July 1998 - XEmacs-Beta - XEmacs Mailing List Archives

Re: Where do you want to go tomorrow? 26 years, 2 months

Hrvoje Niksic

sperber(a)informatik.uni-tuebingen.de (Michael Sperber [Mr. Preprocessor]) writes: > I misunderstood your request for examples; I thought you wanted to > have some proof that call-with-current-continuation is actually useful > rather than understand what it does. I don't remember the flow of all of my messages, but I think the latter is closer to the truth: I would like to know what call/cc does. > Has Justin Sheehy answered your question adequately, or should I > elabore more? Justin's examples are still too hard, and I am obviously unable to understand Scheme just by looking at it, with my knowledge of elisp and some Common Lisp. Rather than making you guys do my homework, I'll try to get hold of a good Scheme book, and learn Scheme. -- Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia --------------------------------+-------------------------------- Your lucky number today is 29842924728. Look for it everywhere.

1 participants
0 comments

Re: Regarding packages .. 26 years, 2 months

Michael Sperber

>>>>> "Holger" == Holger Schauer <schauer(a)coling.uni-freiburg.de> writes: Holger> And as a second point, two weeks ago, I posted a starter document on Holger> packages. What happened to it, did nobody receive it ? I did. I plan to incorporate it into the Info tree as soon as I can catch my breath. Thanks! -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

Re: scheme - my opinion 26 years, 2 months

Michael Sperber

>>>>> "ray" == Raymond Toy <toy(a)rtp.ericsson.se> writes: ray> As an aside, if you want to see what xemacs in lisp might look like, ray> take a peek at hemlock that comes with cmucl. It's a pseudo-emacs ray> written entirely in lisp. It's usuable, and I think some people do ray> use it as their editor. Hemlock is probably at the level of an 18.x ray> emacs, but runs on X. Actually, you might also want to check out Edwin, for which the exact same blurb applies with cmucl replaced by MIT Scheme. http://www-swiss.ai.mit.edu/ftpdir/scheme-7.4/ -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

Re: Obsolescence -- I like `eval-current-buffer' 26 years, 2 months

Hrvoje Niksic

SL Baur <steve(a)xemacs.org> writes: > 1. If a symbol in XEmacs has been renamed, but the old name exists and > is in active usage in FSF Emacs, it is (make-compatible)'d where it > was previously (make-obsolete)'ed. These are mostly done, but some > symbols have fallen through the cracks, like `eval-current-buffer'. > > I do not want package authors turning off bytecompiler warnings due > to obsolescence warnings from the bytecompiler due to portability > with FSF Emacs. I consider this an XEmacs bug to be fixed. Sometimes, not always. `window-system' is an example of otherwise. This variables is obsolete in XEmacs. Period. > I am serious about XEmacs 21 signaling "The bloat stops here." > XEmacs shall not be the prime example of a huge, bloated pig of a > piece of software. Deleting stuff no one has used, and the editor > has signalled for years as being on the verge of imminent deletion > definitely counts. Yes, but the effects of these deletions on lessening the bloat factor are likely negligible. I believe it impossible to unbloat Emacs without removing its greatest virtues. "The bloat stops here" is not the credo I like all that much, unless I know exactly what is meant by it. > We've been doing very well on the performance side. True. Note, however, that 21.0 startup will be very slow -- slower than 20.2, I think. > I've gotten kind of cranky in the last couple of weeks as my nose > has been pushed in the fact that XEmacs 21.0 is going to really > suck. I am still optimistic, except for the packages. They are problematic. -- Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia --------------------------------+-------------------------------- You have an unusual magnetic personality. Don't walk too close to metal objects which are not fastened down.

2 participants
1 comments

Re: scheme - my opinion 26 years, 2 months

Michael Sperber

>>>>> "Lars" == Lars Magne Ingebrigtsen <larsi(a)gnus.org> writes: Lars> Anyway. Few would argue that Scheme isn't more ideologically pure, Lars> but I dislike cleanliness. Common Lisp is a great, towering beast, Lars> but you don't have to be familiar with even a fraction of it to get Lars> lots of mileage out of it. And as you learn more, you find that it's Lars> basically all there. The problem with Scheme is that's there's Lars> generally nothing there, so you have to write all the code Lars> yourself, However, we're dealing with XEmacs here. Of the functionality we need for hacking the editor, almost none of it is either in Common Lisp or Scheme, as Stephen Turnbull pointed out. (Of course it isn't, but this makes the size of the semantic gap between the two smaller.) Lars> With Common Lisp one gets a language environment that is Lars> reasonably standard -- you can expect books like CLtL2, Lars> commercial environments like Allegro, free environments like Lars> CLISP, and (oh, joy of joys) an Emacs that one would know how to Lars> use if one was familiar with any of the others. Problem is, that, as far as I can see, none of the free implementations actually cover the standard. -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

Re: scheme - my opinion 26 years, 2 months

Michael Sperber

>>>>> "Stephen" == Stephen J Turnbull <turnbull(a)sk.tsukuba.ac.jp> writes: Stephen> 5. Do we or CLisp or any of the suggested Scheme variants have an API Stephen> that will allow us to separate the Lisp engine proper from extensions, Stephen> like the C-level support for fontification? This is especially Stephen> important if we hope to offload Lisp engine maintenance, as Bruno has Stephen> offered to do for CLisp. Yes. Elk has something which is close. Gambit has something which is close. RScheme has something which is close. A soon-upcoming release of Scheme 48 has something which was designed specifically with this in mind (by yours truly), which is even closer. CLisp has something which is close. However, it currently requires assembly code to work, if I understand the source right. My plan is to, as completely as possible, separate the FFI issue from the substrate issue, and thus delay the question. Since Steve has moved back the substrate change, we have plenty of time. For all I care, one interface will simply offer CORBA connectivity. -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

Re: scheme - my opinion 26 years, 2 months

Michael Sperber

>>>>> "Martin" == Martin Buchholz <martin(a)xemacs.org> writes: Martin> (call-with-current-continuation) is a very interesting Martin> function, but one cannot expect ordinary humans to understand Martin> it. It is only a tool for system implementors. Ermh, we *are* system implementors. Martin> The scheme community seems to be in a downturn. From casual Martin> visits to the scheme repository, there seems to be little Martin> progress in implementations or standardization. I don't follow that. The latest revision of the Scheme standard just came out in March. (*This* year!) The next round starts off with a workshop in September. I could argue the same about Common Lisp. At least three big Common-Lisp-related companies have folded recently: Lucid (remember?), TMI, and Symbolics. I don't think this is relevant to our discussion. Both languages *obviously* have sizeable followings. -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

Re: My thoughts on package-delete (was Re: Fighting with packages!) 26 years, 2 months

Raymond Toy

>>>>> "sb" == SL Baur <steve(a)xemacs.org> writes: sb> Footnotes: sb> [1] Do we really? It will certainly help testers who already have a sb> populated Lisp package hierarchy and bozos like me who will forget and sb> install things by manual untar in the appropriate directory. We do. If not bozos like you, for bozos like me who can't use package-update because I can't get efs working. I've tried to get it working for quite a while now but I cannot get it to understand our kerberos system here. So, I'm left with ftp'ing what I want, and since I gone that far by hand, might as well untar by hand too. Ray

1 participants
0 comments

Re: Where do you want to go tomorrow? 26 years, 2 months

Michael Sperber

>>>>> "Hrvoje" == Hrvoje Niksic <hniksic(a)srce.hr> writes: Hrvoje> Uh, I am ashamed to admit that I don't understand these Hrvoje> examples. No shame in that. At first, understanding call/cc requires twisting your neck by 180 degrees, which I found hard to do initially. I misunderstood your request for examples; I thought you wanted to have some proof that call-with-current-continuation is actually useful rather than understand what it does. Has Justin Sheehy answered your question adequately, or should I elabore more? -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

Re: Where do you want to go tomorrow? 26 years, 2 months

Michael Sperber

>>>>> "Karl" == Karl M Hegbloom <karlheg(a)inetarena.com> writes: >>>>> "Michael" == Michael Sperber <sperber(a)informatik.uni-tuebingen.de> writes: Michael> (Note that the reverse is not generally possible, since Michael> Common Lisp is not properly tail-recursive and does not Michael> support control abstraction.) Karl> I guess these are the main things that Common Lisp lacks over Scheme. Karl> You're talking about `call-with-current-continuation'. Karl> Clisp is a very rich language, with a large vocabulary, CLOS and Karl> threads. Last I checked, threads weren't in yet. Karl> What Scheme has that? I assume you're referring to "large vocabulary." Schemes that have object systems and threads (and pretty extensive libraries) are: - Scheme 48 - MzScheme - MIT Scheme - RScheme - ? Guile ? I've very probably missed some. -- Cheers =8-} Chipsy Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

XEmacs-Beta July 1998