From: cross@spitfire.i.gajendra.net   
      
   In article <10h3upk$3f20l$1@dont-email.me>,   
   Craig A. Berry  wrote:   
   >[...]  But with people using scripting   
   >languages to process massive vectors to train LLM models, it all seems   
   >pretty puny.   
      
   This is a good point, but, I sometimes wonder if, perhaps, we   
   need to recalibrate what we mean when we say, "scripting   
   language."  I imagine that you are referring to Python here, as   
   that seems to be the thing that the kids are all hip on these   
   days when it comes to model training and such-like, but I think   
   it's fair to say that that language has grown far beyond   
   traditional "scripting" use.   
      
   Python is interpreted, yes, but people who are using it to do   
   numerical analysis are often using the jit-compiled variant, and   
   more often the actual heavy computational lifting is being done   
   in a library that's exposed to Python via an FFI; so the actual   
   training code is in Fortran or C or some more traditional   
   compiled language.   
      
   There is, evidently, a need (or at least desire) for a really   
   good interpreted language to script various system management   
   tasks; I gather folks feel that DCL is a bit long in the tooth   
   and insufficient for that.  As I mentioned before, I feel like   
   that is qualitatively different than using DCL as an interactive   
   CLI; perhaps the solution here is just to build out a really   
   nice set of officially supported modules for, say, Python (or a   
   similar suitable language) and call it a day.   
      
   	- Dan C.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)