On Dec 29, 2006, at 4:14 AM, Luc Heinrich wrote:
On 28 déc. 06, at 00:13, Kevin Ballard wrote:
http://www.opendarwin.org/pipermail/darwinports/2002-October/ 015354.html
In my very own and personal opinion, Tcl is the single reason why MacPorts doesn't get as many external contributions as it should and is therefore so awfully slow to evolve. I have tried to dive in the MacPorts sources many times, read the Welch book, yadda yadda, and I still can't get past 3 or 4 lines of Tcl without giving up in disgust. Ok, maybe that's just me but still...
Hey Luc, Yeah, Tcl is a bit hard to get used to. I had to learn it in order to make any progress on the MacPorts code. I have to admit that once I got into the mind set, I found it not so difficult...but it is different from many other scripting languages, though probably not so different as something like ruby or python. I agree that Tcl has held MacPorts back some. Another problem is just a lack of people with the time or inclination to work on base. I'd encourage anybody so incline to speak up and get involved. If there's enough mass, perhaps there's also some energy to do MacPorts 2.0 in ruby, or pyhon, or C, or... ;) And to make major steps forward in doing so. On "problem" with MacPorts at the moment is that it works "pretty well", which means that there's not as much energy to improve it, since it does almost all the job almost all the time. It's the more exceptional cases that bring in the difficulties. James