I did not expect this to be ever released. Congratulations on being better than the GINGER (who has yet to release CROMA Lisp after all these years) in this aspect!
This completely changed my opinion on you.
Thank you, >>1, and good job.
Name:
Anonymous2011-05-03 20:40
Could you please not use some ghetto amateur file sharing site like MediaFire that is loaded with annoying advertisements, and instead use an actual open source software repository like github or Source Forge?
Name:
Anonymous2011-05-04 2:33
>>6
They require gmail account, which I dont have money or mobile phone to purchase.
http://mail.google.com/support/bin/answer.py?hl=en&answer=114129
The cost of your text or voice messages will vary depending on your plan and provider, but will likely be your standard text message and call charges. Contact your mobile phone provider for details.
There are only 8 files in ``in Lisp'' (they look like they were assembled from the real ``in Lisp'' troll's posts in /prog/), all others (didn't count, but probably more than 30) are in Common Lisp. What the fuck?
Name:
Anonymous2011-05-04 10:35
>>9
Nice data mining scheme(unlike 4chan). I'm deleting my gmail account right now.
Name:
Anonymous2011-05-04 10:37
Thankfully Yahoo is perfectly happy without knowing my real phone number or address, i'll use it for awhile
Name:
Anonymous2011-05-04 10:40
>>11
For prototyping purposes most of implementation uses CL (including pattern mather). A few basic macros are in lib/init. Version 1.0 most likely will be rewritten in Scheme. It's just a preview release.
Name:
Anonymous2011-05-04 10:42
>>13
Some sites (notably torrents) require exclusively google mail account, as it's harder to fake (you have to pay real money for reg).
>>14 Version 1.0 most likely will be rewritten in Scheme.
I thought you hated Scheme.
Name:
Anonymous2011-05-04 10:59
>>16
I hate hygiene obfuscation. But Scheme/Racket uses JIT and supports continuations, which I believe should be included in all future Lisp implementations. It would be easier to just reuse Scheme's framework.
Name:
Anonymous2011-05-04 11:02
I also want to have transparent metadataassociated with each s-exp (like in Clojure), instead of Scheme's syntax-object hack.
>>17-18
Use explicit-renaming or syntactic-closures (Racket doesn't support them, I'm just saying that syntax-case is not the only near-low-level macro system in Scheme world).
And hygiene is necessary:
* (defun f () 3)
F
* (defmacro m () '(f))
M
* (m)
3
* (flet ((f () 2)) (m))
2
>>19
The behavior in your first example is perfectly understood and expected.
Hygiene as you describe it (where the symbol name isn't shadowed) is possible in CL if you decide to make your own macro system like in Scheme, but nobody actually wants/needs it. I read a few papers describing in full the implementation of hygienic macro systems in Scheme and I don't see any issue with bringing it to CL if I needed to, but I really never had any problems as in your case as I use packages.
>>24
I don't think it's a flaw, it's just a different way of doing things. If it's a flaw, I'm right to call the Lisp-1-ness of Scheme a flaw(and having to use lyst, lst and so on). It's a trade-off.
It's not like you couldn't implement a hygienic macro system only given CL, it's just that most CLers don't have a need to.
>>25 I don't think it's a flaw, it's just a different way of doing things. If it's a flaw, I'm right to call the Lisp-1-ness of Scheme a flaw(and having to use lyst, lst and so on). It's a trade-off.
It's understandable, being a Lisp-1 has trade-offs too, but you can't be reasonable with trolls/closed-minded people like >>22,23.
It's not like you couldn't implement a hygienic macro system only given CL, it's just that most CLers don't have a need to.
That would require replacing defun/flet/lambda/...[1], but yes, it's possible.
My point was that Lisp-n plus gensym is not completely immune to unhygienic bindings.
I have two gmail accounts and I've never been prompted for any money or a phone number, are you certain that you need to be able to receive text messages?
Name:
Anonymous2011-05-05 8:37
>>27
It's more high-level. Relies on pattern matching, persistent data structures and most of its syntax is geared towards list-processing.
US PATENT SUBCLASS 525 / 509
.~.~ Solid polymer or SICP derived from at least one amine-, N-C(=X)- or N-S(=O) containing reactant and at least one aldehyde or aldehyde-type reactant (X is chalcogen) http://www.patentec.com/data/class/defs/525/509.html