On Wed, 25 Nov 2020 22:49:32 +0100 Mattias Andrée <maand...@kth.se> wrote:
Dear Mattias, > Concerning farbfeld, it is quite a different thing to create a new > simpler standard than supporting an already existing but complex > standard. Farbfeld was a good first step in moving towards simpler > image formats, although even if would have got exceptionally good > traction it would take quiet some time before support for older > complex formats could be removed, and in the meantime support for an > additional, but simple, format would be required which would add code > (not complexity) to image format libraries. I think this is forth it > in the long run, and it gave us a good standard format for programs, > that don't need to support user provided files, to use, better than > the netpbm formats. yes, well put! It would be different if I extended the NetPBM format with new things that would be incompatible with all other NetPBM-decoders. Of course I wouldn't be able to do it, but the absolute dominance of GNU tools in the 90's and 2000's has given them this opportunity and they seized it. I sometimes wish I would've started programming in the 90's and been able to influence the ecosystem with suckless software. Imagine musl being published in the 90's; it would've been revolutionary and we wouldn't have to deal with the millions of lines of C-code infested with GNU extensions. With best regards Laslo
