Umm .. why does this list only contain clones of commercial softwares? I would rather say that high priority projects are the linux kernel, git, myriad python and ruby interpreters etc which are widely used and do contain some design innovations.
I also feel it a bit disappointing that there were so many "replacements" on this list.
How about something completely original? How about we start racing the commercial sector to a worthwhile goal like computer vision, or decent speech recognition. Anything more inspiring than a "Skype replacement" will do fine.
We just need a few more things where the open source version is the dominate player, instead of a clunky, half working "free replacement" for commercial software that wasn't very good to begin with.
True. Octave is on the list. As mentioned in the Reddit thread though:
http://www.fsf.org/campaigns/priority.html
A lot of companies need to use Matlab because of the large codebase out there that is written in MatLab.
While Octave is good numpy/scipy/matplotlib are also free (in both meanings of the word) and great software.
There is also Sage.
So get started writing machine learning libraries in Python!
That is a strange list. There seems to be a misplaced sense of priorities, which is somewhat alarming. The focus should, as you say, be on building platforms/libraries which can be used to build apps and services, either free or otherwise, instead of just focusing on providing free alternatives to existing commercial software
