On 30 November 2012 15:26, Stephen Montgomery-Smith
<address@hidden> wrote:
And since
Matlab has done such a fantastic job with their documentation, I don't see a
need to reproduce what they have done.
We don't need to reproduce it. We need documentation *for Octave* not
for Matlab. Sometimes the Matlab documentation covers Octave, but this
only happens for the case of functions that are exactly like Matlab.
Most Octave functions have at least some slight discrepancies with
Matlab, sometimes intentional, sometimes not. It's no use for Octave
if the Matlab documentation says the functions do this or that but the
actual Octave function doesn't do match the Matlab documentation.
Matlab's documentation also obviously doesn't cover Octave features
that don't exist in Matlab, such as lsode, double quoted strings, or
broadcasting.
Moreover, Matlab's online documentation changes or disappears without
a trace or is simply wrong or incomplete. We have the full history of
Octave documentation and we can change and modify and amend Octave's
documentation. If there's a problem with Matlab's documentation, on
the other hand, there's nothing we can do about it other than beg and
plea with TMW.
There's an obvious advantage to having good Octave documentation,
please don't dismiss the problem. Free software requires free
documentation, and we are not free to distribute nor modify Matlab's
documentation.