On May 28, 7:28 pm, rjf <[EMAIL PROTECTED]> wrote:
> I think my Tivo DVR runs Linux, and in principle I have access
> to the source, but I would no more consider trying to fix a bug in it than I
> would remove my own appendix.

This is more a problem of commercialization and being a closed system.
There are DVRs/recievers that are Linux based, open for installing new
software and have their own community of people who develop their own
code. So, this is in the wild, just not by TiVo!
The conclusion for me - about that and the current topic - is, that
you not only need a public source code, but also the community, the
ability and tools to develop, (in that case also an open hardware),
howtos, and yes, basically encuragement for those who want to write
software on their DVR and so on. i.e. I know that there are RSS
readers for DVRs with LAN... that's geeky, but cool ;)

About proofs: I think, the current state of understanding software and
hardware is just not mature enough to do this. As mentioned above,
this is a chain of respoinsibilites. It can't be a serious requirement
to proof basic cpu commands for doing abstract algebra - maybe every
time again for each command and each software package. This needs
layers and a new understanding what's going on. My (prbably naiive
understanding) is, that there are much more problems mentioned than in
reality. e.g. if you proof a statement X in pure mathematics, you
don't have to make sure that it is printed correctly in every place in
the universe - it just has to be correct in some virtual space of
imagination! (think of typos, ink problems, ...) The same holds true,
i think, for hardware flaws. Someone just has to trust into the
hardware to write correct algorithms. Also, the software itself is
layered: system libraries, kernel, management tools, APIs, ... It is
impossible to have to proof them just for using them. Therefore, only
the actual implementation using that whole stack should be checked...

At that point, maybe some time in the future, some abstract mapping of
code lines to logic with inferencing for correctness and back to code
will exist? To transform code into something canonical to analyze? I
don't know.
Also, I heared that the language Fortress want's to be close to typed
math (2d formula rendering in source code ...). Maybe that's a first
step towards closeing that gap between pure math and software
implementations?

H
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---

Reply via email to