scope of the project

[Olivier Binda]

webP looks like a nice improvement over jpeg.

Yet, to have my interest as a developper/to gain market traction, it
should have :

1) a very liberal license (BSD/copyleft/public domain...definetely not
GPL) so that it can be used by everybody
2) be usable on most computing platforms, (not only Unix/os X/
windows) including embedded/low resource platforms, arm/x86/x64,
etc...
3) be usable with most languages. I;E. It probably should be a C
library as nearly everything/every language can link C code.
4) be usable in other apps than browsers (if it is only supported by
browsers, it will fail miserably)
5) have technical advantages against the competition in term of size/
quality/speed to compress/decompress


Now, developpers/users would welcome a new image container too that
would


6) allow the efficient processing of most images : this means that the
container should be able to cope with lossless or without loss images
7) have modern features that are expected about images : this means
alpha channel, different color sheme, scaling, etc..
8) provide an uniform/iunified api that allows the treatment of images

9) generally be better/simpler to use

People writing apps out there must already support .png, .bmp, .jpeg,
(sometimes) .gif...and now .webp
This sucks. Those are just goddamn images. We should just have to
support 1 image container, through a high level api,
and get the low level api sort of which compressor/decompressor/format
to use.

If you manage to provide this, you'll be succesfull.
If the scope of this project isn't big enough... you'll probably fail
and bring more fragmentation

[Olivier Binda]

I forgot that it should be

10) be extensible
11) forward compatible