PG: I urge you to look at the rootdir* (docroot) that was added to srv.arc. Adding this in would allow you move the gifs you need out of the base directory for arc. This was added previously and allowed me to easily move the news gifs into arc/news_public_html and serve them from there.
PG: I removed "news.arc" from the "libs.arc". Besides probably unnecessary for most users, this was overwriting the || op which was effecting the "It's Alive" op that used to be there.
One thing that it's not good at (as far as I could tell) but is generally useful for a wiki is file uploads through web. I'm planning to add this myself but it's about 3rd or 4th on my list so I won't be upset if someone else beats me to it. hint hint.
Why would GC be unnecessary for stateless servers? Something needs to free memory once the response is committed. Either the interpreter, the server, or your webapp. I prefer the interpreter - that's why I'm not using C.
I think there is way too much discussion on this forum of new syntax without much clear gain. It's rather overwhelming listening too and I'm afraid it will make arc much more difficult to learn if it all gets adopted.
You have a point, but let's distinguish between optional syntax (like : ' !) and mandatory syntax (as is used copiously in most languages)
Optional syntax has far fewer minuses, I would argue- New users wouldn't need to use it if they prefer not to.
Also, we're not talking about increasing syntax here, just using the existing syntax in a different way.
My personal thinking, though, is that it would be OK to go all out with the other special characters to generate optional syntax, as long as it is really, really, really well thought out. (other arc users, such as yourself, would probably disagree with this argument, however) Some special character for currying (such as described in the current arc.arc comments) would be a strong plus, I think.
the problem is that people will use it. i don't want to read through a:b.c code. my eyes are already optimized for (a (b c)) and i like it thank you very much.
Nice. I was trying to think of an easy way to do this.
In order to not have to muck with the internals of load itself I wonder if you could wrap the saving path name functionality but then call the original load. Maybe this would be some new synatx / ability to add to the language? Of course this is just aspects which I thought Lisp could do so - I'm probably missing something.
