "non-GPL", since GPL restricts not only more strict licenses, > but also less strict. I understand why it's done, but let's be honest - > don't say "we don't allow proprietary software to take advantage", > omitting the part of "we also don't allow more free software to take > advantage, because of our political agenda". Not being honest here results > in unfair advertising, which leads people to troubles. That is not true. You CAN take GPLed code and combine it with code licensed under a less restrictive license. > AS>> But there is no virus, and there is no contamination: just remove that > AS>> person's code from the program, rewriting it if necessary. The rest of > AS>> the program is not affected by the removed code's license. It never > AS>> was, either. > > Yeah, but I though whole point of "free software" was sharing, right? > Getting the world better, maybe... And here we see typical EULA argument - > "you don't like it, you don't buy it". So we're back to square one, right? > No free software revolution anymore, just another kind of EULA? Sharing the code with users and other free software developers, not necessarily with proprietary software. You need to rewrite the code only if you want to place more restrictions on it than are in the GPL. > AS>> the FSF can release a more liberal version of the GPL and it will > AS>> automatically apply. And just to make sure, there is always dual > AS>> licensing (invented by Larry Wall, a classical TIMTOWDI approach :) > > Well, dual licensing code as GPL and BSD (or GPL and PD, for example) is a > real way to make GPL look stupid. I wonder why nobody tried it just to see > what it will do - just for curiosity. Or someone did and I missed that? I > wonder what would be FSF reaction on that. But you do understand that dual > licensing actually makes GPL void and RMSs fear of proprietizing GPLed > software come true? You take GPL code, wrap it into a package, license it > under GPL+PD, then take new package under PD (forgetting about GPL > part) and do whatever you like. GPL even hints you could do that, in > paragraph 5: > > ==quote== > 5. You are not required to accept this License, since you have not > signed it. However, nothing else grants you permission to modify or > distribute the Program or its derivative works. These actions are > prohibited by law if you do not accept this License. > ==end quote== > > So, if there was anything else that allowed you to distribute and modify > that program (like dual licensing under PD), you perfectly might do > anything you want, right? No, that would not work. You can license, dual-license or triple-license your own code any way you like, but another author's code stays with the license that he or she gave it. You can write a readline-using program dually licensed under BSD/GPL, but if you want to make it proprietary you will have to remove readline first. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ipchains
On Fri, Dec 29, 2000 at 08:44:52PM +0200, Stanislav Malyshev a.k.a Frodo wrote: > NH>> What kind of judge is going to make a decision against a company > NH>> when in a 100,000 line code, 50 lines "somehow distantly > NH>> resemble" code from a GPLed program? If the developer only looks > NH>> at the code, that's what going to happen - he won't suddenly > NH>> have 10,000 lines identical to a GPL program. If he does have > NH>> such 10,000 lines, it means he copied them, not just looked. > > There's such concept as "negative knowledge". Like, you look in other's > code and see comment "I'm doing 'foo' because we tried 'bar' and 'baz' and > this doesn't work". Voila - you gained knowledge that saved you thousands > workhours. Even if you don't copy any of the code, still this code > influenced your code, thus making it kind of derived work. This doesn't > mean that somebody can _successfully_ sue you on this, but chances that > you can be just sued on this basis are good enough so that most companies > don't want the trouble. That's only one example of concepts that can be > applied given there's a lawyer who wants to make money on it. The "Negatice Knowledge" precendent, to grossly understate my opinion here, was thought in malice, argued in lies, accepted in ignorance and maintained by extortion. And since it has such broad, overreaching effects, it can be used for almost anything, regardless of the licnese of the software. For example, BSD could sue because their ideas were used somewhere yet they were given no credits in advertisemenets like the copyright license demands. > Why I am telling this - because for a serious company even a possibility > of lawsuit is often frightening enough to avid the trouble of touching GPl > code. Given the publicity that each "GPL violation" case receives, you > better avoid this trouble even if you know you'd win the suit for sure. No GPL code or BSD code would ever sue over such a thing. If even they had, they would have much less resources than either of the two companies mentioned (I think they were Microsoft and Novell?). These companies are afraid because a) the precedence is ridiculous, and b) they expect others to do what they would in their shoes. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ipchains
On Fri, Dec 29, 2000 at 07:27:34PM +0200, Nadav Har'El wrote: > But the GPL causes the following sort of "comtamination": Take any of the > important pieces of GPL software on the Internet. Most, if not all, of them > have been written by more than one person. Some of them have been written > or update by dozens of people over the years. Each of these people are not > only forbidden from releasing the improved version under a more restrictive > license, but also from releasing it under a less restrictive one! That is necessary for copyleft. If you could take Linux and release it under a less restrictive license, you could also immediately relase it under a MORE restrictive license. Just release it under a license that does not enforce the "no more restrcitions" part, and it's not copylefted anymore. The idea of the GPL was to add the minimal set of restrictions that would not allow anyone to add further restrictions. > So any > program that once caught the GPL "virus" will stay a carrier for ever, or > until every one of the authors agree to change the license, which isn't likely > to ever happen in a program with many authors. What happens happens when > a program has 20 authors, and 19 agree to change the license but 1 doesn't, > and the small part that guy wrote is so strongly intertwined in the program > it can no longer be easily removed? (e.g., this guy added a new, trivial, > function to some API, and any attempt to redo his work will obviously end > up looking the same thing). Well, if someone does not want his or her code distributed in a certain way, and has not given you permission to do so, all you can do is respect his or her wishes... But there is no virus, and there is no contamination: just remove that person's code from the program, rewriting it if necessary. The rest of the program is not affected by the removed code's license. It never was, either. By the way, that is one of the major reasons the FSF accepts copyright transfers of software: so that it can relicense them under a more liberal license if the original author is unreachable. Also, that is why most people put their software under "GPL version 2 or later" -- the FSF can release a more liberal version of the GPL and it will automatically apply. And just to make sure, there is always dual licensing (invented by Larry Wall, a classical TIMTOWDI approach :) > > Also, nowhere does the GPL limit "interoperability". Only use code as > > part of another program. Whether or not you consider an executable and > > a dynamic library accompanying it two parts of the same program or not > > is a different issue. GPLv3 intends to clarify this and other points. > > There's another problematic issue about the GPL. It's quite clear how it > applies to software companies, but how does it apply to Hardware companies? > > For example, some companies are selling a VCR-like product that records on > a hard-disk and that allows cool features like playing back while recording > (you can skip commercials if you start watching the movie 20 minutes after > starting recording it!). Anyway, some of these products are actually small > modified PCs (with extra, special hardware), running Linux plus the company's > proprietary software. Now, how does the GPL apply here? > > If needed, the company can supply the product with a CD of the GNU/Linux > sources (but it's probably not needed because it is found on the Internet). > But do they have to also give away the sources of the proprietary VCR-interface > program? It doesn't look that way. I actually find this a good thing, because > the GPL was meant to stop the tyranny of the software companies, not of the > consumer-electronics companies. On the contrary: the consumer-electronic > companies are damaged by the software giants in the same way us invidual users > are, and should be allowed to benefit by the same solution. > Besides, all the "crap" about your rights to make copies doesn't really apply > to consumer electronics: you can't (yet) copy a physical machine. I'd say that as soon as a company releases software, it doesn't matter whether the company's core business is hardware or not. The software is governed by the same laws. I'm not sure I understand: do you mean the GPL should or should not be applicable to hardware? Or firmware (there was a discussion on debian-devel about firmware not long ago, unfortunately without conclusive findings). And by the way: has anyone in Israel tried any of these machines? I suppose the schedule planning features cannot be used, because they are tried to a service that is only given in the US, but it still sounds cool enough. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ipchains
On Fri, Dec 29, 2000 at 09:29:51PM +0200, Stanislav Malyshev a.k.a Frodo wrote: > AS>> Have you actually READ the GPL? It does not define "derived work" > AS>> anywhere, leaving that to copyright law. RMS has said as much, too. > > For GPL, RMS is the copyright law. Since if RMS thinks it's violating GPL, > you probably will very soon forced to GPL it or pull it, if only you don't > want to get a load of bad publicity and a costy lawsuit. Is your problem here the GPL's dry words? Or the FSF's interpretation of it? Or the fact that the FSF eforces the GPL? Every license has its terms, and its interpretation by those who apply the license, and is often enforced. I doubt very much that if the FSF's interpretation were legally unreasonable it would have still been protected after 26 years. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ipchains
7;s not > many organisations that can afford being at war with FSF now) > _distribution of source_, which doesn't have a line of GPLed code. I > understand why he does this - because otherwise GPL and LGPL would be the > same for libraries, and he's actively against it - but the fact that RMS > doesn't like it shouldn't automatically make it illegal. I am not a lawyer so I cannot argue the finer legal points. But consider examples: can I make a patch to Linux that fixes some bugs and only sell it under a license that does not allow redistribution? I cannot. Can Microsoft add to IE a feature that automatically downloads Sun's JVM and patches it to contains its own extensions? They cannot. Neither the patch nor the extentions stand on their own. They both consist a part of a larger program that is covered by a copyright license and has to abide by it. > AS>> I do think a program's libraries should be considered part of the > AS>> program. Or else, if someone wanted to turn GIMP into a proprietary > > You mean, if you source has possibility to link it to GPLed library, it > has to be GPL? In other words, if you allow possibility to attach some > GPLed code in some way to your system, all your system should be GPL? > Let's say we are talking Java and not C - you don't have any includes, you > can directly call GPLed library in runtime without ever mentioning it in > the source. By this point of view, any program that allows user to do this > (call GPLed library in run-time) and isn't GPLed is illegal. Which pretty > much makes all Java code illegal, since you can just pass text string > obtained from the user to class loader and say "load class with this > name". > > Now that can be extended, for example, to make any GPL software on > Windows illegal - since Windows can take advantage of GPLed code, and > obviously Windows isn't GPL, so it's illegal by RMS's point of > view. That only depends on what you call the library and what - the > executable, and that's just technicality - you can call Windows being > executale and running program being library - actually, some systems do > work this way. See my point here? You seem to be saying that: 1. Distributing proprietary code linked to GPLed code is illegal. 2. Therefore: anything that _can_ link proprietary code to GPLed code is illegal. 3. Therefore: Java is illegal. And since 3 is obviously not true, 1 is not true. But this is the same as saying: 1. Distributing Microsoft Office is illegal. 2. Therefore: anything that _can_ distribute MS-Office is illegal. 3. Therefore: FTP is illegal. This does not hold. Distributing MS-Office, or Solaris' bourne shell linked to readline are both illegal, because they both do something in practice. Not because they _can_. > AS>> Everyone has their own definitions. > > Yeah, MSIE is "free" too - look on the Microsoft site. Free as in beer. A distinction added by RMS to separate the two things. You can always add your own distinctions, if you like. What is your point? - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Problem booting with Kernel 2.4.0-test12
On Fri, Dec 29, 2000 at 05:17:58PM +0200, Noam Meltzer wrote: > > Content-Type: text/plain; charset=us-ascii > Content-Transfer-Encoding: 7bit > > Hello! > I encountered a very strange problem after I installed Kernel > 2.4.0-test12. > After doing the normal routine for building a kernel (make menuconfig, > make dep, make bzImage, adding it to 'lilo.conf 'and doing 'lilo') > Ok, I reboot, start the new message and then i get this message (I quote > the last 3 lines of all the log showed on the screen, like in any normal > boot procedure): > > VFS: Cannot open root device "1607" or 16:07 > Please append a correct "root=" boot option > Kernel Panic: VFS: Unable to mount root fs on 16:07 The kernel cannot mount your root filesystem. I can think of two possible reasons for this: 1. The kernel is configured to mount the wrong block device. I don't know what 16:07 is, but you can change it in /etc/lilo.conf or, like the message suggests, add a root=/dev/whatever at the lilo prompt. 2. The device is correct but unusable. Did you remember to compile IDE/SCSI, ext2/reiserfs and ELF support into the kernel? A module won't help; the kernel needs to access the filesystem before it can load any modules. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ipchains
s of Net*, Open*, etc. One of the major goals of BSD is to raise the software quality bar for proprietary products by letting vendors base such code on BSD and add proprietary code to it[1]. I find BSD's goal admirable but it does not settle well with GPL software author's goals, who do NOT want their software used in proprietary products. If a project's license gives users all rights the GPL gives, there is no problem using readline. But you won't be able to take the project and turn it into a proprietary one without taking out the GPLed code first. If this is the case, BSD had to reimplement readline so that it can be used in proprietary products. This is different from the case of programs whose license actually have restrictions NOT in the GPL. AFAIK FreeBSD no longer does. See SunOS, NeXT, MacOS X for examples of complete systems based on BSD, and numerous smaller parts of code used by proprietary systems, e.g. sockets. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ipchains
On Thu, Dec 28, 2000 at 11:07:47PM +0200, Stanislav Malyshev a.k.a Frodo wrote: > With our GPLed software the matter is pretty complicated. Generally, GPLed > software is referred to as a "free software". But, in fact, it's not free > at all, in the common meaning of the word "freedom". You cannot use it in > any project whose views on software licensing is not the same as RMS's. > And "use" here includes interoperability, by RMS views, i.e. linking with, > etc. See RMSs commentaries on readline library. That's a common misconception. It should have been obvious, but somehow never is, that no amount of licensing trickery can make one program be considered a derivative work of an unrelated program. And the GPL doesn't attempt to, either. Likewise, no program can "contaminate" other programs and change their license, whether or not you link them together. What the GPL is saying that you cannot distribute someone's GPLed code as a part of a larger program unless you give users at least the same rights they would get were the program GPLed. So you can take GPLed Linux drivers and use them in FreeBSD, because the amended BSD license gives users more rights than the GPL. But you cannot use them in AIX. I think that is reasonable to many authors, who would not want their products to be used by proprietary products. If you think otherwise, use a different license for your own code. If you want to use others' GPLed code in more restricted programs, go ahead and ask the authors. I fail to see why this makes you so furious. Also, nowhere does the GPL limit "interoperability". Only use code as part of another program. Whether or not you consider an executable and a dynamic library accompanying it two parts of the same program or not is a different issue. GPLv3 intends to clarify this and other points. > Some commercial companies > explicitly prohibit their workers even to touch GPLed software or enter it > into the campuses. Not because they hate free software so much - but > because current copyright system can make all their code "derived work" > just because some of the developers took a look into GPLed code. I've heared similar stories, and even worse ones, told about proprietary products. It is true that the legal system is confused by new challenges to the old "intellectual property" laws, and that several bad judgements where lawyers' ridiculous claims won cases caused companies to become worryingly paranoid. All this has literally NOTHING to do with the licensing of code. The only reason why companies are more worried about GPLed programs than about proprietary programs is that with GPLed code there is a risk of some employees actually reading it. > That's > not exactly how I understand "freedom". Recently, I start to think that if > people would know and think a bit more on RMSs interpretation of GPL > (which is pretty much of "you cannot touch it unless you are with us in > GPL") this license probably won't get so popular. GPL became yet another > modern myth. Not true. See above about using GPLed code with non-GPLed code. People who don't want to restrict the use of their code in proprietary programs or with somewhat more restrictedly license code should not use the GPL, and usually don't. For example, most of Gnome is under the LGPL (beats me why KDE chose the GPL while opposing it so badly). But I think there are many people who want to contribute to free software and not necessarily to proprietary software, and that exactly is the reason for the GPL. I can't see what your problem is here, except for the linguistic definition of "free". > Also, this copyright business got really out of proportion. I cannot just > take some "free" software and use it, I need to check license and all > legalese and get almost a degree in law to understand what's written > there. And I saw a number of times that people had to reimvent the wheel > and reimplement existing "free" code just because it was GPLed. I'm not > talking here about commercial software, I'm talking here about open source > project that happen to disagree with RMS on how to license their own > program products. Is it really useful for anyone? I doubt it. No, you don't have to. If you want to put your code under the LGPL, or the public domain, or the amended BSD license, or whatever, you are free to do so and are free to combine your program with any GPLed code you can lay your hands on. (I don't know what projects you are referring to that had to reinvent the wheel, but I gather they were either misinformed about the GPL's restrictions, or specifically wanted their project to be forked into proprietary products, like the BSDs do.) - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Sending SMS messages by E-mail + Sending E-mail from an SMS cellphone
On Thu, Dec 28, 2000 at 05:28:38PM +0200, Ely Levy wrote: > at worse you can originize a patition to cancel thier licence. > and give it when/if we have a new goverment;) You mean, BEFORE we have a new government :) Listening to citizens right after elections is not very productive... - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: spaces in filenames
On Wed, Dec 27, 2000 at 12:55:19AM +0200, Nadav Har'El wrote:
> On Tue, Dec 26, 2000, Adi Stav wrote about "Re: spaces in filenames":
> > Speaking of, by the way, I wonder why almost everybody I've seen who
> > digs smart xterm titles uses the cd trick instead of something like:
> >
> > PS1_SIMPLE="\h: \w "
> > PS1="\[\033]0;${PS1_SIMPLE}\007\]${PS1_SIMPLE}"
>
> Maybe because your solution (at first glance - I didn't try it) changes the
> title on every prompt? This is a waste of bandwidth (though you won't notice
> it unless you have a slow line - e.g., log in to a computer abroad, or
> use a slow modem, or worse - a slow X connection). In some window managers it
> may also cause the title to flash on every prompt (though if it does, it's a
> bug that needs to be corrected in the window manager).
>
> Anyway, your solution isn't bad - it's just that there's a more efficient
> solution: to only change the title when it needs changing, i.e., when the
> current directory has changed.
Ah, no. We are talking about five bytes here (plus the length of the
title). And since it will arrive through telnet (or equivalent)
together with the rest of the prompt, in normal use it will occupy a
very small single packet that has to be sent anyway, if only for the
newline. So there will generally be no latency problem even with very
slow modems.
I don't like the cd aliasing solution because it feels kludgy: what it
the cwd changes not because of cd (like choo pointed out)? What if you
want to display data other than your cwd in the title? Anyhow, if
efficiency is an principal thing, I suggest caching:
CURRENT_TITLE=""
retitle() {
OLD_TITLE="$CURRENT_TITLE"
CURRENT_TITLE="$*"
[ "$CURRENT_TITLE" != "$OLD_TITLE" ] && \
echo -n "\033]2;${CURRENT_TITLE}\007"
}
export PS1='\[`retitle whatever`\]whatever'
Warning: I haven't checked this one, don't have X at the moment.
> By the way, in zsh there's an even more elegant solution (if anybody got
> the impression that I'm an avid zsh advocate, they are right): zsh's
> standard 'cd' builtin runs a user-define function 'chpwd' after changing
> the directory, if such a function exists. So all you need to do is to
> add something like the following to your .zshrc (no need for 'cd' functions
> or aliases):
>
> chpwd(){
> case "$TERM" in
> xterm|screen) print -n '\033]2;'"$LOGNAME@$SYS":$PWD'\007';;
> esac
> }
> chpwd # run this once in the .zshrc to show the initial current directory
Did you manage to get it working with screen? Screen ate my control
characters when I tried it...
- Adi Stav
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: spaces in filenames
On Tue, Dec 26, 2000 at 08:15:39PM +0200, Oleg Goldshmidt wrote:
> Brilliant! Stupid me! I had
>
> # let the xterm titlebar show the current dir
> xtitle() {
> if [ "$TERM" = "xterm" ]; then
> echo -n -e "\033]0;$@\007" >/dev/tty
> fi
> }
>
> # Change the 'cd' 'pushd' 'popd' commands to use xtitle()
> cd() {
> builtin cd "$@"
> xtitle ${HOSTNAME%%.*}: $PWD
> }
>
> in .bashrc! Thanks!
Ahh... I once got bitten by such, and since then I always surround my
arguments with double quotes in scripts. I find it just as easy to
type and to read as the braces people usually put in.
Speaking of, by the way, I wonder why almost everybody I've seen who
digs smart xterm titles uses the cd trick instead of something like:
PS1_SIMPLE="\h: \w "
PS1="\[\033]0;${PS1_SIMPLE}\007\]${PS1_SIMPLE}"
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: XWindows with masquerading
On Fri, Dec 22, 2000 at 12:52:34AM +0200, guy keren wrote: > > On Thu, 21 Dec 2000, Arieh Bibliowicz wrote: > > > I intalled in my house a Mandrake7.2 linux as a masquerading server that > > connects to the internet using ppp and is connected to a home net. > > I wan to run remote X applications and use ANOTHER computer in the network > > (not the server) as the DISPLAY. Howcan I do this? > > one option is to install a port redirector on the masquerading server, > have it bind to port 6001 on the server machine, and forward connections > to port 6000 on the other local machine. then login to the remove > machine, setenv DISPLAY masquerading_server:1.0 > > and then try launching an xterm. if that work, you're done. ofcourse, > you'll still need to mess with x authorization, but i guess you're > familiar with that. > > btw, with netfilter (kernel 2.4), there is kernel-support for port > redirection. Alternatively (easier IMHO), use ssh -X. More secure, too. Login into the router, set DISPLAY to the location of the X server, and ssh -X anywhere. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: 2 more questions, if you please -
> 1. how can I upgrade RPM from 3.x to 4 without losing the database of all > the RPMs installed ? (as I understand, the DBs arent' compatible) can you > supply a step-by-step ? RPM 3.0.5 is supposed to be able to handle the transition. Not that I'm crazy enough to try it. ObRTFM: http://www.rpm.org > 2. whenever I install something using RPM, I get the following error (after > all the files have been copied, but I'm not sure about the scripts being > run) : > error getting record /bin/sh from //var/lib/rpm/requiredby.rpm > then is pauses for a few seconds before it seg faults. > when uninstalling (same package that was installed) I get the same error > message, but no core dump. > any clue ? Sounds like db corruption, but you already knew that. Sorry, can't help you there. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: biditext
On Thu, Dec 14, 2000 at 11:50:25PM +0200, Tzafrir Cohen wrote: > Anybody wants to rewrite this as a dock-applet/epplet/kde-applet/ > gnome-applet ? > > Anyway, I've packaged it along with a couple of menu items (I put r2l and > the menu items in seperate packages to avoid unnecessary dependency > problems): > ftp://linux.org.il/pub/Hebrew/Rpms/srpms/biditext-0.0.3-2.src.rpm > > ftp://linux.org.il/pub/Hebrew/Rpms/i386/biditext-0.0.3-2.i586.rpm > ftp://linux.org.il/pub/Hebrew/Rpms/i386/biditext-menu-0.0.3-2.i586.rpm > ftp://linux.org.il/pub/Hebrew/Rpms/i386/biditext-r2l-0.0.3-2.i586.rpm > > Enjoy I should probably be doing this instead of talking, but since you've brought it up here's a suggestion: Don't use a system-global file in /tmp, use a file in $HOME. Better yet, have it overridable by a file whose name include the PID of the process, so that people can run one netscape for logical Hebrew and another for Visual Hebrew for example. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Me
On Wed, Dec 13, 2000 at 04:54:07PM +0200, Oleg Goldshmidt wrote: > Adi Stav <[EMAIL PROTECTED]> writes: > > > Why aren't they? Unix, BSD and SysV are/were actual branded OSs that > > were sold under that brand, no less than Red Hat or HP/UX are now. > > I suppose UNIX was a brand when it was controlled by a single company > and was immediately associated with that company. What is your > immediate association when you say UNIX today? SCO? Sun/Solaris? > HP/UX? IBM/AIX? DG/UX [G-d forbid!]? Now UNIX is more like jeans, > which is not a brand, while Levi's or Wrangler are. Or a hamburger as > opposed to McDonald's or Burger King. At least that's how I think > about branding. Any of the above. Sun, IRIX etc still have still have respectably large chunks of AT&T code in them. They are Unix descendants, unlike Wrangler that has nothing to do AFAIK with Levi's. GNU/Linux (separately or together) are not really "a Unix" in the Sun or IRIX sense of the word, also by the way people refer to them. But I don't know how we got to this subject :) > > the term "operating system" meaning a complete and branded software > > system has become far too common by now, so I don't really mind it. > > I suppose we should allow the terminology to evolve. > > > I don't usually make a fuss over it. > > No fuss, just let's define the terms. Ok. "software system" or "distribution" vs "kernel" good enough for me :) > > But whichever definition you choose, saying that Red Hat is a variant > > of Linux is not more true than saying that a Compaq PC is a variant of > > Intel's Pentium. > > I am not sure I am getting it. PC is not a brand. Intel is, and Compaq > is. A Compaq PC can be a brand but it is not a brand of Pentium. Just like Red Hat is a brand, but not exactly of Linux (certainly not a VARIANT of Linux), but of a complete software system of which the Linux kernel is a major component. The Pentium is the same in all Pentium-based PCs, and Linux is the same in all Linux-based distributions (the occasional minor patch notwithstanding). > And assuming that there are AMD-based Compaqs, and potentially > Transmeta-based Compaqs (not too soon, I guess), there are different > combinations of brands, all of which as associated with the notion of > "PC". Ah, yeah. This is true. Forgot about AMD and friends. Anyway, you see my point. The BSDs often contain much of the same applications as the Linux-based distributions, and they can even run Linux binaries unchanged. But they have entirely different kernels. > > You don't see the PC forking anytime soon, do you > > I think we've seen the PC forked many times already. Even if we do not > consider Apple products PC's, when you buy a "PC" you'll boast (or be > ashamed of, as the case may be) of buying a Compaq PC with an Intel > CPU and this motherboard and that videocard et caetera. All these are > forks in the same sense as But people are not concerned about a Compaq'a Pentium CPU being incompatible with Dell's. There might be several alternatives to each component, but mostly the different alternatives are not associated specifically with a PC brand. Compaq or Dell have no incentive to develop their own forks/clones of Pentiums or hardisks. Why should they? Intel and Seagate already do a good job at that, and all they have to do is pick. Similarly, Red Hat has no incentive to have its own version of Gnome or Linux. This is all very different from Unix where each fork had its own special versions of each of the components. That's what I am arguing -- that the GNU/Linux software systems model parallels the PC model in that aspect, not the Unix forks model. The media's confusion and hysteria about a "Linux incompatible forks" is caused by misunderstanding of the model, partly due to bad phrasing and definitions like you pointed out earlier about "OS" :) = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Me
On Wed, Dec 13, 2000 at 03:51:26PM +0200, Oleg Goldshmidt wrote:
> Adi Stav <[EMAIL PROTECTED]> writes:
>
> > But Linux is not a brand.
>
> No, it isn't. Neither is UNIX. Nor are BSD or SysV. Red Hat is. HP/UX
> is. What has it got to do with OS?
Why aren't they? Unix, BSD and SysV are/were actual branded OSs that
were sold under that brand, no less than Red Hat or HP/UX are now.
> > Most of the Linux-based distributions include the Linux kernel as it
> > is, or with relatively negligible changes. Linus has neither the
> > power nor the desire to influence "userland" applications. It is Red
> > Hat, Debian etc who brand the complete systems. So I would not say
> > Linux is a single operating system, but a family of operating
> > systems, because there's no OS you can point at and say "this is the
> > original Linux", unlike Unix or BSD.
>
> I suggest that you guys agree on the definitions of "OS", "Desktop
> Environment", "Application" etc. It seems to me that the layman's (no
> Adi, I don't mean you, don't jump :) understanding of the term OS is
> stroingly influenced by the notion of a "Windows Operating System"
> that includes the whole preinstalled package. Many components of that
> package (and of Linux distros) rightfully belong to the "Applications"
> category, I believe. From a more professional/CS point of view, in the
> "Operating Systems" course students study Linux the Kernel, I imagine
> (those who took an OS course please correct me if I am wrong :).
>
> My personal bias is that not only KDE or GNOME, but X, emacs, even ed
> or find are not parts of UNIX or Linux. Call them environments,
> applications, utilities, tools as appropriate. To support this point
> of view here's http://info.astrian.net/jargon/terms/o.html#operating_system
I agree Oleg, the traditional definition makes more sense nowdays. But
the term "operating system" meaning a complete and branded software
system has become far too common by now, so I don't really mind
it. It's not only laymen who use this definition anymore, and since
there's an easy word to replace the original meaning of OS ("kernel"),
I don't usually make a fuss over it.
But whichever definition you choose, saying that Red Hat is a variant
of Linux is not more true than saying that a Compaq PC is a variant of
Intel's Pentium. I see these definitions all the time in the trade and
mainstream media and I think that this is one of the reasons why
outsiders often think that Linux is in the process of a disasterous
fork. (You don't see the PC forking anytime soon, do you :)
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: Me
On Wed, Dec 13, 2000 at 01:46:53PM +0200, Boaz Rymland wrote: > Tzafrir Cohen wrote: > > > > With the current amount of differences between various linux distros, > > calling each of them "operting system" is more correct. > > I wouldn't join your terminology. > IMO, > * Linux is the Operating system brand, > * RedHat,Debian are distributions, meaning a typical, not identical to other, > type of Linux. They differ in numerous minor/less minor ways. Each dirto' is > like a special flavour of Linux. > > I wouldn't call each distro a different OS. They are all able to run the same > binaries (true they are packaged differently, but this is just the *package*, > not the content), and they simply share much more than they differ, if you look > at the whole computers OS world, not only at the Linux world. But Linux is not a brand. Most of the Linux-based distributions include the Linux kernel as it is, or with relatively negligible changes. Linus has neither the power nor the desire to influence "userland" applications. It is Red Hat, Debian etc who brand the complete systems. So I would not say Linux is a single operating system, but a family of operating systems, because there's no OS you can point at and say "this is the original Linux", unlike Unix or BSD. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Me
On Wed, Dec 13, 2000 at 11:51:45AM +0200, Shaul Karl wrote: > > Now. let's see if I got it right: > > RedHat is the operating system, and the > > KDE, GNOME... are the envyroments - > > Which means: I'll have the Linux Redhat 6.2 > > and will be able to function on several > > envyroments. > > > > Hope I'm not a slowlearner... > > I prefer to say that Linux is the operating system, RedHat is the > distribution, 6.2 is the distribution version, and Kde or Gnome are the > desktop environments. You should be able to use both Gnome and Kde, but not > concurrently, unless you have several monitors. ..But you will be able to run any applications in any environment. A desktop environment only means what your desktop set up will be: how the windows looks like, what is the default file manager etc. You can still run Gnome programs in KDE or KDE programs in Gnome with no problems. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Debian in Israel
On Tue, Dec 12, 2000 at 11:53:43AM +0200, Alex Shnitman wrote: > Hi, > > I've registered the domain debian.org.il, and I'd like to gauge the > interest for creating a local Debian community. The goal would be > Debian support in Israel, advocacy, experience sharing and the like. > If you're interested in anything like that, please send me e-mail. If > there is enough interest we'll open a mailing list. Count me in. Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Need Help Installing Redhat 7.0
On Wed, Dec 06, 2000 at 03:08:35PM +0200, Aharon Schkolnik wrote: > > Hi. > > > I just downloaded the RedHat image 7.0-respin-disc1.iso. I checked the > md5checksum, and it is correct. I burned it using the command: ..snip... > However, after I choose Local CDROM as my installation media, I get a > screen with the message: > > I could not find a Red Hat Linux CDROM in any of your CDROM drives. > > > On console 3 I see: > > * trying to mount device hdc > * loopfd is 6 > * failed to mount loop: Invalid argument > * 92 keymaps are available > * loaded 9 keymap tables > * trying to mount device hdc > * got url (null) > * trying to mount device hdc > > > On console 4 I see: > > <4>md.c: sizeof(mdp_super_t)=4096 > <4>Partition check: > <4>hda: hda1 hda2 > <5>RAMDISK: Compressed image at Block 0 > <4>VFS: Mounted root (ext2 filesystem) > <6> ... USB stuff ... > > > <7> ISO 9660 Extenstions: Microsoft Joliet level 1 > <7> ISO 9660 Extenstions: RRIP_1991A > <4> Unable to identify CD-ROM format > <4> VFS: Can't find an ext2 filesystem on dev loop (7,0) ^^^ Looks like there is nothing wrong with your CD. Sounds like Red Hat thinks /mnt/cdrom or whatever is supposed to carry an ext2 filesystem rather than iso9660. I vaguely remember similar problems with me having to modify /etc/fstab after every Red Hat 6.2 installation so that it knows the cd is iso9660, but I could be misremembering. What you could do: find where Red Hat stores the "ext2" bit (fstab? a script?) and change it to "auto" or to "iso9660". Or, maybe you could mount the cd manually from the console and then have Red Hat continue installation. Debian supports such manual intervention, but I don't remember if Red Hat does. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: wine/ does it work ?
On Mon, Oct 16, 2000 at 02:14:43PM +0200, Ben-Nes Michael wrote: > Hi > > I tried to install office on wine but I had few problems. > > So before i dig my self into it, did any one had a good experience with > wine + windows office (2000 or 95) + hebrew ? AFAIK (had the same problems, but don't know much more) the Wine project has not (yet?) implemented the BiDi functions necessary for true Hebrew support, including the support needed for MS-Office. So I think you can't. Will be glad to be proved wrong. What you can do: 1. Find someone to implement BiDi in Wine. I don't know how hard or easy this will be, it would involve some Windows internals. 2. Use QText or Dagesh. AFAIK both don't use Windows' BiDi functions and use their own, so might work on a non-Hebrew versions of Windows, including Wine. Help the local industry and all :) 3. Run MS-Office using VMWare or Plex86 (if it's good enough yet -- worth trying). Never tried Win4Lin, it might be useful too. Regarding non-Hebrew MS-Office, I've run Office '97 successfully without Hebrew. It was a little quirky, though. Wine is supposed to be able to run Office 2000 now but I never tried it myself. Good Luck, Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: number of CPUs
On Tue, Nov 07, 2000 at 07:24:35PM +0200, Ely Levy wrote: > I'll take that as a no? > geezz.. > > > Ely Levy > System group > Hebrew University > Jerusalem Israel > > > > On Mon, 6 Nov 2000, Adi Stav wrote: > > | On Mon, Nov 06, 2000 at 10:55:28PM +0200, Ely Levy wrote: > | > how do I get the number of CPUs a computer has > | > without parsing any file. > | > something like sysctl on BSD.(just in linux it's not there) > | > | /proc/cpuinfo Oh sorry, I skipped the "parsing" line. Anyhow... I reckon Linus has expressed before his dislike for special system calls to get kernel information, and wants people to use proc only. So unless I'm missing something I guess you're out of luck. But why is it so crucial not to parse a file? I can't imagine anyone checking for the number of CPUs in the inner loop of their programs :) system( "exit `grep ^processor /proc/cpuinfo | tail -1 | cut -d : -f 2`" ) Would probably work (I don't have SMP machine to check it). = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: number of CPUs
On Mon, Nov 06, 2000 at 10:55:28PM +0200, Ely Levy wrote: > how do I get the number of CPUs a computer has > without parsing any file. > something like sysctl on BSD.(just in linux it's not there) /proc/cpuinfo = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Cool shell tricks (was: Re: file system questions)
On Sat, Oct 21, 2000 at 08:23:44AM +0200, Shaul Karl wrote: > > Also, seems that not many people are aware of "process substitution" > > in bash: you do something like "diff <(ls dir1) <(ls dir2)", and bash > > runs two ls's, each writing to a FIFO, and diff gets the names of > > these two FIFOs as parameters. Output redirection also works. I'm not > > sure how portable it is, though. > > 1. > [08:05:37 /tmp]$ diff < (ls ~/setiathome/) <(ls ~/setiathome/amos/) > bash: syntax error near unexpected token `(l' > [08:06:04 /tmp]$ > > Ache, the space there is significant: > > [08:17:12 /tmp]$ diff <(ls ~/setiathome/) <(ls ~/setiathome/amos/) Yeah... Otherwise it's redirection. > 2. > This will only work if the "main" process (diff in this example) does not need > to seek the input since it can not do it with a FIFO. Or am I missing > something? True... You can't. It would've been nice if Bash had the option to use temporary files rather than FIFOs... But I don't know if it's needed that often. Just tried: both nvi and vim cope fine, Emacs refuses to read it (at least with default setup and switches). Less requires you to use -f. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Cool shell tricks (was: Re: file system questions)
On Tue, Oct 17, 2000 at 04:59:48PM +0200, guy keren wrote: > there is a dirty method you could use. every once in a while (perhaps even > via a cron job), execute the command: > > :>/path/to/.xsession-error Cool. I love these little shell tricks... I'm personally particularly fond of the old "(cd dir1; tar c) | (cd dir2; tar xv)" thing, but it's been discussed here extensively before. Also, seems that not many people are aware of "process substitution" in bash: you do something like "diff <(ls dir1) <(ls dir2)", and bash runs two ls's, each writing to a FIFO, and diff gets the names of these two FIFOs as parameters. Output redirection also works. I'm not sure how portable it is, though. Does anyone else like these tricks? I'd love to hear about other people's. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: LGPL question
On Wed, Oct 18, 2000 at 11:47:54AM +0200, Felix Shvaiger wrote: > Hi All ! > > May I develop proprietary application (executable) that uses LGPLed library > in form of shared library and distribute nothing but: > 1. executable file itself > 2. my proprietary license for this executable > 3. notice that this executable uses some LGPLed libraries > 4. shared object file of LGPLed library (to make my executable working) > 5. LGPL license file > 6. link do LGPLed library's source code download site > 7. instructions about making shared library from source > > Have I got it right ? > > Do not ask, why I don't GPL my applications - it is not up to me. If I'm not mistaken (IANAL), if you provide the LGPL library commercially (as opposed to free as in *beer*) you need to also provide the source together with the binary, or an offer to provide it for no charge. There is a clause allowing you to settle for a link, but it (IIRC) does not apply for commercial distribution. Of course, you don't have to GPL your source if the library is only LGPL... Also, see the http://www.gnu.org site. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Web search tools
Perhaps you could help me: I'm looking for a simple web search tool or spider to map all the links from a certain URL starting point(s) to a certain depth, and print or log the URLs it encounters. I've been using "wget -r -l 5" so far but it cannot parallelize its requests, visits the same URLs several times and cannot be told to refrain from downloading files unnecessarily, which makes it far too slow. Does anyone have experience with or knows a tool that does have these features (especially parallelizing requests)? TIA, Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: fsck and async mode
>Adi once suggested printing a T-shirt with the following: > >LILO boot: linux init=/usr/bin/emacs > The Return of the LISP Machines > >Anyone still interested? ;-) That was Moshe Zadka, not me... But I'm still interested :) - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: OT: BeOS
On Tue, Apr 04, 2000 at 02:13:54AM +0200, Ereli wrote: > Hi > did any one here tried BeOS 5? > I did and i saw it uses many GPLed tools, stuff like bash, and gcc. > is any one found the license packed somewhere in the image? if it isn't there, isn't >that a violation of the GPL? > can any one comment on this issue? Really? The GPL has an exception clause that allows it to be linked to non-GPL'd software if that non-GPL'd software is an integral part of the OS, but that exception does not hold if the OS is distributed together with the GPL'd software. So the GPL would require that either most of BeOS is GPL'd as well, or that these tools are distributed separately from BeOS... I'm curious... > Ereli - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: (no subject)
On Mon, Apr 03, 2000 at 12:46:49PM -0400, Webmaster; IT3 Weaver wrote: > Hi, my name is Dan Weaver, I'm a sailor on the USS Dwight D Eisenhower. I > run apache web server on a Mandrake 7 box and was wondering if you are > having any kind of meeting around the 14th of this month. We'll be in Haifa > then for a port visit. Hi Dan! It's very cool to receive such emails... Where did you hear about us? I'm afraid we don't have any meeting planned in Haifa in the 14th... There are two meetings in Haifa planned for the 9th or the 11th, though. I'm not sure what parts of them might be in English. We do have a meeting which we might hold in the 14th, but it is not final yet, and will probably take place in some place other than Haifa. It is going to be mostly about developing Hebrew and bidirectionality support in Linux. Visit http://www.linux.org.il as updates and planned events are posted there on a regular basis. > Dan Weaver, MCP+I, MCSE - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: OT: Amazon.com - don't buy from Nazi supporters
On Mon, Apr 03, 2000 at 11:00:09AM +, Avi Berliner wrote:
> ps. If you are boycotting amazon than I suggest that you also don't buy from
> barnes and npble as they also sell the book.
Do we need to find new reasons to boycott amazon.com? I'm amused :)
http://www.gnu.org/philosophy/amazon.html
- Adi Stav
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: kerneli
On Sat, Apr 01, 2000 at 01:25:07AM +0200, Shaul Karl wrote: > > Linux Kernel Mailing List. I recommend the excellent summaries at > > http://kt.linuxcare.com . > > I looked for these summaries but could not find them. Can you be more > specific? Can these summaries be subscribed to so that it will be obtained > automatically every day or every week? I am thinking about the DWN (Debian > Weekly News) model. That's the model. I'm not sure who copied whose ideas, KT or DWN. Try the links on the left bar... KT/KC news (also summaries from Wine, Gimp, Hurd...) or Kernel Traffic Home: Latest Issue. > Then the main purpose is that the kernel src will be signed and have an MD5 or > similar checksum. The fact that it gets encrypted is only a by product. Am I > right? Ah... AFAIK, the idea was to put in the kernel code that does encryption. Not encrypting the kernel source itself. Like an encrypted filesystem for example, encrypted block devices... For the kernel itself to do encryption of other data. That would make the kernel an encryption device. Some countries put problems on distribution of encryption devices. -- Shaul Karl <[EMAIL PROTECTED]> - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: kerneli
On Thu, Mar 30, 2000 at 09:02:06AM +0200, Shaul Karl wrote: > > On Tue, Mar 28, 2000 at 09:30:49PM +0300, Gaal Yahas wrote: > > > Hi! > > > > > > I'm not following lkml. Could anyone who is say why aren't the > > > international and mainstream kernel trees being merged? (Or > > > perhaps they are, but it'll take some time?) > > > > > What is lkml? Don't I have only one kernel src on my machine? Then why are > you talking about merging the kernel trees? Linux Kernel Mailing List. I recommend the excellent summaries at http://kt.linuxcare.com . > > It was suggested, but it was turned down because there are still many > > countries in the world (I think Sweden was the example used) that have > > encryption problems and they didn't want to limit Linux's distribution > > or make mirroring illegal. > > > > > What encryption has to do with it? Lots of countries have legal limitations on the use, export or import of encryption software. The US recently lifted some of its own limitations but many countries still have them. The Linux Kernel guys want to make sure that people can download or create mirrors of the kernel without risking themselves. So http://www.kerneli.org holds an "international" (i.e., non-us, because of the US limitations in the past) tree adding encryption to the main kernel. Apparently Linus now wants to merge the trees, but to make sure that people who can't take the encryption stuff can remove it easily. > -- Shaul Karl <[EMAIL PROTECTED]> - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Lilo & > 1024 Cylinder Disk.
On Thu, Mar 30, 2000 at 05:09:15PM +0300, Ely Levy wrote: > Lilo or any other booter for that matter can load from after the 1024 > cylinder (about 8 gig) limit make the boot partition before.. IIRC, GRUB can do that. Never used it though. I think it actually reads the filesystem without the BIOS's assitence to reach the kernel. Anyone with details / corrections? > Ely Levy > System group > Hebrew University > Jerusalem Israel - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: kerneli
On Wed, Mar 29, 2000 at 08:36:40AM +0200, Adi Stav wrote: > On Tue, Mar 28, 2000 at 09:30:49PM +0300, Gaal Yahas wrote: > > Hi! > > > > I'm not following lkml. Could anyone who is say why aren't the > > international and mainstream kernel trees being merged? (Or > > perhaps they are, but it'll take some time?) > > It was suggested, but it was turned down because there are still many > countries in the world (I think Sweden was the example used) that have > encryption problems and they didn't want to limit Linux's distribution > or make mirroring illegal. Update, from the latest kernel traffic (http://kt.linuxcare.com): > Lars Marowsky-Bree asked if crypto support would make it into 2.4, or > would it have to wait until 2.5; H. Peter replied, "I talked with > Linus about this, and he says he thinks it can get merged into 2.4 > assuming it is reasonably separate, but "probably not for 2.4.0."" - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: kerneli
On Tue, Mar 28, 2000 at 09:30:49PM +0300, Gaal Yahas wrote: > Hi! > > I'm not following lkml. Could anyone who is say why aren't the > international and mainstream kernel trees being merged? (Or > perhaps they are, but it'll take some time?) It was suggested, but it was turned down because there are still many countries in the world (I think Sweden was the example used) that have encryption problems and they didn't want to limit Linux's distribution or make mirroring illegal. > Thanks! > > -- > believing is seeing > [EMAIL PROTECTED] > http://www.forum2.org/gaal/ - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: X116.4 vs Xfree86-4
On Mon, Mar 27, 2000 at 11:39:10PM +0200, Game Wizard wrote: > Hi!! > > I have a question. What is the difference if any, between XFree86 4.0 and > X116.4 ?? > > They have completely different installations and documentation. From the > docs I understood that X116.4 supports all kinds of UNIX's and NT. > > But is it still the same X or not ? X11 is a standard for X servers which provide GUIs. You're referring to revision 6.4 of it. XFree86 is a specific (and free) implementation of X11. If you're curious about these things, I think you'd find the HOWTOs to be very inetresting of informative... They come with most distributions (notably excluding Mandrake 7.0) and are also available on http://www.linuxdoc.org . Usually you'd find you don't need to ask most of the questions in the list because they're all in the HOWTOs. > Blessed be, > GW. - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: MySQL and encryption
On Thu, Mar 23, 2000 at 02:58:31PM +0200, Yuval El-Hanany wrote: > Hi, > a customer asked me whether MySQL can keep its data encrypted on the disk. > Does anyone know if that's possible in MySQL or perhaps there's a file system > driver which can do that? IIRC, there are kernel patches that act as encryption translators for block devices, so you can use any filesystem, encrypted on disk. They're not integrated in the standard kernel for legal reasons but are available on http://www.kerneli.org. I've never checked these out so I'm not sure if this information is accurate, but do tell me if it works :) > -- >Yuval El-Hanany | Kawasaki GPZ500 '97 | > [EMAIL PROTECTED] | stolen :-( | Have backpack, > Home : 972-3-6993301| Debian Linux Inside |will travel > Work : 972-9-9586077-227| | - Adi Stav = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
