Timeline for Licence of open source software re-implementation
Current License: CC BY-SA 4.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 15, 2022 at 0:30 | comment | added | bryc | At some point, this becomes a very grey area. Even "Clean room" involves uncovering third party works. Whether from raw transistors, pulses over a wire, binary data, disassembly output, or decompiled bytecode, you're piecing together someone's original work. Studying C code is no different, sometimes you need to clone functionality in ways not comparable to a straight logical port, either by verifying an algorithm (which cant be copyrighted), or a complete rewrite/reimplementation with well-defined differences. It's just whether you are comfortable parking in a grey area. Just use attribution. | |
| Jul 26, 2021 at 22:38 | comment | added | supercat | In many cases, chip manufacturers will assign names to registers within their chips' published datasheets/reference manuals. If a GPL header file defines constants for addresses and bit masks using names published in a manufacturer's datasheets, any copyright interest that might exist in those names would belong to the chip manufacturer, and publication in a datasheet would almost certainly be construed as granting a license for people to use those names in writing software for use in that manufacturer's chips. | |
| Feb 7, 2020 at 13:54 | comment | added | vonbrand | @E.Körner the "compatibility" table says if you can combine pieces you got from somebody else and dostribute the result. Only the owner of the code can distribute under a different license. | |
| Jan 10, 2020 at 11:46 | vote | accept | E. Körner | ||
| Jan 8, 2020 at 17:26 | comment | added | E. Körner | Thank you, I thought so too but hoped otherwise. I will search a bit or even post a question. The wording was not really that clear and more about other software that is depending on a GPL licensed software if I understood it correctly. -- I will wait til the end of the week to accept your answer. Maybe some other replies will come but if not, yours was really helpful. | |
| Jan 8, 2020 at 15:46 | comment | added | planetmaker | I find that table colouring a bit mis-leading and I think the answer is 'no'. The corresponding matrix says in those cases "combination (of GPL and LGPL) is ok under GPL". | |
| Jan 8, 2020 at 15:29 | comment | added | E. Körner | According to this gnu.org/licenses/gpl-faq.html#AllCompatibility - would that mean I can also use the LGPLv2* licences? | |
| Jan 8, 2020 at 15:02 | comment | added | E. Körner | Thank you. A clean-room implementation at this point is impossible and unpracticable as I would have to test even more to have it working correctly. There are parts written by me that are completely new (e. g. config parsing, device acces via USB), some are heavier modified, but I had to look at the original source to understand what choices were made for the given results. One reason for the 'rewrite' is also that I needed it in Python with additional functionality. Trying to get it working with the original language "C" and using it would have been even more time-consuming. | |
| Jan 8, 2020 at 14:43 | history | edited | planetmaker | CC BY-SA 4.0 | added 387 characters in body |
| Jan 8, 2020 at 14:36 | history | answered | planetmaker | CC BY-SA 4.0 |