Lambda the Ultimate

If you have a language with extensible records (like those of Gaster and Jones), and you view functions like a -> Maybe { r } for some row r as patterns, are there practical obstacles that prevent you from extending the environment of a case alternative with the row of the result type of its controlling pattern?

I would appreciate pointers to anywhere that this is discussed in the literature, or an explanation of why it is a bad idea. Google hasn't been especially helpful.

-- concrete syntax for records is entirely made-up

wildcard_pattern = const Just {}

-- without first class labels, the compiler would have to make a whole family
-- of declarations like this, one for each variable bound by a pattern
variable_pattern_x val = Just { x := val }

-- same comment applies
as_pattern_x pat val = case pat val of
                         Just rec -> { x := val, rec }
                         Nothing -> Nothing

-- I'm not 100% sure that this handles laziness correctly, but you get the idea
-- declarations like this would be automatically generated from the datatype declarations
cons_pattern pat_head pat_tail [] = Nothing
cons_pattern pat_head pat_tail (x:xs) = case pat_head x of
                                          Just rec_head -> case pat_tail xs of
                                                             Just rec_tail -> record_join rec_head rec_tail
                                                             Nothing -> Nothing
                                          Nothing -> Nothing

nil_pattern [] = Just {}
nil_pattern _ = Nothing

I was just forwarded a message that Peter Landin passed away yesterday.

From: Edmund Robinson
Date: 4 June 2009 09:10:11 GMT+00:00
Subject: Peter Landin

I am very sorry to inform you that Peter Landin died yesterday of natural causes.

For those members who are several generations away from Peter's early contributions, he was one of the major figures in the UK at the time that Computer Science was beginning to establish itself as a discipline. Some of his papers from 40 years ago are essential reading for any serious student of programming languages as still the simplest and clearest exposition of ideas that remain fundamental. The ideas in his papers were truly original and beautiful, but Peter never had a simplistic approach to scientific progress, and would scoff at the idea of individual personal contribution. Some of his own greatest ontribution to the field was as part of a golden nexus of work on programming languages in the UK in the late 60's and early 70's, containing Dana Scott and Christopher Strachey and others as well as Peter. The ideas they developed through their discussions truly lifted the study of programming languages to another level, and are now part of the bedrock of the subject.

Landin was one of the founders of our field, and did a lot of work of lasting value. We've discussed his papers here many times before, even though some of them were written decades ago. He did work that cast a long shadow, or phrased better, did work that illuminated wide vistas.