David Bakhash <ca...@bu.edu> writes: > Does ilisp have a mechanism to lookup the documentation of the function at > (point) in the CLHS and then present it to you using w3?
> is it an add-on? I'd love to have it (otherwise, I'll probably write it).
David Bakhash <ca...@bu.edu> writes: > this is really excellent.
Indeed it is. But beware: This is addictive! Once you've gotton used to the idea of getting excellent and accurate documentation at the touch of a button, you'll never want to be without it...
This brings me to something I've been pondering about: As part of (re-)writing documentation for MaiSQL (and several other packages), I've started on some work of customizing DocBook to make it more suitable for (Common) Lisp (i.e. stripping out useless C/Unix centric stuff (synopsis-related) and introducing instead things of use to Lisp developers), as well as some work on Norman Walsh' excellent DocBook DSSSL StyleSheets, which produce nice HTML and printed output from the DocBook source.
It might be useful to try to "standardize" some form of DocBook customization, and write a couple of simple tools/framework stuff, that would allow documentation written using this to be easily plugged into Emacs for lookup (i.e. since all symbols will be marked up in the documentation stuff anyway, it should be simple to produce the symbol->URL mapping needed for something like Erik's hyperspec.el while generating the HTML output).
If combined with some standardized defsystem stuff, this would make for easily "pluggable" packages...
Anyone interested in this kind of stuff? (Beware: nothing of the stuff imagined here is anywhere near alpha-stage, even design-wise, and I probably won't have time to do something about it in the next 1-3 months. Just interested in opinions at the moment...)
Regs, Pierre.
-- Pierre Mai <p...@acm.org> PGP and GPG keys at your nearest Keyserver "One smaller motivation which, in part, stems from altruism is Microsoft- bashing." [Microsoft memo, see http://www.opensource.org/halloween1.html]
> I've started on some work of customizing DocBook to make it more > suitable for (Common) Lisp (i.e. stripping out useless C/Unix centric > stuff (synopsis-related) and introducing instead things of use to Lisp > developers), as well as some work on Norman Walsh' excellent DocBook > DSSSL StyleSheets, which produce nice HTML and printed output from the > DocBook source. [...] > Anyone interested in this kind of stuff?
[...] I'd certainly be interested in seeing this. -- T. Kurt Bond, t...@access.mountain.net
Marco Antoniotti <marc...@copernico.parades.rm.cnr.it> writes: > I have never used hyperspec.el and now I hear that there are two > versions of it floating around.
Two versions or two completely different implementations? I wrote my own hyperspec.el once (http://www.telent.net/lisp/hyperspec.el) purely because I couldn't find Erik's at the time. The basic difference in normal use, I think, is that when looking up a symbol with multiple references (eg LAMBDA), Erik's implementation visits each page in turn so that the user can use netscape's Forward/Back buttons to switch between them, whereas mine goes to the HyperSpec/Body/any_lambda.html page which says 'Please select which reference to LAMBDA you intended'
His will also work with a remotely-hosted hyperspec, whereas mine requires that you download it first.
> I'd like feedback on which people deem the best before I go ahead and > includ it in 5.9.
hyperspec-a-la-dan has three users that I know of. I won't weep if ilisp obsoletes it.
Daniel Barlow writes: >Two versions or two completely different implementations?
It looks like there are THREE completely different interpretations. (Talk about great minds working along the same lines....) The one I use is by Stephen Carney (car...@gvc.dec.com as of July 8 1997).
-- Fred Gilham gil...@csl.sri.com "Come to me, all who labor and are heavy laden, and I will give you rest. Take my yoke upon you, and learn from me, for I am gentle and lowly in heart, and you will find rest for your souls. For my yoke is easy, and my burden is light." --Jesus of Nazareth
gil...@snapdragon.csl.sri.com (Fred Gilham) writes, in <u7emh1koqv....@snapdragon.csl.sri.com>:
>Daniel Barlow writes: >>Two versions or two completely different implementations? >It looks like there are THREE completely different interpretations. >(Talk about great minds working along the same lines....) The one I >use is by Stephen Carney (car...@gvc.dec.com as of July 8 1997).
My HyperSpec hack is at <http://www.scubadoo.com/carney/emacs/index.html> under "Miscellaneous", and my email is now <Steve.Car...@altavista.com>. This implementation is unrelated to the others AFAIK.
Back around Jul-1997, I posted the sources to gnu.emacs.sources, and RMS responded. Although the HyperSpec was interesting/neat, he didn't welcome my posting of a HyperSpec interface on that newsgroup, because the HyperSpec itself couldn't be incorporated into FSF sources/control/licensing/whatever. So, further posts for yet another HyperSpec interface might be best kept off of gnu.emacs.sources (unless something has changed).
Steve Carney <Steve.Car...@altavista.com.NOSPAM> writes: > Although the HyperSpec was interesting/neat, he didn't > welcome my posting of a HyperSpec interface on that newsgroup, because > the HyperSpec itself couldn't be incorporated into FSF > sources/control/licensing/whatever. So, further posts for yet another > HyperSpec interface might be best kept off of gnu.emacs.sources (unless > something has changed).
Yes, he had a hissy fit over the fact that somewhere in the world there was a content creator that did not, upon creation of content, hand over to Stallman all rights to do whatever they wanted to the Hyperspec. Well, the HyperSpec wouldn't have been created at all if that had been a necessary condition of its creation, so he got a local optimum. I had a long conversation about this by phone in which he told me he was upset that the spec wasn't "free". I explained to him that it was "free", but not "free" in the meaning he wanted to assign to that word--more in the sense that the word has traditionally been assigned: without cost. Yes, CLHS still has restrictions. But that doesn't mean the dictionary word "free" does not apply, nor do I accept Stallman as a superior authority on the meaning of the word "free".
My impression of Stallman's position is that he wants to reassign the meaning of that word to be something else, but *I* think he should get another word besides free for what he wants. I further feel like (and it's just my personal impression) he wants everyone to have full rights to content EXCEPT the content creator. Because when it comes to a test of wills over how to use what is created, he always wants to use content in ways the author didn't intend. And I don't like that not because I don't like to see him do interesting things, but because I fear that the kind of world he lives in will provide no incentive for the creation of content. It means one can never establish a mark of quality on something. It's the same world that would have color versions of Casa Blanca at the theaters, and edited for TV versions of Eyes Wide Shut. Right now, anyone who wants to can create content that is free for anyone to modify, but I don't think it's reasonable to assert that all people should have to do that.
The specific restrictions on modification of the HyperSpec were created to assure quality control (since the Harlequin name is affixed) and to assure "advertising effect" (since the HyperSpec cost $$ to produce, and Harlequin had to recover the cost somehow; the other way to do so would have been to package it into a specific product, but I argued [eventually successfully] that the advertising value, even if it could not be accounted for specifically because the document was moving freely around, would more than pay for itself). Making it "Stallman-Free" (sorry, not "free of Stallman" but "free according to Stallman") would have made it harder for me to convince anyone of this claim, and probably would have meant there was no cost-free way to get access.
Stallman grumbled that he would just go off and write his own processor to produce the equivalent result. To my knowledge, he didn't ever do that. Franz essentially did the effort over again themselves, and it's part of their product--not "free", which is completely understandable: They had to pay money to get it done, and they can't afford to just give away that investment to potential competitors any more than Harlequin could. It's tough being a commercial entity and having to care about these issues, but these issues can make the difference between a company making money and not.
Franz, Harlequin, and others donated lots of dollars toward making sure the ANSI standard happened at all. They also have cost-free variants of their commercial software available--again under license restrictions. At any given time, there is a limit, and reasonably so, to how much they can do for the community without charging. It might be great for Stallman if they made all their software open, too, but it would not necessarily be great for Lisp because two important commercial sources of Lisp would be out of business.
As to whether he can include the HyperSpec in FSF sources, I'm not at Harlequin so cannot speak for them (if I ever could), but my *impression* casually looking at the license (which I admit to having partly myself originally drafted) is that he certainly can as long as he doesn't modify the sources or charge money or various other specific things specifically enumerated in the license. His problem is not with his inability to incorporate it, it's with his inability to CHANGE it during said incorporation. It's certainly possible for him to want to deny his users access to the HyperSpec based on this lack of flexibility, but I think he does his users a disservice in exercising that possibility, since this facility would be of use to some of them.
We each create our contributions to the world in different ways. I can understand and respect Stallman's vision for what he wants to do with his time and energy, but it is only one possible vision and I can and do have respect for other people's views of how they are trying to make the world better, too. What really bugs me about Stallman's view is that in the process of what he does good, he and his movement seem to breed a set of people who are taught to have disdain for others who are doing nothing more than to try to improve the world in a different way than he is. I think that kind of closed-mindedness that hurts his cause a great deal. It's more important that people think hard about the social consequences of their actions and strive every day to do well by the world than that we all agree on a specific set of social patterns that are Always Good and another set that are Always Bad. Almost by definition, I think, dogma is going to lose. And in my opinion, Stallman's approach, by refusing to admit the possibility that certain specific other approaches have some legitimacy, tends toward dogma. He's got some good ideas bound up in there, but they're just some of the many options people should consider in trying to decide how best they can contribute to the world.
> My impression of Stallman's position is that he wants to reassign the > meaning of that word to be something else, but *I* think he should get > another word besides free for what he wants.
It seems that you are intent on denying the meaning that "free" has had for a very long time, insisting that it means "no cost". As if the hegemony of political-economy should include not only our understanding of the creative process thru the blind assumption of the model of economic incentive as fact, but also our language.
From Webster's Revised Unabridged Dictionary (1913) [web1913]:
Free \Free\ (fr[=e]), a. [Compar. {Freer} (-[~e]r); superl. {Freest} (-[e^]st).] [OE. fre, freo, AS. fre['o], fr[=i]; akin to D. vrij, OS. & OHG. fr[=i], G. frei, Icel. fr[=i], Sw. & Dan. fri, Goth. freis, and also to Skr. prija beloved, dear, fr. pr[=i] to love, Goth. frij[=o]n. Cf. {Affray}, {Belfry}, {Friday}, {Friend}, {Frith} inclosure.] 1. Exempt from subjection to the will of others; not under restraint, control, or compulsion; able to follow one's own impulses, desires, or inclinations; determining one's own course of action; not dependent; at liberty.
That which has the power, or not the power, to operate, is that alone which is or is not free. --Locke.
Yes, "free" also means, no cost, no money need be paid. But I don't think that Richard is re-assigning the word to mean something new, it has always meant liberty, independent action. I think it disingenuous to claim that his use of the word "free" in this well-established manner is an attempt to "re-assign" the meaning of a common word. We all know what a Free Man is afterall.
There is a line of argument that says that the restrictions which the GPL place on distribution of derivative works disqualified it from being considered Free, but my impression is that that is hardly the direction which you were coming from. There are counter-arguments, and counter-counter arguments to this that I don't really want to cover, and anyone can look these up in DejaNews on gnu.misc.discuss
I've attached the definition of the word "free" as obtained from various Free sources, to make clear that "no cost" is hardly the only, or even dominant understanding of the word. I still like the typographical convention of capitalizing it when refering to the higher notion of liberty, and so I try to be consistent in that.
> Franz, Harlequin, and others donated lots of dollars toward making > sure the ANSI standard happened at all. They also have cost-free > variants of their commercial software available--again under license > restrictions. At any given time, there is a limit, and reasonably so, > to how much they can do for the community without charging. It might > be great for Stallman if they made all their software open, too, but > it would not necessarily be great for Lisp because two important > commercial sources of Lisp would be out of business.
True, one can't really expect those who base their business on proprietary software to be able to make a transition to other models. It's especially true when they are already on shaky ground.
But as we draw closer to a single commercial vendor for the Language, with the purchase of Harlequin by a company not really in the software development business, I really wonder what good the vendors have done. What is their legacy? How much work, time, intellect, blood, and sweat are now tied up in dead companies? Who is going to use a gigantic language with a single viable vendor who charges royalties for deployment?
Am I being to bleak?
I really see no future in the commercial vendors. If Lisp is going to have any real longevity it is going to be in the Free variants. Franz might hold out for a decade, maybe more, but I don't see more vendors joining the ranks of lively Lisp implementors anytime soon, and a single vendor language is a dead language. But I am not going to turn around and say that Free Software will "save" lisp. The Free implementations have their own set of problems that I don't see any easy solutions for either.
Most importantly, thanx for the HyperSpec Kent.
From Webster's Revised Unabridged Dictionary (1913) [web1913]:
Free \Free\ (fr[=e]), a. [Compar. {Freer} (-[~e]r); superl. {Freest} (-[e^]st).] [OE. fre, freo, AS. fre['o], fr[=i]; akin to D. vrij, OS. & OHG. fr[=i], G. frei, Icel. fr[=i], Sw. & Dan. fri, Goth. freis, and also to Skr. prija beloved, dear, fr. pr[=i] to love, Goth. frij[=o]n. Cf. {Affray}, {Belfry}, {Friday}, {Friend}, {Frith} inclosure.] 1. Exempt from subjection to the will of others; not under restraint, control, or compulsion; able to follow one's own impulses, desires, or inclinations; determining one's own course of action; not dependent; at liberty.
That which has the power, or not the power, to operate, is that alone which is or is not free. --Locke.
2. Not under an arbitrary or despotic government; subject only to fixed laws regularly and fairly administered, and defended by them from encroachments upon natural or acquired rights; enjoying political liberty.
3. Liberated, by arriving at a certain age, from the control of parents, guardian, or master.
4. Not confined or imprisoned; released from arrest; liberated; at liberty to go.
Set an unhappy prisoner free. --Prior.
5. Not subjected to the laws of physical necessity; capable of voluntary activity; endowed with moral liberty; -- said of the will.
Not free, what proof could they have given sincere Of true allegiance, constant faith, or love. --Milton.
6. Clear of offense or crime; guiltless; innocent.
My hands are guilty, but my heart is free. --Dryden.
7. Unconstrained by timidity or distrust; unreserved; ingenuous; frank; familiar; communicative.
He was free only with a few. --Milward.
8. Unrestrained; immoderate; lavish; licentious; -- used in a bad sense.
The critics have been very free in their censures. --Felton.
A man may live a free life as to wine or women. --Shelley.
9. Not close or parsimonious; liberal; open-handed; lavish; as, free with his money.
10. Exempt; clear; released; liberated; not encumbered or troubled with; as, free from pain; free from a burden; -- followed by from, or, rarely, by of.
Princes declaring themselves free from the obligations of their treaties. --Bp. Burnet.
11. Characteristic of one acting without restraint; charming; easy.
12. Ready; eager; acting without spurring or whipping; spirited; as, a free horse.
13. Invested with a particular freedom or franchise; enjoying certain immunities or privileges; admitted to special rights; -- followed by of.
He therefore makes all birds, of every sect, Free of his farm. --Dryden.
14. Thrown open, or made accessible, to all; to be enjoyed without limitations; unrestricted; not obstructed, engrossed, or appropriated; open; -- said of a thing to be possessed or enjoyed; as, a free school.
Why, sir, I pray, are not the streets as free For me as for you? --Shak.
15. Not gained by importunity or purchase; gratuitous; spontaneous; as, free admission; a free gift.
16. Not arbitrary or despotic; assuring liberty; defending individual rights against encroachment by any person or class; instituted by a free people; -- said of a government, institutions, etc.
17. (O. Eng. Law) Certain or honorable; the opposite of base; as, free service; free socage. --Burrill.
18. (Law) Privileged or individual; the opposite of common; as, a free fishery; a free warren. --Burrill.
19. Not united or combined with anything else; separated; dissevered; unattached; at liberty to escape; as, free carbonic acid gas; free cells.
{Free agency}, the capacity or power of choosing or acting freely, or without necessity or constraint upon the will.
{Free bench} (Eng. Law), a widow's right in the copyhold lands of her husband, corresponding to dower in freeholds.
{Free board} (Naut.), a vessel's side between water line and gunwale.
{Free bond} (Chem.), an unsaturated or unemployed unit, or bond, of affinity or valence, of an atom or radical.
{Free-borough men} (O.Eng. Law). See {Friborg}.
{Free chapel} (Eccles.), a chapel not subject to the jurisdiction of the ordinary, having been founded by the king or by a subject specially authorized. [Eng.] --Bouvier.
{Free charge} (Elec.), a charge of electricity in the free or statical condition; free electricity.
{Free church}. (a) A church whose sittings are for all and without charge. (b) An ecclesiastical body that left the Church of Scotland, in 1843, to be free from control by the government in spiritual matters.
{Free city}, or {Free town}, a city or town independent in its government and franchises, as formerly
wtank...@dolphin.openprojects.net (William Tanksley) writes: > On 19 Aug 1999 19:55:01 -0500, Craig Brozefsky wrote:
> >It seems that you are intent on denying the meaning that "free" has > >had for a very long time, insisting that it means "no cost". As if > >the hegemony of political-economy should include not only our > >understanding of the creative process thru the blind assumption of the > >model of economic incentive as fact, but also our language.
> No, he's denying that "free" makes much sense when applied in that sense > to inanimate objects.
Well that's why I included the whole dang definition; we have Free School for instance, Free Country, Free Church, Free City, Free ships, and Free Thought, and of course, Free Speech. We have both things, and acts being prefixed with "Free". Free Software covers both the thing, as well as the act, or mode of production, so it straddles the two.
I think that Free Software can indeed be more confusing than Free Man or Free Country, because your average computer user prolly does not think of Man or country as being objects of commerce the "cost free" meaning does not ever come into play. Most users see the software industry and all the people supposedly getting rich off of it, and the only software they know is that which they buy or is bundled 'free' (ahem, microsoft, ahem) with their machine. And since they prolly do not understand the literary aspects of software development, the idea of Free, in terms of liberated, software never enters their mind, while the "cost free" meaning is the most prevelant one they are exposed to.
But once this literary aspect of software is explained, and that usage of Free propogated (as it is now in all the articles on Linux and Apache and Perl) both meanings become apparent to the user.
-- Craig Brozefsky <cr...@red-bean.com> Free Scheme/Lisp Software http://www.red-bean.com/~craig I say woe unto those who are wise in their own eyes, and yet imprudent in 'dem outside -Sizzla
On 19 Aug 1999 19:55:01 -0500, Craig Brozefsky wrote:
>Kent M Pitman <pit...@world.std.com> writes: >> My impression of Stallman's position is that he wants to reassign the >> meaning of that word to be something else, but *I* think he should get >> another word besides free for what he wants. >It seems that you are intent on denying the meaning that "free" has >had for a very long time, insisting that it means "no cost". As if >the hegemony of political-economy should include not only our >understanding of the creative process thru the blind assumption of the >model of economic incentive as fact, but also our language.
No, he's denying that "free" makes much sense when applied in that sense to inanimate objects.
>Yes, "free" also means, no cost, no money need be paid. But I don't >think that Richard is re-assigning the word to mean something new, it >has always meant liberty, independent action. I think it disingenuous >to claim that his use of the word "free" in this well-established >manner is an attempt to "re-assign" the meaning of a common word. We >all know what a Free Man is afterall.
Right -- it's a man that anyone can copy so long as they distribute the source.
I should add that I think Free Software is just about the most harmless redefinition I've heard of. It works okay, so long as we don't accidentally let people get confused with its obvious (but wrong) meaning as in costless software.
>There is a line of argument that says that the restrictions which the >GPL place on distribution of derivative works disqualified it from >being considered Free, but my impression is that that is hardly the >direction which you were coming from. There are counter-arguments, >and counter-counter arguments to this that I don't really want to >cover, and anyone can look these up in DejaNews on gnu.misc.discuss
Good idea. I've made that argument, but I'm not against the GPL -- I think the GPL is about pragamtism in the pursuit of Free Software, and I don't have anything against pragamtism so long as the people using it are honest.
>I've attached the definition of the word "free" as obtained from >various Free sources, to make clear that "no cost" is hardly the only, >or even dominant understanding of the word. I still like the >typographical convention of capitalizing it when refering to the >higher notion of liberty, and so I try to be consistent in that.
That's very good, and will generally cause people who were about to take it wrong to question whether they really understood.
>I really see no future in the commercial vendors. If Lisp is going to >have any real longevity it is going to be in the Free variants. Franz >might hold out for a decade, maybe more, but I don't see more vendors >joining the ranks of lively Lisp implementors anytime soon, and a >single vendor language is a dead language. But I am not going to turn >around and say that Free Software will "save" lisp. The Free >implementations have their own set of problems that I don't see any >easy solutions for either.
Agreed on all points. Sigh.
>Most importantly, thanx for the HyperSpec Kent.
Again agreed. And thanks for making it free (lower case common meaning), even if you couldn't also make it Free.
> if that had been a necessary condition of its creation, so he got a > local optimum. I had a long conversation about this by phone in which > he told me he was upset that the spec wasn't "free". I explained to > him that it was "free", but not "free" in the meaning he wanted to > assign to that word--more in the sense that the word has traditionally > been assigned: without cost.
Ahem! Just to be exceedingly annoying :) The controversy about "free" is inherent in the evolution of the semantics of the term withn the English language. In Italian (and many other languages) there would be no confusion. (not (equal "Gratis" "Libero")). :)
> My impression of Stallman's position is that he wants to reassign the > meaning of that word to be something else, but *I* think he should get > another word besides free for what he wants.
I disagree. "The (semantic) right thing" would be to stop to use "free" for "without cost".
English is a "worst is better" language. :)
Cheers
-- Marco Antoniotti =========================================== PARADES, Via San Pantaleo 66, I-00186 Rome, ITALY tel. +39 - 06 68 10 03 17, fax. +39 - 06 68 80 79 26 http://www.parades.rm.cnr.it/~marcoxa
> of quality on something. It's the same world that would have color > versions of Casa Blanca at the theaters, and edited for TV versions of > Eyes Wide Shut.
You have "edited for the US" "Eyes Wide Shut".
I.e. things get murky when lawyers get in the middle. I bet Kubrick had "final cut" privileges, but he died before distribution....
Cheers
-- Marco Antoniotti =========================================== PARADES, Via San Pantaleo 66, I-00186 Rome, ITALY tel. +39 - 06 68 10 03 17, fax. +39 - 06 68 80 79 26 http://www.parades.rm.cnr.it/~marcoxa
* William Tanksley | No, he's denying that "free" makes much sense when applied in that sense | to inanimate objects.
* Craig Brozefsky | Well that's why I included the whole dang definition; we have Free School | for instance, Free Country, Free Church, Free City, Free ships, and Free | Thought, and of course, Free Speech. We have both things, and acts being | prefixed with "Free". Free Software covers both the thing, as well as | the act, or mode of production, so it straddles the two. | | I think that Free Software can indeed be more confusing than Free Man or | Free Country, because your average computer user prolly does not think of | Man or country as being objects of commerce the "cost free" meaning does | not ever come into play.
I think it's because in all of the above instances of "Free" you actually refer to _people_ who are free, and that's a much harder leap for Free Software, which is about programmers being freed of the need to ask the author for permission to use their works in any way they like.
(now, since Free Software proponents tend to cast everyone who does not agree with their full philosophy as enemies, let me say that I joined the Free Software movement for the wrong reason many years ago: I thought it was about educating people, not about producing "alternative" software. I have since found other ways that are more effective in educating people than relinquishing the author's rights to his works and removing his say in its use. I also think there are better ways to beat the "hoarders", if that is a worthy goal at all.)
In article <87ogg38e56....@duomo.pukka.org>, Craig Brozefsky <cr...@red-bean.com> wrote: > I really see no future in the commercial vendors.
I disagree.
> If Lisp is going to > have any real longevity it is going to be in the Free variants.
Sorry, I don't want to be rude - so I better say nothing.
> Franz > might hold out for a decade, maybe more, but I don't see more vendors > joining the ranks of lively Lisp implementors anytime soon, and a > single vendor language is a dead language.
Last time I looked people at Harlequin, Digitool, Symbolics, ... etc. were selling and maintaining their Lisps.
> > I really see no future in the commercial vendors.
> I disagree.
I also disagree; I see a bright future for us all; there are a lot of people using lisp. Those who don't see it that way, don't see it. Harlequin's future (in their new identity) re lisp is in their own hands, as well as other commercial vendors, _and_ there is plenty of room and use for free versions as well.
> > If Lisp is going to > > have any real longevity it is going to be in the Free variants.
> Sorry, I don't want to be rude - so I better say nothing.
Heh, heh. I find it amusing to see the term "real longevity" here, as being applied to the second oldest language in existence.
But I think the emphasis in the original statement was on the "Free", rather than the "longevity", and it is true that the free/open software is very popular at this time, so to those involved it may seem as though it might take over all software, not just Lisp. But I am sure that it will never take over 100%, because there will always be many customers of software that would rather not have to deal with their language or their tools in detail, but would rather treat it as a black box, and as long as they get a good value for their money in support of their product, they would rather turn their attention to and put their effort into their application which will make them some money.
> > Franz > > might hold out for a decade, maybe more, but I don't see more vendors > > joining the ranks of lively Lisp implementors anytime soon, and a > > single vendor language is a dead language.
> Last time I looked people at Harlequin, Digitool, Symbolics, ... > etc. were selling and maintaining their Lisps.
Yes, but a decade from now, the nay-sayers will give us another ten years, telling us how dead lisp is and how it won't last ...
-- Duane Rettig Franz Inc. http://www.franz.com/ (www) 1995 University Ave Suite 275 Berkeley, CA 94704 Phone: (510) 548-3600; FAX: (510) 548-8253 du...@Franz.COM (internet)
Back around Jul-1997, I posted the sources to gnu.emacs.sources, and RMS responded. Although the HyperSpec was interesting/neat, he didn't welcome my posting of a HyperSpec interface on that newsgroup, because the HyperSpec itself couldn't be incorporated into FSF sources/control/licensing/whatever. So, further posts for yet another HyperSpec interface might be best kept off of gnu.emacs.sources (unless something has changed).
Well, I recently posted a hyperspec-like browser for Harlequin's online CLIM documentation on gnu.emacs.sources, and I didn't hear anything from RMS. Perhaps he's mellowed on this topic???
Larry
-- Lawrence Hunter, Ph.D. Chief, Molecular Statistics and Bioinformatics National Cancer Institute email: lhun...@nih.gov Federal Building, Room 3C06 phone: +1 (301) 402-0389 7550 Wisconsin Ave. fax: +1 (301) 480-0223 Bethesda, MD 20892 USA
Duane Rettig <du...@franz.com> writes: > Heh, heh. I find it amusing to see the term "real longevity" here, as > being applied to the second oldest language in existence.
Yah, I must admit, I think it's the curse of being a bit on the young side. But I was also talking about commercial vendors, not the language itself. Franz is 16, Harlequin is 14, Digitool is 13, so while these are respectable ages, they don't really compare to the "real longevity" of Lisp, 40 some years. So perhaps the phrase is not as absurd sounding when applied to the commercial vendors of proprietary Lisps, as I originally intended.
> But I think the emphasis in the original statement was on the "Free", > rather than the "longevity", and it is true that the free/open software > is very popular at this time, so to those involved it may seem as though > it might take over all software, not just Lisp.
I don't think it will take over everything, but it's where I feel the most long term stability is for building my own software. It's immune to vendors going bankrupt, or dropping product lines, or making broken changes to their products. So as a programmer, the problems Free Software face seem surmountable, I can contribute to their maintenance and development with my own work, and can never have the code taken away from me. I can also contribute financially, directly to the people responsible for the maintenance of the product, rather than having to go thru a corporation. On the other side, the problems inherent in commercial enterprise are not something I can directly address, except maybe in my capacity as a "consumer" but that is ultimately an empty and meaningless power. I'll admit that this may be an artifact of my present perspective as a software developer working with mostly Free Software tools.
> But I am sure that it will never take over 100%, because there will > always be many customers of software that would rather not have to > deal with their language or their tools in detail, but would rather > treat it as a black box, and as long as they get a good value for > their money in support of their product, they would rather turn > their attention to and put their effort into their application which > will make them some money.
Free Software does not preclude this, ask Cygnus for instance. I'm sure you've heard that argument a thousand times before, but it appears that you still think that Free Software cannot provide for those types of customers. Do you have time to tell us why you feel that way, as I'd be interested in hearing the opinion of someone working on just such a "black box" Lisp implementation. Perhaps there are issues I am unable to see.
> > Last time I looked people at Harlequin, Digitool, Symbolics, ... > > etc. were selling and maintaining their Lisps.
Digitool supports a single platform (very well tho), Harlequin went bankrupt and was bought by a company that doesn't do software development, and Symbolics bobs up and down like a drowning rat. I'm not expecting any more advancements from Harlequin since they relinquished a large part of their Lisp staff, and I am not even sure Symbolics has a staff anymore. I'd be ecstatic if either of those two proved me wrong.
So who besides Digitool and Franz have the resources to move Lisp forward and help it keep up with the changes in the computing environment? How many developers of Lisp implmenetations do these organizations presently support? If there is noone else, can we rely on those two organizations? Do you guys think that is a valid question? I'm genuinely interested in those answers.
> Yes, but a decade from now, the nay-sayers will give us another ten > years, telling us how dead lisp is and how it won't last ...
I'm not nay-saying Lisp tho. The difference is that I don't think it's "dewmed". This does not mean that I think the current situation is one that can carry it along the furthest. I think that directing funds towards the maintenance and development of Free implementations is the most fruitful route, but I don't begrudge Franz or Digitool or others their income.
-- Craig Brozefsky <cr...@red-bean.com> Free Scheme/Lisp Software http://www.red-bean.com/~craig I say woe unto those who are wise in their own eyes, and yet imprudent in 'dem outside -Sizzla
Craig> Duane Rettig <du...@franz.com> writes: >> Heh, heh. I find it amusing to see the term "real longevity" here, as >> being applied to the second oldest language in existence.
Craig> side. But I was also talking about commercial vendors, not the Craig> language itself. Franz is 16, Harlequin is 14, Digitool is 13, so Craig> while these are respectable ages, they don't really compare to the Craig> "real longevity" of Lisp, 40 some years. So perhaps the phrase is not Craig> as absurd sounding when applied to the commercial vendors of Craig> proprietary Lisps, as I originally intended.
How about the oldest language, Fortran? Are there any Fortran compiler companies older than 16? I remember some PC Fortran compilers from about 85 or so. I know IBM has been making Fortran compilers for a long time, but I don't know when the first was was done.
Despite what some people may think about Fortran, I don't think it's not going away soon either. I don't even know of any PC Fortran compiler companies anymore. (Mostly because I don't need one anymore.)
> > Heh, heh. I find it amusing to see the term "real longevity" here, as > > being applied to the second oldest language in existence.
> Yah, I must admit, I think it's the curse of being a bit on the young > side. But I was also talking about commercial vendors, not the > language itself. Franz is 16, Harlequin is 14, Digitool is 13, so > while these are respectable ages, they don't really compare to the > "real longevity" of Lisp, 40 some years. So perhaps the phrase is not > as absurd sounding when applied to the commercial vendors of > proprietary Lisps, as I originally intended.
It would probably be interesting if anyone had any data on ages of comercial or free software vendors, of _any_ product. I don't myself know, but my gut feel is that an average of 14 years over three vendors of the same standard product is pretty good.
> > But I think the emphasis in the original statement was on the "Free", > > rather than the "longevity", and it is true that the free/open software > > is very popular at this time, so to those involved it may seem as though > > it might take over all software, not just Lisp.
> I don't think it will take over everything, but it's where I feel the > most long term stability is for building my own software. It's immune > to vendors going bankrupt, or dropping product lines, or making broken > changes to their products.
You had me with you until the last phrase. Unless the vendor of the software has controls in place for accepting software changes, breakage is inevitable. This is no different in "Free" (i.e. open) products than in "commercial" products.
> So as a programmer, the problems Free > Software face seem surmountable, I can contribute to their maintenance > and development with my own work, and can never have the code taken > away from me.
I also use open software, and changes I make to my own personal copy (i.e. the one that can't be taken away) tend to get dropped on the floor when I update to the next version. Obviously, I have a lot of choices: I can either hope that the new code works better, or I can merge (i.e.re-implement) my changes into the new version, or I can feed back my changes to the vendor (there's that change control structure I was talking about) or I can refuse to upgrade. More and more, as my brain gets filled with my own product and issues thereof, I tend to let "someone else" do the job of maintaining the free/open software - in other words, I treat it as a black-box.
> I can also contribute financially, directly to the > people responsible for the maintenance of the product, rather than > having to go thru a corporation.
Say the word, and I'll give you a bank account number to which you can contribute directly to me :-)
> On the other side, the problems > inherent in commercial enterprise are not something I can directly > address, except maybe in my capacity as a "consumer" but that is > ultimately an empty and meaningless power. I'll admit that this may > be an artifact of my present perspective as a software developer > working with mostly Free Software tools.
You sell yourself short as a consumer. There are times when you have to walk away from a non-responsive vendor, (be it software or household commodities or automobiles). But most of the time, there are many things you can do that might surprise you. Even huge, impersonal companies are sometimes responsive if you know how to approach them.
> > But I am sure that it will never take over 100%, because there will > > always be many customers of software that would rather not have to > > deal with their language or their tools in detail, but would rather > > treat it as a black box, and as long as they get a good value for > > their money in support of their product, they would rather turn > > their attention to and put their effort into their application which > > will make them some money.
> Free Software does not preclude this, ask Cygnus for instance. I'm > sure you've heard that argument a thousand times before, but it > appears that you still think that Free Software cannot provide for > those types of customers.
I don't know why you are making this assumption about my thinking. Perhaps you are thinking in terms of black-and-white wrt Free/open vs Proprietary/closed.
> Do you have time to tell us why you feel > that way, as I'd be interested in hearing the opinion of someone > working on just such a "black box" Lisp implementation. Perhaps there > are issues I am unable to see.
Perhaps it is the shades of gray that are missing from your vision. If you can't see the differences, for example, between Franz and Microsoft, or between Cygnus and FSF, or the similarities between Franz and Cygnus, then the grays are missing.
> So who besides Digitool and Franz have the resources to move Lisp > forward and help it keep up with the changes in the computing > environment? How many developers of Lisp implmenetations do these > organizations presently support? If there is noone else, can we rely > on those two organizations? Do you guys think that is a valid > question? I'm genuinely interested in those answers.
The excellent thing about lisp is that anyone can develop it, without the need for source code. It is like your example above, where you take source code from an open implementation, modify it, and no one can take it away from you (but of course, as in the example, this has its limitations). The point is, many developers of Common Lisp are not necessarily employees of Franz, or of other lisp vendors, or _even_ of any lisp shop. In that sense, lisp is by nature open.
-- Duane Rettig Franz Inc. http://www.franz.com/ (www) 1995 University Ave Suite 275 Berkeley, CA 94704 Phone: (510) 548-3600; FAX: (510) 548-8253 du...@Franz.COM (internet)
Duane Rettig <du...@franz.com> writes: > > Yah, I must admit, I think it's the curse of being a bit on the young > > side. But I was also talking about commercial vendors, not the > > language itself. Franz is 16, Harlequin is 14, Digitool is 13, so > > while these are respectable ages, they don't really compare to the > > "real longevity" of Lisp, 40 some years. So perhaps the phrase is not > > as absurd sounding when applied to the commercial vendors of > > proprietary Lisps, as I originally intended.
> It would probably be interesting if anyone had any data on ages of > comercial or free software vendors, of _any_ product. I don't myself > know, but my gut feel is that an average of 14 years over three > vendors of the same standard product is pretty good.
Yah, I am coming around into agreement on this. I think that you are correct below in that my initial assesment was based on a reasoning which set the Free implementations and the vendors at odds, and so my personal preference for Free Software asserted itself. And in that either/or space, that meant that the vendors would appear to be much worse off than they are. That black/white understanding is not one that I can support once I become aware of it, but I was not vigilant enough to catch it creeping into the background of my argument.
I personally want to work primarily with Free Software, as I think that the mode of production it supports has advantages over proprietary modes. We've heard all the arguments about this already, so I'll leave them out here. For this reason the Free implementations are most important to me. They give me the best place to learn about Lisp, how to implement and use it, and I can always use them in a business setting without having to shell out money (both Free and free is an attractive combination). Combine that sentiment with my misconception of the community as a split between Free and vendors, well, you can see what happened.
My original intent was not to set the Free and Proprietary implementations against one another. That would waste all the effort that went into making ANSI Common Lisp such a full language, as well as the flexibility of the language itself. The better path would seem to be one which uses the advantages of Lisp to support both types of products. I think getting the Free implementations polished would be a useful first step. Then we could build tools that everyone can use.
> > I can also contribute financially, directly to the > > people responsible for the maintenance of the product, rather than > > having to go thru a corporation.
> Say the word, and I'll give you a bank account number to which you > can contribute directly to me :-)
Well, I'm relocating to Albany next month, right next to you guys, and one of the things I wanna try and do once I get settled in, is generate a revenue stream for just this type of work.
> The excellent thing about lisp is that anyone can develop it, without > the need for source code. It is like your example above, where you > take source code from an open implementation, modify it, and no one > can take it away from you (but of course, as in the example, this has > its limitations). The point is, many developers of Common Lisp are > not necessarily employees of Franz, or of other lisp vendors, or > _even_ of any lisp shop. In that sense, lisp is by nature open.
Lisp may be by nature open, but it is not written in a vacuum. If your extension to Lisp is proprietary, I need to either re-implement it entirely (and knowing the way Lispers work, there is little chance I'll agree you did the 'right thing' and be compatible with your product) or I need to produce a competing product. We're presently suffering from this, with no universal GUI, or FFI, or DB interface, or regexp package. Perhaps something like the SRFI mechanism is called for?
This is the way vendors end up cast against one another, and the Free lisp comunity. The best licensing for a tool as far as benefit to the whole of the community is IMO, either public domain or a Free license that is palatable to the commercial vendors. The details of the license are dependent upon the work itself, and how it integrates with existing toolsets. Vendors say they need to reclaim their investment for these works, and that is how they justify proprietary licenses, but we can see these types of products being produced by other language communities with open licenses? How are they managing that?
-- Craig Brozefsky <cr...@red-bean.com> Free Scheme/Lisp Software http://www.red-bean.com/~craig I say woe unto those who are wise in their own eyes, and yet imprudent in 'dem outside -Sizzla
