3 ms·
"No breakage" is possible in a Lisp dialect. With each rev of the dialect, put all new functions and variables into a new namespace like "std2019:...". Old co
by kazinator 8y ago
"No breakage" is possible in a Lisp dialect.
With each rev of the dialect, put all new functions and variables into a new namespace like "std2019:...".
Old code that doesn't know about new features doesn't run into them at all.
Keep the old namespaces untouched: old code finds everything it expects.
There is no bullshit like reserved keywords: all symbols are namespaced in packages. New syntax is just macros named by symbols.
The only kind of breakage left is system requirements (new tools don't target old hardware well due to larger footprints; new dev environments don't run on old dev hardware; implementations drop support for unpopular platforms).
- coldtea 8y agoThat's only if you make superficial changes, and is already possible in non-Lisp languages. What about changing core semantics though?
- kazinator 8y agoSuch as what? Say, change strict evaluation to lazy? Core semantics can be wrapped in an operator that is in some namespace. We develop (lazy ...) and everything in it is lazily evaled. File-level annotations are possible. Emacs Lisp introduced lexical scope instead of dynamic scope as a file-level option.
- lispm 8y agoIn Symbolics Genera stuff was originally implemented in ZetaLisp with Flavors. Then there is CLtL1, Symbolics Common Lisp (a rather large version of Common Lisp), ANSI CL, ... There is quite a non-trivial difference between ZetaLisp and Symbolics Common Lisp: both a very large Lisp dialects. ZetaLisp is dynamically scoped, has base 8 integers, the old flavors system is widely used, Fexprs, ... All these Lisp dialects are provided in the same system and one switches the reader/printer and the packages when changing the language: on can set the language context in a listener and also per file.