Re: [Axiom-developer] Re: OpenAxiom-1.2.1 released

[Scott Morrison]

I actually sent that email prematurely last night, hence it ended with a colon.

I said that Boot's pattern matching allowed us to get by without structured data for the whole project.  I meant to follow that up, with "but that's not a good thing."  The interpreter exposes it's internal representation in a very bad way.  If you want to change the data structures the interpreter uses, you need to find all the places in the code which use pattern matching to destruct data, and change them.  I did this several times and it is no fun at all.

Any modern language uses structured data to hide those sorts of implementation details.  Tim's project to rewrite the system in CL is a very good thing if he replaces Boot pattern matching with structured data.  The important issue is hiding implementation details.  To me the language used is secondary.  If Tim wants to use CL, that's fine with me.  If other projects do the same thing with Boot, that would be good too.

Tim's point about the heavy use of global variables in Boot is a valid point.  That makes the code hard to understand and change.  But to me that's not a language issue, but an implementation choice.  Any good rewrite of the interpreter would remove those.  That could be done in Boot or CL.  If Tim likes CL, and he does that work to add structured data, I think Axiom would be much easier to maintain.

-- Scott

On Sat, Apr 11, 2009 at 12:37 AM, root <address@hidden> wrote:

> As Dick Jenks explained it to me when I joined the Axiom project in 1984,

Back in the late 80s I set out to remove boot from Axiom.
It is a clearly stated goal of the Axiom project.
It is happening as I write this.

Tim