tag:blogger.com,1999:blog-5592542.post5949033357506356812..comments2024-03-18T17:28:44.693-07:00Comments on Rondam Ramblings: My take on ArcRonhttp://www.blogger.com/profile/11752242624438232184noreply@blogger.comBlogger13125tag:blogger.com,1999:blog-5592542.post-52367674989476326832008-02-11T13:29:00.000-08:002008-02-11T13:29:00.000-08:00First, Paul has specifically said that unhygienic ...First, Paul has specifically said that unhygienic macros were a deliberate decision because he thinks name capture is a purely academic concern that doesn't actually arise in practice. And the whole unicode thing has been fixed (and even before that I agreed with Paul that it was all much ado over nothing).<BR/><BR/>Second, I'd say Arc is already a "proper Lisp" (to the extent that I can glean Ronhttps://www.blogger.com/profile/14719368822663798864noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-67188002201142606762008-02-11T13:08:00.000-08:002008-02-11T13:08:00.000-08:00What if pg and rtm are performing a giant "head fa...What if pg and rtm are performing a giant "head fake" on the lisp community by releasing a version of Arc based on Scheme that they promise to break in fundamental ways?<BR/><BR/>If true, then the lack of hygenic macros, Unicode, etc are just ways of coaxing people into the upgrade. I wouldn't put it past pg and rtm to build some "can't live without it" web thing in Arc, then charge for the Jimhttps://www.blogger.com/profile/06650027377863169558noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-34283812384642759482008-02-06T07:46:00.000-08:002008-02-06T07:46:00.000-08:00I strive to become a Lisp aficionado and so am dis...I strive to become a Lisp aficionado and so am disheartened that someone could lose their faith in Lisp. Now that you have done considerable work in Python, do you still feel that way?Unknownhttps://www.blogger.com/profile/01627999279978758135noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-40377391287094614232008-02-05T23:54:00.000-08:002008-02-05T23:54:00.000-08:00> I see I was in err, I was under the impressio...> I see I was in err, I was under the impression that B, Forth, Fortran, or Algol could do that; I'm afraid I'm not too familiar with any of those languages though, leading to my misconception.<BR/><BR/>In fact, Burroughs used (a dialect of) Algol to implement the OS of the B5000 in 1961, whereas C was invented just about a decade later. C really <EM>wasn't</EM> all that innovative. It won forChris Bartshttps://www.blogger.com/profile/13381221381435216868noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-55948308708273008062008-02-05T12:26:00.000-08:002008-02-05T12:26:00.000-08:00I submit that what you want from a programming lan...<EM>I submit that what you want from a programming language is not one that makes programs shorter, but one that makes programs easier to create.</EM><BR/><BR/>i think the word <EM>expressiveness</EM> is often used. some people in the programming language community [who had fed up with all the informal claims in this area] had tried to identify and quantify this in the distant past. my favorite ozhttps://www.blogger.com/profile/06062585218369513464noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-74043585235216998572008-02-02T16:19:00.000-08:002008-02-02T16:19:00.000-08:00> But Common Lisp is also not the 100-year languag...> But Common Lisp is also not the 100-year language, at least not according to Paul Graham. (In fact, Paul has said that Common Lisp "sucks".) Paul's idea of the 100-year language involves more than mere longevity. If that's all there was to it then Fortran would be a serious contender too (and it *was* innovative in its day).<BR/><BR/>Ah, he and I define it differently. I consider a 100-year roothttps://www.blogger.com/profile/17676417938259782648noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-72507257644918695422008-02-02T12:14:00.000-08:002008-02-02T12:14:00.000-08:00> Common Lisp has lasted a while and there wasn't ...> Common Lisp has lasted a while and there wasn't anything very innovative about it.<BR/><BR/>But Common Lisp is also not the 100-year language, at least not according to Paul Graham. (In fact, Paul has said that Common Lisp "sucks".) Paul's idea of the 100-year language involves more than mere longevity. If that's all there was to it then Fortran would be a serious contender too (and it *was*Ronhttps://www.blogger.com/profile/14719368822663798864noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-45471519473576499612008-02-01T22:57:00.000-08:002008-02-01T22:57:00.000-08:00>> Why should it have to be innovative?>>>> One of...>> Why should it have to be innovative?<BR/><BR/>>>>> One of Arc's stated goals is to be "the 100 year language." I think meeting that goal is going to take some innovation.<BR/><BR/>Common Lisp has lasted a while and there wasn't anything very innovative about it. Zeta-Lisp and others had already implemented most of what was in the CL spec; though maybe not all in one language.<BR/><BR/>I do notroothttps://www.blogger.com/profile/17676417938259782648noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-28364453294539537182008-02-01T21:06:00.000-08:002008-02-01T21:06:00.000-08:00> Common Lip doesn't have an OO layer either, that...> Common Lip doesn't have an OO layer either, that didn't stop someone from making CLOS though.<BR/><BR/>Yes, and the fact that CLOS is grafted on as an afterthought is the source of many of its shortcomings. Math functions are not generic. You can't inherit from built-in types like list. (You can in Python, and this turns out to be tremendously useful.)<BR/><BR/>> Why should it have to be Ronhttps://www.blogger.com/profile/14719368822663798864noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-67483537433823505962008-02-01T12:00:00.000-08:002008-02-01T12:00:00.000-08:00>The Right Thing ... is to provide an abstract ass...>The Right Thing ... is to provide an abstract associative map whose underlying implementation can be changed.<BR/><BR/>You might be interested in <A HREF="http://clojure.sourceforge.net" REL="nofollow">Clojure</A>,<BR/>a Lisp for the JVM where sequences (first/rest), maps, vectors, invoke-ability etc are all abstractions that support multiple implementations.RH Trialhttps://www.blogger.com/profile/17189173699994999680noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-52787287036127072112008-02-01T11:16:00.000-08:002008-02-01T11:16:00.000-08:00> In particular, it seems a safe bet that Arc will...> In particular, it seems a safe bet that Arc will not have an OO layer, which means...<BR/><BR/>Common Lip doesn't have an OO layer either, that didn't stop someone from making CLOS though.<BR/><BR/>In response to one of the comments:<BR/><BR/>Arc isn't meant to be innovated, it's just meant to be fun to program in. Why should it have to be innovative? All it needs to do is address problems roothttps://www.blogger.com/profile/17676417938259782648noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-32120258633317602302008-02-01T04:19:00.000-08:002008-02-01T04:19:00.000-08:00> If Arc doesn't save Lisp it's hard to imagine wh...> If Arc doesn't save Lisp it's hard to imagine what would.<BR/><BR/><A HREF="http://www.lambdassociates.org/" REL="nofollow"><BR/>Qi</A>. It seems much more innovative than Arc.danbhttps://www.blogger.com/profile/00888160127591318240noreply@blogger.comtag:blogger.com,1999:blog-5592542.post-19954645039487673442008-01-31T19:30:00.000-08:002008-01-31T19:30:00.000-08:00Thank you for your informative article.I do wonder...Thank you for your informative article.<BR/><BR/>I do wonder if what PG meant by "sortable" assoc lists is "ordered". You can sort a list, but if you don't it keeps its own order, which can be meaningful. You can sort hash keys, but if you don't they're not kept in meaningful order.<BR/><BR/>I'm not sure that's what he meant, since he didn't actually say it.<BR/><BR/>But all your other comment Chouserhttps://www.blogger.com/profile/01980936182490649442noreply@blogger.com