Future ....

[Dirk Hohndel]

On Tue, 25 Oct 2011 19:13:00 +0200, Ďoďo <dodo.sk at gmail.com> wrote:
> I would like to know where is the development of subsurface heading to ....

Where the developers take it. Seriously, there is no Five Year Plan or
anything. For me it's mostly things that I notice that don't work or
that I want to work differently. So all the recent work on multi tank
dives and interpolation of air consumption for tanks with no pressure
data or incomplete pressure data is basically a result of diving with
twins and a stage and a dive buddy who's computer at times stops
receiving pressure data from the transmitter...
 
> During the weekend I implemented air consumption curve based on
> average air consumption and I found out now, that something similar
> was recently added.

Oops - send patches early and often; that's the best way to avoid things
like this. Or tell us what you think you'd like to work on. John Van
Ostrand for example told us a while ago that he was going to work on
manually entering dives and sent a first cut of the code a few days
ago. That tells me that I'll stay away from this area for now.

My immediate todo list includes context menues in a few areas (in the
dive list in order to be able to merge dives, in the profile in order to
make changes to events).

Linus just opened the Linux 3.2 merge window, so my guess is he will be
a bit distracted for a couple of weeks. :-)

> So, I do not want to do parallel unnecessary programming ...

No one does. So please, tell people here what you are thinking of doing
and send code once you have something that works...

BTW: the other thing I'm working on is getting more binaries built in an
automated fashion. I have Fedora and OpenSUSE builds at
subsurface.hohndel.org. We have a Gentoo ebuild from Bearsh. Roland
Dreier has mostly completed a Debian build and David Woodhouse has
continued my early efforts to get a Windows build going - but instead of
my native approach he's doing so with the MinGW cross tools which is
just brilliant as it means we should be able to produce an installable
package in OBS as well 

The one platform where I don't know if/when we'll get an installable
binary package is Mac, but Henrik Brautaset Aronsen has at least shown
that it's easy to build from sources, so all it takes is someone who
knows how to assemble an installer from that.

/D