Slashdot Mirror


User: hummassa

hummassa's activity in the archive.

Stories
0
Comments
2,076
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 2,076

  1. Not really... on Why One Man Got a Guerrilla RFID Implant · · Score: 1

    But it would also make identity theft not as painful.

  2. Let me try to understand you... on Hurd/L4 Developer Marcus Brinkmann Interviewed · · Score: 1

    How is the IPC scheme under L4? Why isn't it comparable to QNXs?

    Aren't the problems you mentioned just fixable in the scheduler?

    Isn't it enough to boost the priority in the queue for recently-activated threads and to un-boost the prio for threads that recently activated other threads? The former are likely to receive CPU time soon, and the latter are likely to relinquish CPU soon, anyway?

  3. Back in 1992... on Hurd/L4 Developer Marcus Brinkmann Interviewed · · Score: 1

    I worked at a firm that made payroll and accounting software to various Unix flavours. When I got there, none of our machines even had a network adapter. IIRC we had DEC Ultrix, HPUX8, Solaris, AIX, SCO Xenix, and others. The code was sometimes out of sync from one machine to another, and the fixes and upgrades were rolled from machine to machine via removable storage.

    Well, I networked the whole bunch, unified the source in NFS, and started to unify the build toolchain to gcc, autoconf, and others. Why? Because we had a lot of different tools, different makefiles, it was a living hell.

    In the end of this project, we had one source tree, with 100x less #ifdefs, and even if our systems got bigger in some archs, they were at least 10x cheaper to maintain. Thanks to GNU.

  4. FYI on Debian Release Mgr. Proposes Dropping Some Archs · · Score: 1

    Even NT/XP can be installed with FAI.

  5. Re:No, pkgsrc does... on Debian Release Mgr. Proposes Dropping Some Archs · · Score: 1

    pkgsrc is released separately from NetBSD, and it has a good QA done on its 5300 packages.

  6. No, pkgsrc does... on Debian Release Mgr. Proposes Dropping Some Archs · · Score: 1

    Oh, sorry... :-)

  7. Re:Dropping ARM??? on Debian Release Mgr. Proposes Dropping Some Archs · · Score: 1

    But there are a lot of ARM palmtops and many run Debian, and benefit for instance from the security updates and stuff...

  8. I disagree wholeheartedly. on Debian Release Mgr. Proposes Dropping Some Archs · · Score: 4, Insightful

    Someone at Debian is finally getting a fucking clue. I've been telling stupid Debian zealots this for years... your distro is dying because everything has to move in lockstep.
    Interesting, from where I am it seems to be pretty much alive, thank you.
    Take a look at the Linux kernel -- it's x86, and yet there are loads of ports which move at their own speed. Debian is a slug of a distro because it moves at the speed of the absolutely *LEAST* developed port.
    There is always sid.
    Split them off focus on the x86 distro... and let the other catch up or die off.
    And then the only thing that sets Debian apart from the other distros (quality, determined by lots of portability issues spotted, bad code spotted this way, lots of different archs using the same distro, etc. will be dead. People will just use Ubuntu, if they want to use something x86-ppc only.
    Debian is smothering... and all the puffed up insane zealotry about how other platforms are supported just as well as x86 is worthless if your distro is 5 years out of date.
    Interesting, I run Debian, with kde 3.4 over kernel 2.6.10 and my distro does not feel 5 years out of date.

  9. Some remarks. on Finding the Pits In CherryOS · · Score: 1
    The header is a work of intellectual creation when its author writes it. Thus, it is and remains copyrightable art.

    There lies your first mistake. A program is copyrightable art, but a header file is considered (at least by USofA's case law and by Brazilian Computer Programs Act) uncopyrightable. Brazilian law makes a special exemption that normally applies to header files: they are, as a rule, the definition of an interface (what I called API in the course of this conversation with you) that admits little variation of expression is not copyrightable.

    Hence, for instance, errno.h:
    // ...
    #define EDEADLK 35 /* Resource deadlock would occur */
    #define ENAMETOOLONG 36 /* File name too long */
    // ...
    cannot be expressed much differently, so, it's not copyrightable.

    In no moment I said, under a blanket, that the output of a program cannot be copyrighted. I said the output of a program is not per se copyrightable. Obviously, if it's a transliteration of a copyrighted work, it is for the the purposes of copyright law, the same work. Mind you, what I said is that automated-automatable works are not copyrightable, but if the input to an automated work is copyrighted, the output is just a transliteration of the same work, so it's under the same copyright.

    When my program #includes errno.h, this (and errno.h contents -- which are uncopyrightable) is just an instruction to the compiler/preprocessor of how to transliterate (automatedly transform) the work. The work is my (source) program. My binary program is just another aspect of my work. Get my source program, the same toolchain, compile... you have a copy of my binary program. This is very different than a derivative work.

    A translation, for instance, is a derivative work, that involves transformation of the original work and the transformation is, per se, a work of the spirit. Which is why we do have a lot of crappy translations in our .po files :-) A good translation is made by a specialized professional that is skilled in the art.
  10. And that is the problem I have... on Finding the Pits In CherryOS · · Score: 1
    With your interpretation... That work is textually included in compiling the driver...

    You know, compiling something is not an act of intellectual creation... it's just a step in an automated process, that is, copyright law does not kick in this particular step. So, when I write (pseudo-dd-code):
    1 #include <linux/pci.h>

    3 int my_device_init(struct dev_t *d, struct bus_t *b) {
    4 dev_t *e = pci_enum(b, NULL);
    5 while( e && e->device_type != 0x04747 )
    6 e = pci_enum(b, e);
    7 if( e )
    8 outp(e->port, 0x04747);
    9 }
    I am not, in any moment, pulling in to my work "linux/pci.h" nor the definitions of the structs dev_t and bus_t. I am referring to them, and this is what case law already permits. The step that "includes" effectively the bits you mentioned is an automated step, and is not made by me, it's made by cpp.

    And no, you cannot copyright the output of a program because it's never an intellectual novel creation (Brazilian copyright law -- and I'm sure others based in Berne's convention -- mentions explicitly that "protected works are those that are creations of the spirit).
  11. Drivers needing more than API to work on Finding the Pits In CherryOS · · Score: 1

    Do you care to elaborate on this, Bruce? What textual inclusion of non-interface material is required by the driver to work?

    Regards,
    Massa

  12. Sorry, Bruce, but you are wrong. on Finding the Pits In CherryOS · · Score: 1

    I want to start this comment stating that I have nothing but deep admiration for you and your work. So, I am not flaming.

    But the fact is: when you link only to "EXPORT_SYMBOL" symbols in the Linux kernel, there is a promissory estoppel from the person that wrote the exporting module, saying: "the use of the entity (function) referred by this symbol does not constitute evidence that the importing module is a derivative work of the exporting module; as a matter of fact, I, the exporting module author, chose to make this 'EXPORT' and not 'EXPORT_GPL' because I promiss any importing modules that they won't be held liable... any use of the referred entity (function) must be removed in the Filtration phase when determining the derivative status by Abstraction-Filtration-Comparison."

    That IS, after all, the whole point of 'EXPORT_SYMBOL_GPL'... to document that some symbol refers to a function that is intrinsecally connected to the "intimate parts" of the kernel.

    Now, AFAIK, the nVidia binary kernel module just uses EXPORT_SYMBOL symbols. Furthermore, AFAIK again, their driver is a derivative on their proprietary Windows driver, just transformed to use the linux driver API instead of the Windows driver API.

    And, to boot, APIs are not copyrightable, as someone else already stated.

  13. Brasil doesn't. on DrinkOrDie Warez Trader to be Extradited to U.S. · · Score: 1

    Extradiction of nationals is forbidden by our Constitution.

  14. It is a Political Correctness thing... on DrinkOrDie Warez Trader to be Extradited to U.S. · · Score: 1

    If you were a shoplifter, would you like to be called a rapist? Even if the populace in general started calling all shoplifters "rapists"? And if this happened because some Sears-CEO showed on TV saying "those shoplifters are raping me, they are rapists"?

    That is the question. Calling copyright infringers "pirates" is a marketing ploy (brought to you by your *AA and other copyright-holder-associations in the world) to make other people think: (a) that copyrights as implemented today is a perfect system and (b) that people who infringe copyrights are the worse scum on Earth.

  15. Re:HP innovation! on HP Introduces New Technology to Save Mobile Battery Life · · Score: 1

    how do you activate transparency in all inactive windows? (I am running k3.4 too)

  16. You're wrong. The difference... on Militants Planned Attack On Indian Software Firms · · Score: 0, Flamebait

    is that USofAns think "oh, outsourcing! the end of the world! lost American jobs!" and the rest of the world thinks "globalization is about if you send McD's to India, suits that you have to send some jobs too."

    Only USofAns are bigots about outsourcing. Nah, not only them, but they are the worse bigots IRT outsourcing.

  17. To be precise, on MS-DOS Paternity Dispute Goes to Court · · Score: 1

    America was already an independent continent soon after it got loose from Eurasia and Africa some millions (billions?) of years ago.

  18. The GPP on Cox on Torvalds and Linux Kernel Development · · Score: 1

    summarizes the text you referred to with both simplicity and precision: "Gandhi had his moments of pettiness and just plain tom-foolishness, but the sum of his efforts changed the way people gain power back from those who would usurp it for their own."

  19. Care to elaborate? on Cox on Torvalds and Linux Kernel Development · · Score: 1

    I think, you don't know much about Gandhi. (sic)

  20. Not to mention... on ClearLooks to be Default Theme on Gnome 2.12 · · Score: 1

    that this theme is not a lot different than the old one... it is different, mind you, just not a lot.

  21. Even with "%u"? on KDE 3.4 RC1 Released · · Score: 1

    Anyway, in T-bird you would have to set its default handling of URIs too... (T-bird would not use the setting in kcontrol)

  22. You miss the point on John Gilmore's Search for the Mandatory ID Law · · Score: 1

    The fact is that, historically, the position earned as "representatives" (head) of State by the Crown is derived from the fact that they owned (as in they were proprietors) all the country at some point in time.

  23. I have never been to the USofA on John Gilmore's Search for the Mandatory ID Law · · Score: 1

    But when I was in the Antilles, 50-70% of all cars I saw were automatic.
    In Europe (I lived for a year in Spain) 30%.
    Here in Brasil, less than 1% of the cars are automatic.

    There is no special habilitation "automatic cars only" class.
    If you can only drive an auto car, a manual gear is very difficult to handle.
    So, people cannot make the exam in automatic cars.

  24. Component Chooser on KDE 3.4 RC1 Released · · Score: 1

    don't forget to put

    sensible-browser "%u"

    so it starts firefox with the URL and not with the kio-supplied filename.

  25. Nazi aliens on TrekUnited Reports Mission Successful at Trek Rallies · · Score: 1

    I had completely erased those from my memory... :-) I was considering them as being the worse 3rd season ones :-)

    But they brought an end (albeit a non-sensical one... *) to the temporal sh*t.

    * = if the temporal cold war has re-set, shouldn't Enterprise still be in drydock instead of flying the Klingon home?