Paper/Paper.thy
changeset 1493 52f68b524fd2
parent 1491 f970ca9b5bec
child 1506 7c607df46a0a
equal deleted inserted replaced
1492:4908db645f98 1493:52f68b524fd2
     1 (*<*)
     1 (*<*)
     2 theory Paper
     2 theory Paper
     3 imports "../Nominal/Test"
     3 imports "../Nominal/Test"
     4 begin
     4 begin
       
     5 
       
     6 notation (latex output)
       
     7   swap ("'(_ _')" [1000, 1000] 1000) and
       
     8   fresh ("_ # _" [51, 51] 50) and
       
     9   supp ("supp _" [78] 73) and
       
    10   uminus ("-_" [78] 73) and
       
    11   If  ("if _ then _ else _" 10)
     5 (*>*)
    12 (*>*)
     6 
    13 
     7 section {* Introduction *}
    14 section {* Introduction *}
     8 
    15 
     9 text {*
    16 text {*
    20 
    27 
    21   Contributions:  We provide definitions for when terms
    28   Contributions:  We provide definitions for when terms
    22   involving general bindings are alpha-equivelent.
    29   involving general bindings are alpha-equivelent.
    23 *}
    30 *}
    24 
    31 
    25 section {* A Brief Overview about the Nominal Logic Work *}
    32 section {* A Short Review of the Nominal Logic Work *}
       
    33 
       
    34 text {*
       
    35   At its core, Nominal Isabelle is based on the nominal logic work by Pitts
       
    36   \cite{Pitts03}. The central notions in this work are sorted atoms and
       
    37   permutations of atoms. The sorted atoms represent different
       
    38   kinds of variables, such as term- and type-variables in Core-Haskell, and it
       
    39   is assumed that there is an infinite supply of atoms for each sort. 
       
    40   However, in order to simplify the description of our work, we shall
       
    41   assume in this paper that there is only a single sort of atoms.
       
    42 
       
    43   Permutations are bijective functions from atoms to atoms that are 
       
    44   the identity everywhere except on a finite number of atoms. There is a 
       
    45   two-place permutation operation written
       
    46 
       
    47   @{text [display,indent=5] "_ \<bullet> _  ::  (\<alpha> \<times> \<alpha>) list \<Rightarrow> \<beta> \<Rightarrow> \<beta>"}
       
    48 
       
    49   \noindent 
       
    50   with a generic type in which @{text "\<alpha>"} stands for the type of atoms 
       
    51   and @{text "\<beta>"} for the type of the objects on which the permutation 
       
    52   acts. In Nominal Isabelle the identity permutation is written as @{term "0::perm"},
       
    53   the composition of two permutations @{term p} and @{term q} as \mbox{@{term "p + q"}} 
       
    54   and the inverse permutation @{term p} as @{text "- p"}. The permutation
       
    55   operation is defined for products, lists, sets, functions, booleans etc 
       
    56   (see \cite{HuffmanUrban10}).
       
    57 
       
    58   The most original aspect of the nominal logic work of Pitts et al is a general
       
    59   definition for ``the set of free variables of an object @{text "x"}''.  This
       
    60   definition is general in the sense that it applies not only to lambda-terms,
       
    61   but also to lists, products, sets and even functions. The definition depends
       
    62   only on the permutation operation and on the notion of equality defined for
       
    63   the type of @{text x}, namely:
       
    64 
       
    65   @{thm [display,indent=5] supp_def[no_vars, THEN eq_reflection]}
       
    66 
       
    67   \noindent
       
    68   There is also the derived notion for when an atom @{text a} is \emph{fresh}
       
    69   for an @{text x}, defined as
       
    70   
       
    71   @{thm [display,indent=5] fresh_def[no_vars]}
       
    72 
       
    73   \noindent
       
    74   A striking consequence of these definitions is that we can prove
       
    75   without knowing anything about the structure of @{term x} that
       
    76   swapping two fresh atoms, say @{text a} and @{text b}, leave 
       
    77   @{text x} unchanged. For the proof we use the following lemma 
       
    78   about swappings applied to an @{text x}:
       
    79 
       
    80 *}
       
    81 
    26 
    82 
    27 section {* Abstractions *}
    83 section {* Abstractions *}
    28 
    84 
    29 section {* Alpha-Equivalence and Free Variables *}
    85 section {* Alpha-Equivalence and Free Variables *}
    30 
    86 
       
    87 section {* Examples *}
    31 
    88 
       
    89 section {* Conclusion *}
    32 
    90 
    33 text {*
    91 text {*
    34   Acknowledgements: We thank Andrew Pitts for the many discussions
    92 
       
    93   \noindent
       
    94   {\bf Acknowledgements:} We thank Andrew Pitts for the many discussions
    35   about the topic. We thank Peter Sewell for making [] available 
    95   about the topic. We thank Peter Sewell for making [] available 
    36   to us and explaining some of the finer points of the OTT tool.
    96   to us and explaining some of the finer points of the OTT tool.
    37 
    97 
    38 
    98 
    39 *}
    99 *}