Mark bond wrote:
Henning Nielsen Lund wrote:
Mark bond wrote:
[snip]
- We do not know how much work it will take to port OOo.
- More targets will slow down the porting - I for one will not
support 4 different OS at a time - take a look at AmiZilla/MorphZilla, I am almost sure that there would have been more people working on it, if only OS4 (or MorphOS) would have been a target. 3. We should not use MUI or Reaction. OOo has its own graphical toolkit. And if not, would I say that we should use Reaction, as this is the official toolkit for OS4 - else, people would need to pay money to be able to use our work with its full power (and did MorphOS not lose MUI in the "genesi does not pay it's developers" case?)
Lots ;-)
No comment
But surely said toolkit will need "translating" to the amiga version,
as such it would make sense to use MUI/Reaction so it can take advantage of the global prefs for these two (no they lost Ambient as I understood it).
I will point out I was using htat as an example, just because we are not supporting other platforms, doesnt mean we cant choose to port things in a particular way to make it easier for it to be ported to the similar systems. Or even that we have to consider this, as I've said previously its something that needs discussing. When it comes right down to it, I use OS4, the only version I want is an OS4 version, but if making the AROS/MOS porters lives a bit easier is what it takes, to get the thing done quicker, then I say at least consider it.
Mark
All- 100% in agreemenet with OP at least with respect to #s 1 and 2. OOo is a huge effort, and it needs to not be 'distracted' by porting to multiple platforms. OO is _beyond_ unlikely to run on the 68k platform anyways.
Having said that, we should take reasonable steps to not make re-use of some of the code difficult for future porting efforts, whether it's to MOS, AROS or whatever.
The bigger the project, the better defined the focus needs to be for success, whether it's in breaking the project down into manageable chunks, or limiting initial scope.
3. RE: translating toolkit. Again, we should make reasonable efforts to allow re-use here, and a _second_ or child project could always come in and port our toolchain and other bits to other (68k, OS3.X, MOS?) platforms...then we merge the changes back in if they do in fact still work for all platforms at that point. Actually, this would be better to do only once the entire toolchain and supporting tools have stablized, but then someone is welcome to do it..but not as a primary focus.
Scott