On 28 Jan 2010, at 15:20, Ron Jensen wrote: > Actually, I disagree with this statement, and it represents a > fundamental shift in attitude from the way I've seen flightgear's > development progressing over the past year or two.
It wasn't intended to be anything so fundamental, and I'm surprised to see it described that way, see below: > - Hard-coding every instrument in C++ instead of nasal means only > developers following/building the latest cvs head code get to use > whatever until the next release cycle. Absolutely, but that's a release-cycle problem, not a Nasal vs C++ one. This becomes a cyclical problem - I'd like a faster release cycle, so more aircraft developers used a current release, so there was more testing of current code, so more aircraft worked with CVS, so CVS was the most usable version, etc, etc. > - Hard coding every instrument/flight control in C++ means my WW-II > storch (et.al.) is stuck with an autobrake functionality it doesn't have > nor need. Indeed, but the overhead for the autobrake is extremely minimal, and a C++ interface was the easiest way to implement the required control behaviour. Arguably the real solution is a much more extensive input interface on the Nasal side, between the input driver and the FDMs, but that simply doesn't exist right now. > - The pool of people with commit rights to C++ code is very, very small. Sure, but generally such changes are fairly straightforward - again largely because most of this data already exists in C++ land. I'm definitely *not* suggesting to do everything in C++, what I would ideally like to see is most computations taking place in the C++ side, and Nasal providing the behavioural control. The Navradio flag is a good example - only the instrument designer knows what changes are appropriate based on the flag, but to have each aircraft encode the range of ILS frequencies seems unfortunate, at best. As Curt said, for each specific case, there can be a discussion - Nasal is great for plenty of things, and C++ is great for plenty of other things - and there is a grey area where it's something unclear which is better. What I'm most concerned about is that where the C++ coders can trivially make some information available via a property, new nasal function, or similar, people on the Nasal side at least take the time to ask, if it can be done that way. If that means Nasal developers are demanding a software release so that new properties / functions are 'available' to more people not using CVS, well, personally I consider that a good thing :) Regards, James ------------------------------------------------------------------------------ The Planet: dedicated and managed hosting, cloud storage, colocation Stay online with enterprise data centers and the best network in the business Choose flexible plans and management services without long-term contracts Personal 24x7 support from experience hosting pros just a phone call away. http://p.sf.net/sfu/theplanet-com _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
