On Thu, Jun 27, 2013 at 3:04 AM, Gary Trakhman <gary.trakh...@gmail.com>wrote:
> The first thing I think of along these lines is codeq > http://blog.datomic.com/2012/10/codeq.html > Codeq seems very good for querying and manipulating the history of the project, but i was thinking along the lines of executing the whole history of the project. Being able to execute the whole history is not always meaningful, there is always a kind of global state, but for deterministic functions that operate only on their arguments, that shouldn't be impossible to do. > When I first heard about it, I thought it might have potential for code > reuse and whole-program optimization. If you've ever imported something > that causes irrelevant classes to take up space in your application's > distribution, you might also wonder if it's possible to have reusable > clojure code that isn't tied to java classes and classloaders. > > There is that thing called Church Numerals that does something like that. http://en.wikipedia.org/wiki/Church_encoding#Computation_with_Church_numerals > On Wed, Jun 26, 2013 at 7:01 PM, emporas <empo...@gmail.com> wrote: > >> Hello, I have a question that is more of a Lisp question, than Clojure >> specific, but it is in the true spirit of things as 'values'. >> >> Now that the practice of applying collision resistant hashes (ie unique >> strings), to save multiple >> versions of structures with the possibility of overlapping, is becoming >> less deficient thanks to better >> hardware, i am wondering if there is a compiler in lisp or anything >> else, that has the ability to save >> multiple versions of functions, that operate only on their arguments, >> better known as deterministic or >> pure functions. >> >> A compiler like that, it could keep all of the deterministic functions >> ever written, the number of the them could only increase and never >> decrease. That would not solve the versioning problem, but it could >> sure alleviate it. In theory, at least, integers do not vary a lot from >> one platform over another, only their >> performance characteristics. That could enable the compiler to keep all >> the versions of those functions >> the operate on integers around, and the programmer to move backwards and >> run whichever function he prefers, from all their different versions >> over time. >> >> With my poor knowledge in Lisp, i think that Lisps-2 CL/Lfe are more >> appropriate for this task than >> Lisps-1 Clojure/Scheme. A language like Haskell that has explicit >> annotations for the functions that >> are not deterministic seems very close to that idea, too. In dynamic >> languages that there is no out of >> the box solution for this, one could simply attach an image of a >> chimpanzee scratching it's head to the >> metadata of the impure functions, and solve the problem instantly! >> >> Does anyone know if there is any work in that area, such as compilers, >> academic papers or am i horribly disguided? >> >> -- >> -- >> You received this message because you are subscribed to the Google >> Groups "Clojure" group. >> To post to this group, send email to clojure@googlegroups.com >> Note that posts from new members are moderated - please be patient with >> your first post. >> To unsubscribe from this group, send email to >> clojure+unsubscr...@googlegroups.com >> For more options, visit this group at >> http://groups.google.com/group/clojure?hl=en >> --- >> You received this message because you are subscribed to the Google Groups >> "Clojure" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to clojure+unsubscr...@googlegroups.com. >> For more options, visit https://groups.google.com/groups/opt_out. >> >> >> > > -- > -- > You received this message because you are subscribed to the Google > Groups "Clojure" group. > To post to this group, send email to clojure@googlegroups.com > Note that posts from new members are moderated - please be patient with > your first post. > To unsubscribe from this group, send email to > clojure+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/clojure?hl=en > --- > You received this message because you are subscribed to the Google Groups > "Clojure" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to clojure+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
