I'm currently using mark and sweep in my language and thinking of switching to incremental tricolor for better real-time performance. Does anyone have any...
I'm sure you know more about efficient collectors than I do. But I think a cool thing about scavenging is it provides a natural way to get transparent...
Hi Patrick, Thanks alot for this post. I've been digesting it! I have some comments inline. ... Thanks, I managed to find some powerpoint slides. It seems...
Hi Jack, ... First of all, there is a short paper outlining how to do this in a generic way. See http://tunes.org/~eihrul/pmd.pdf. However, this paper does not...
Very cool. How do you handle argument positions that don't have "@", i. e., wild- card positions? Is it (or do you see it as) a typical programming practice to...
Hello again, ... Dispatching is an annotation that you put on objects to refer to that method at a particular (role, selector) pair. So if an argument-position...
... Rice <water@t...> wrote: . . . ... dispatch is ... available ... basic type ... already ... I thought you were de-emphasizing the concept of type? ... Yes,...
On Fri, 13 Sep 2002, Jack Waugh wrote: [Omitted information about type-annotations and inference.] ... Sure, but it *does* matter at some level (it saves on...
... Bawden's writing style is so clear and readable! Did he write this near the time the concept of linearity entered the literature on computing? On June 11,...
I am in the middle of reading Bawden's paper and have come to where he starts to talk about state, saying that the viewpoint of linear graph reduction sheds...
at present I am designing a language which I am unsure whether it is even really scheme anymore, or whether I should try to keep it as scheme... largely it is...
Hi Jack, ... It means that there's an intersection of lists here, with each placing each method in its own 'precedence' with respect to the others. A stable ...
In http://groups.yahoo.com/group/langsmiths/message/117 , Brian T ... idea of ... ideas. ... a ... I don't understand; Bawden's paper does not describe a...
Okay, I realize that I had previously misunderstood your explanations... ... Yes, that's correct, but with certain modifications that I believed would help,...
I am not sure how similar this is to what "Jack Waugh" was talking about (given I did not fully understand). I had been sitting around with an idle thought...
... OK, here's the manual: http://groups.yahoo.com/group/public_links/files/People/jack_waugh/Com puters/symmetry.html Yahoo! will have broken up the above...
Anyone know of a memory allocator optimized for locality? Ideally, I'd like one that I could give hints to like: void *malloc(size_t numBytes, void ...
... I don't think it would be too hard to pull off, though I don't really see a point and it would likely hurt efficiency (and could cause heigher memory...
... Right, but nearest fit to a given point, instead of the last point. ... I understand. That's what I'm going for but this variation on nearest fit seemed...
ColdStore had something a little like this for C++ -- it used a BTree for the implementation and you could pass a locality hint to the new operator. Take a...
ryan daum
opiate@...
Sep 28, 2002 1:47 pm
134
... agreed. ... ok. I am not sure but I think copy collectors usually get good locality in many cases, but could be optimized for better locality. funny: not...
... Yes, I really like that about them. I worry about the added complexity and memory usage(for dual memory spaces) though. I'm considering a sort of partial...
... it also depends on language, it would likely be a problem for a c like language. I wasn't really sure of the indended usage of the allocator. ... I don't...
ok, I had started coding a gui for my vm (yesterday). it is kind of cool so far. I am using gl, mostly for ease and also because it matches the way in which I...
... I did not catch this one, shows what happens if I post code fragments I had not really tested (this was moved to script so now 'render-obj' is not...
I'm sorry for not replying sooner. There was a bit of initial digestion and amazement at such a divergent extension to the Smalltalk model of programming. The...