Am 18.10.2009 um 16:04 schrieb D'Arcy J.M. Cain:
On Sun, 18 Oct 2009 00:22:11 -0400 Thor Lancelot Simon <tls%panix.com@localhost> wrote:I have to say, I like Python as a language for many purposes. But as an embedded scripting language, I don't like Python. It is very, very hard to cross build, its runtime is huge, and its maintainers have basicallyno self discipline, changing the API and ABI at the drop of a hat forCan you support that statement? My experience has been the opposite. Within the 2.x series everything has been backwards compatible,sometimes painfully so. The major changes are in the 2.x ==> 3.x changefor which they are giving people years to adopt and including tools to make the migration easier.I don't want to see this become a language war - I think the discussionshould focus on whether we need a scripting language in base, not what it should be - but the above statement seemed very unfair given theamount of wotk that Python developers have done to make upgrades happenas smoothly as possible.
I second that. As a long time python user (I think ten years or so) I did never run into substantial trouble with the language, and I created some fairly big pieces of software with it.
otoh, I would like to point out that the topic of this discussion is actually way more focused than if we want a scripting language in base. The question is "Do we want Lua in base". We don't need a programming language that can cover all aspects for informatics, we have C and C++ for that already. What I need is a reasonably small, fast, embeddable language that can be used to extend software.
There are options. I could pack Lua to my software locally, then only my software would "benefit" from it. Or we could add it to the system so that others can use it, too. So this discussion is about finding out IF and HOW we want to add Lua.
Because the language is so powerful, I think it will be benefitial to users of NetBSD, especially those that are not so technically skilled, but want a solid foundation for their applications. But this is a pure hypothesis, of course.