todd (04/22/83)
og, do you have any comments about it? Would you recommend it? Are there other implementations other than the above that we should be considering? Any other advice? Todd B. Knoblock
PEREIRA@SRI-AI.ARPA (04/26/83)
A few points on C-Prolog that might help you decide: Syntax: The syntax is supposed to be the same as DEC-10 Prolog. However, there are a few exotic ambiguities in the definition of the DEC-10 syntax, so there may be obscure differences. It would be very unlikely the other systems are better in this respect, given the ambiguities in question are undocumented in the original system. There are a few deliberate differences in syntax: floating point numbers are available, so . cannot be used as an infix operator on numbers unless surrounded by blank space; the / operator is floating division, and a new operator // is integer division. DEC-10 Prolog allows | to be used as an alias for ;. C-Prolog does not. Debugger: The debugger is similar to the DEC-10 one, but less powerful and less convenient. I really miss some of the features of the DEC-10 debugger, such as the redo port and the retry command. Compiler: None of the systems you mentioned has a compiler. I do not intend to add one to C-Prolog. Various people are working on more or less portable compilers for Prolog, and I expect to see one within 1 year. Therefore, any of the current systems is likely to be a short-term investment. Support: I keep improving and debugging C-Prolog, but unfortunately I don't have the time or resources to distribute updates. This may change, however. Documentation: C-Prolog has a reference manual, which is complete if rather terse. It is an adaptation of the DEC-10 manual. Fernando Pereira (the C-Prolog implementer) -------