#tyndur · 2010-01-16

90 Nachrichten · Seite 1 von 1 · Original-CSV herunterladen

  1. taljethMoin
  2. FreakyPenguinMoin taljeth
  3. XanClicquok: Are you already awake?
  4. quokXanClic: I am awake now!
  5. XanClicHey! :D
  6. quoki've was up most of the night. On-call for work and they kept calling me with "issues"
  7. quokmost of it was "Ignore it, it's fine" and then trying to go back to sleep, only to get called again
  8. XanClicNo problem, I got up 12:50 this "morning" ;-)
  9. XanClicBecause of your contest and the ISC: Someone told me that this is not a public domain license
  10. XanClicThe closest you could get is WTFPL or CC-zero
  11. taljethXanClic: You can dual-license ISC and WTFPL
  12. XanClicI don't plan to license it as WTFPL if I'm not forced to do, I just wanted to say that ;-)
  13. XanClicThough I have no problem in doing so
  14. quokISC isn't "public domain" as you still retain copyright
  15. quokin "public domain" you give up copyright
  16. taljethWikipedia says "functionally equivalent to the 2-clause BSD license". Which is far from being PD, but still quite permissive.
  17. quokbut ISC allows anyone to do anything at all with the code, except remove the copyright header
  18. XanClicWTFPL allows anyone to do anything ;-)
  19. XanClicto do "what the fuck he wants to" ;-)
  20. quokwith public domain, technically I can claim copyright on the code, as the original author gave up copyright
  21. XanClicThat's right, it's your decision ;-)
  22. XanClic(Just wanted to say that because you said ISC to be something as PD)
  23. taljethquok: wtf? You can claim copyright for PD code in US law?
  24. XanClicWTFPL clearly states so *g*
  25. XanClic"Do what the fuck you want to" *g*
  26. quokbecause in US law, PD code is not copyrighted, so you can try to claim copyright for it, yes. and in 99% of cases you'll succeed
  27. quoksomeone would have to notice that you took PD code and claimed copyright for it, and then file a lawsuit over it, but I doubt a court would do anything about it
  28. quokit is slightly different from when copyright on something expires
  29. quokvs it never being copyrighted in the first place
  30. quokor so I've had it explained to me
  31. quokalso I don't like the WTFPL because it contains the word 'fuck' in it. :)
  32. XanClic*g*
  33. XanClicCC0 is available, too ;-)
  34. quokISC is quite permissive, I really don't see anything wrong with it. You keep copyright, but anyone can do anything else they want with the code.
  35. XanClicI don't see so, too, just wanted to tell this to you in case you didn't know ;-)
  36. taljethI have released quite some code under 2-clause BSD, which seems to be basically the same
  37. taljethSo I don't think there's anything wrong with it
  38. quokyeah
  39. quok"A person using CC0 (called the “affirmer” in the legal code) waives all of his or her copyright and neighboring and related rights in a work, to the fullest extent permitted by law. If the waiver isn’t effective for any reason, then CC0 acts as a license from the affirmer granting the public an unconditional, irrevocable, non exclusive, royalty free license to use the work for any purpose"
  40. quokthat's from http://wiki.creativecommons.org/CC0_FAQ
  41. quokso in countries that don't allow you to give up copyright, CC0 is basicaly the same as the ISC / 2 clause BSD
  42. taljethNo, it doesn't even require to retain to keep the copyright notice
  43. XanClicbtw, quok, I may safely assume that the "hidden sectors" value in that FAT BPB is equal to the LBA of the boot sector on the medium?
  44. XanClicBecause my mkdosfs simply set it to zero...
  45. quokXanClic: I will use mtools to generate my testing images. mtools is aware of paritioning, and I do believe that the hidden sectors is equal to the number of sectors before the boot sector on the medium
  46. quokbut I'd have to re-read the FAT spec to make sure of that. :) I'll likely do that tonight
  47. XanClicAs I read on osdev.org or wikipedia, it was just that value
  48. quoktaljeth: ah, you're right, it doesn't require you to keep the copyright notice. Oh well. Minor difference. :)
  49. XanClicosdev.org says: "Number of hidden sectors. (i.e. the LBA of the beginning of the partition.)"
  50. XanClicAnd WP says: "Count of hidden sectors preceding the partition that contains this FAT volume. This field should always be zero on media that are not partitioned."
  51. quokand the MS spec says? :) I think the MS spec agrees with WP
  52. XanClicIt's the same ;-)
  53. XanClicNo difference if I say "number of sectors preceding the partition" or "the LBA of the first sector of that partition"
  54. XanClicbtw, I had problems with mtools on creating a 4 GB medium...
  55. quokIMO the way it is stated on wiki.osdev.org is a bit ambiguous
  56. XanClicIt simply created FAT32 and when I forced it to use FAT16, it complained about that
  57. quokyeah, MS has "rules" on how to create FAT filesystems. Over 4MB or someting like that, they say to use FAT16 instead of FAT12
  58. quokand over 500MB to use FAT32 instead of FAT16. Mtools will do what you tell it to, but will complain
  59. XanClicBut it complained and didn't do anything ;-)
  60. quokoh, maybe I hacked my mtools then? :) or I used apple's disk utility to do it...
  61. XanClic*g*
  62. quokI need to get back to work though. Doing after-hours support sucks :(
  63. quokif I get time today I'll generate disk images and post them for people to test against.
  64. XanClicHm, unzipping 8 GB of data takes a lot of time, doesn't it? *g*
  65. quoknot if it's a sparse image :)
  66. XanClicSeems to be Mac OS X related...
  67. XanClicOh, got that know: Don't think I didn't got 4 GB FAT working, I used mkdosfs as stated earlier ;-)
  68. quokI just prefer mtools over mkdosfs for updating the bootsector code. it's much simpler :)
  69. XanClicI wrote a shell script for that so it's not a big problem ;-)
  70. quokif you have the room in your code, you should calculate if it's fat12 vs fat16 by the sector count rather than relying on the filesystem type id thing
  71. XanClicI know
  72. XanClicBut I don't have that space ;-)
  73. quokbut for the purposes of this competition, since requirements are quite tight, I'm going to allow relying on that silly field :)
  74. XanClicthx :-)
  75. quokI was going to make it a 1024 byte contest, but then you need even more arguments to special formatting tools like mformat and mkdosfs :)
  76. quokand messing around with reserved sectors and such
  77. XanClicAnd that would be too easy, too ;-)
  78. quokbut 1024 bytes also makes it too easy, IMO
  79. quokanyone can right a bootsector that does all of this properly in 1024 bytes. With that much space, you should be able to locate the file ANYWHERE on disk, and load it at or above 1MB as well. :)
  80. XanClicI agree to the latter one ;-)
  81. quokas well as test for and use EDD extensions when available, heh
  82. XanClicThough I think only my FAT capabilities prevent me from doing the first thing
  83. XanClicWhy not also loading ELF64 files? *g*
  84. XanClicAnd executing them in LM, of course
  85. quokin 1024 bytes? probably can do that :)
  86. quokalthough I personally allow my stage2 to do all of that. stage2 runs in 32bit mode and then will initialize LM and jump to it before executing the ELF64 file
  87. quokso my stage2 will do ELF64 files but they must be fully located in the first 4GB, which I think is reasonable
  88. XanClicI don't do anything with LM, but I think this model is often choosed, especially when using GRUB (if you don't plan to patch it) - theoretically I collaborate at an OS project, which uses this one, too
  89. quokit also will only identity map the first 4GB, so the kernel would need to fix that ASAP
  90. quokanyway, I'm out of here for a while. have things I need to get done.