Andre van Meulebrouck wrote: > I feel frustrated that LISP vendors don't want to target wider audiences by > targeting platforms like Microsoft and dovetailing with them.
> For instance, .Net isn't language centric; how about targeting .Net with CL?
.Net _is_ language centric - please read the CLR specs. .Net only supports a limited object model, essentially the same as Java's object model, with only single inheritance. This already makes it hard to implement several languages on top of .Net. Microsoft's solution is to change language definitions in order to make them fit. For example, "managed C++" is not C++ anymore. The Eiffel implementation on top of .Net comes with its own platform-dependent libraries in order to have a full implementation (so why do they need .Net in the first place?). The people who port Beta to both the JVM and .Net say that there's no real difference. The language neutrality of .Net is just a marketing stunt, born out of Microsoft's desire to beat Java.
Please don't misunderstand me - I think some of your ideas are good. However, you have to remember that companies like Microsoft, Sun, Oracle, IBM and the like can throw loads of money out of the window in order to support and promote their favorite tool of the week. I don't think anyone in the Lisp world can do that. (And I doubt that it would make sense.)
Pascal
-- Given any rule, however ‘fundamental’ or ‘necessary’ for science, there are always circumstances when it is advisable not only to ignore the rule, but to adopt its opposite. - Paul Feyerabend
* Andre van Meulebrouck | I agree that consumers vote with $$ and reputations, but I think they are | voting against LISP not for it.
So much group think. Is there no word for "individual" in your language?
| For instance, .Net isn't language centric; how about targeting .Net with CL?
.NET is too lacking.
| How about adding extensions to Scheme to make it of interest to real | world applications?
Ask someone who cares over in comp.lang.scheme.
| When you see the herd is going Microsoft, target Microsoft technologies: | dovetail CL with COM!
When you see the herd going in any particular direction, choose another.
| I would especially like to see .Net targeted. Microsoft might seem like | the Roman Empire to LISP; but the good news is it is very LISP friendly | (if the LISP world only knew that!).
It may be Scheme friendly (JVM is), but it is not Common Lisp friendly.
Please upgrade your brain with this patch: "Scheme and Common Lisp are not interchangeable languages in /any/ respects". Thank you.
-- Erik Naggum, Oslo, Norway
Act from reason, and failure makes you rethink and study harder. Act from faith, and failure makes you blame someone and push harder.
> > This whole thread, but these bits of this reply in particular, really > > remind me of a part of a speech by James P. Cannon that I read > > recently. He's specifically talking about revolutionists and tired > > once-revolutionsts in the Socialist Workers Party in mid-century US, > > but the point he makes applies to any partisan in any struggle:
> > The surest way to lose one's fighting faith is to succumb to one's > > immediate environment;
> Oh my God! That sounds like the Communist Manifesto!
> Wow, you lost me with that because I detest communism, and any form > whatsoever of Socialism. I'm extremely conservative and anti-communist.
And apparently anti-thought, too. Would you have read what I wrote, complete with disclaimers giving context to the quote, if it had been by a Catholic? I certainly could have dug one up, covering the same topic -- but I kind of expected to get a thought-free response from you.
-- /|_ .-----------------------. ,' .\ / | No to Imperialist war | ,--' _,' | Wage class war! | / / `-----------------------' ( -. | | ) | (`-. '--.) `. )----'
> I feel frustrated that LISP vendors don't want to target wider audiences by > targeting platforms like Microsoft and dovetailing with them.
I'm absolutely sure that the Lisp vendors want to target the widest audience they can. Or at least, I'm absolutely sure that the Lisp vendors want to make lots of money, since they are capitalist money-grabbing slime to a man[1]. What makes you think that they don't?
> For instance, .Net isn't language centric; how about targeting .Net > with CL?
The problem is that it's not trivial to do this (even if .NET wasn't language-centric, which it is). The Lisp vendors have finite money, and they also are concerned with risk: they want to get rich - don't we all? - but they need to be concerned with making sure the possible downsides aren't too bad. Vendors can't simply throw resource at something which may seem like a good idea, because they don't have that resource to throw. They need either to sacrifice something else, or get new resource. New resource means new people, which means money: either they need to burn cash they have in the bank, or they need to get funding. Well, this probably isn't a great time for high-tech companies to go looking for funding, so let's rule out that last one. Now look at what happens if they fail - either .NET fails, or they don't get any sales for their .NET product, or they fail to complete the port before the money runs out. Another dead Lisp company.
You have to be very convinced of something before betting the livelyhoods of yourself and many of your friends on it. I have an idea (not a software or IT idea, but I can't tell you what it is since I don't have the patent yet) which, if it succeeds, will likely make me rich and (more likely) make a lot of other people rich, *and* make the world a better place. I should just go for this right? Well, I had this idea a couple of years ago, and I haven't done anything yet. Why not? Well, first I have to pay for the patent (cheap enough (but not trivial) to get, but expensive in time to write the proposal), then I have to do a whole lot of development work to get a demo of the thing, then I have to hawk it around companies to try and get them to license it. Maybe 50,000 to 100,000 pounds investment. I could mortgage my house, perhaps. But I get rich! I should just do it! Well, *maybe* I get rich, maybe I get ripped off and have to defend the patent against Megacorp PLC who have some really good lawyers, and I lose my house. What would you do.
(mind you: if anyone wants to invest, you know where to contact me (:-)).
> How about adding extensions to Scheme to make it of interest to real world > applications?
Matthew Danish wrote: > I very much agree. You can imagine my surprise when I tried to do:
> for item in vec[:].sort():
> in Python. ... > Convenience. Let us consider the above example again; instead of > writing that I have to write:
> new_vec = vec[:].sort() > for item in new_vec:
Er, you don't mean that. You mean (alas!)
new_vec = vec[:] new_vec.sort() for item in new_vec: [stuff]
The Python answer to this is very similar to the Lisp answer to "Lisp is no good because doing X is verbose": you can make it less verbose at a one-off cost in code. (And, in Python but not in Lisp, a small cost in efficiency. No macros.)
def sorted_copy(x): result = x[:] result.sort() return result
for item in sorted_copy(vec): [stuff]
Guido van Rossum considers it a *good* thing that lists' "sort" method doesn't return the list, because that might encourage some people to think it returns a sorted *copy* instead of mutating the list. I'm no more convinced than I expect everyone else in c.l.l to be :-), but it is a deliberate choice rather than a random blunder.
-- Gareth McCaughan Gareth.McCaug...@pobox.com .sig under construc
"Erann Gat" <g...@jpl.nasa.gov> wrote in message news:gat-1011020951080001@192.168.1.51... > > You seem to blame it on the nebulous "political" problems, ideas > > or the pressures of failing projects and tight timelines. But > > from reading it it seems quite obvious that the problems are > > from personal failings of the development groups at JPL.
> Oh dear, I hope it's not obvious because that's definintely not true. JPL > is full of very smart, very good, very busy people. But many of them are > not computer scientists, they are *real* scientists -- the kind who do > science. Or they are "rocket scientists", which is to say, spacecraft > engineers. Many of them know just enough about programming to be > dangerous (much like I know just enough about spacecraft to be dangerous) > :-) The situation is a microcosm of the world at large. The problems > really are political, which is to say, they arise from the complex > dynamics of many people working together to try to get something very hard > done. There is almost nothing going on that I would attribute to anyone's > "personal failings." Or perhaps a better way to say that would be that > everyone has personal failings, including myself, and they all contribute > more or less equally to the situation.
But the top manager getting up and esposuing that getting rid of Lisp is the one thing that he would like to improve things shows a lack of introspection and thoughfulness. Why would ANY manager say that at any time in a public forum (it's akin to flogging). The only thing I can think of is a personal failing, a failing to get his mouth under control and his brain in gear. Perhaps you can come up with some excuse?? From your writing you are harder on yourself than other people and it seems that you do want to use the words that accurately describe poor human behaviour. You are a target because of that those very behaviours, they are using it against you. Quit giving them excuses and a way out that justifies their stupid behaviour in the face of difficulties. Its in the difficult situations that a person's real character comes out. What is theirs? (Sounds like they eat their young).
Don't take that crap from anyone, even real scientists. It is also not true that everyone more or less contributes equally. Do your personal failings make the politics worse? It sounds like you try to learn from them, as opposed to your co-workers. Ignorance is no excuse.
> > Their budgets may have been tightened because they are viewed > > as being incompetent and failures by the powers that be.
> No, quite the opposite happened on DS1 actually. The budget was tightened > because management believed we could get the job done with fewer resources > (they were wrong, but that's another story).
> > With the > > inability of the "Top" managers to perform and without having > > reasonable explanations as to why, the people in control of the > > money have resorted to just financially squeezing them until > > they either find a way (and smarten up) or fail miserably.
> Well, the "top managers" *are* the people in control of the money, so that > doesn't really make any sense. What would you expect them to do > differently?
I thought the people in control was the Congress (or whatever other branch of government). Don't the managers just budget and plead their case before some committee? In light of the recent failures (spacecraft missing their orbits, or crashing into their destination, or the boondoggle with the space station) this would certainly give me pause.
> "Brian Palmer" <br...@invalid.dom> wrote... > > Andre van Meulebrouck wrote:
> > > I feel we have an obligation to serve.
> > You presented your talk and paper not out > > of a sense of selfless service but to sell to others your own selfish > > view of how you think things should be done in the hopes that others > > will find it useful and will join in to promote /your/ cause, which > > ultimately benefits you.
> I disagree with your second guessing of my motives and wonder how you think > you can make such determinations. (It seems very presumptive to me.)
Your motives don't concern me. It's your statement. What you have done is not the act of a servant so you have contradicted yourself. One who is a servant acts on the orders of a master. You acted on the desires of your self. Hence, this is far from serving or being a servant.
> The purpose of conferences is to present ideas as food for thought. Take it > as that. Take it or leave it.
Exactly. So why this socialist crap about being a servant? That you feel you have an obligation to serve suggests a guilt complex. To not serve would mean that you have neglected your duties.
> As to who is ultimately to benefit: it would be the world at large first > and foremost (a world drowning in needless complexity) and the LISP > community secondarily. A win for LISP would be a tide that floats all > boats.
> My only benefit would be less frustration as the tools get better.
So you presented your talk and wrote your paper for the benefit of the world at large, and the community second. I thought you "detest communism, and any form whatsoever of Socialism". Yet here you espouse the exact views of the ultimate socialist -- your work is to benefit the world, with benefit to yourself being merely a side effect. How does this fit into the notion of being "extremely conservative"?
> > This isn't service -- it's salesmanship.
> There is an element of both. (There darn well should be!)
So why call it an "obligation to serve"? If you feel like being a true servant, your needs shouldn't enter into the equation at all.
> I have no problems with salesmanship and wish the LISP community didn't > either!!! =:0)
Well, this is a different matter altogether. To whom do you refer when you mention the Lisp community? Is this the faceless masses you are unable to identify who have somehow not lived up to your notion of what you think should be done? Doesn't that seem very presumptive to you?
> > If > > you truly feel the obligation to serve, then you should post a message > > saying "tell me what to do and I'll do it".
> Why do I need others to tell me what to do?
That is the nature of being a servant, of feeling the obligation to serve others. Or have you forgotten what you wrote:
"I feel we have an obligation to serve. But if being of service means being a target; I no longer feel any obligation to serve (at least not in that type of venue). There is a difference in being a servant and being an abused toadie."
Erik Naggum <e...@naggum.no> writes: > Thank you for pointing out the misuse of "take" over "make", a remnant of > my native tongue which I now cannot imagine how slipped in there.
AFAIK, "to take a decision" is perfectly standard British English. Americans tend to use "make" instead.
> You are very CL-centric! I am very Scheme centric. > If I was [sic] a professor, I'd use Scheme as the lingua franca > MzCOM. This is Rice's Scheme wrapped up as a COM object. > ... I intend to show examples in both Scheme and Javascript. > In addition I'm working on a Scheme implementation in Javascript coded in a > purely object oriented fashion. > I've done it in Javascript. I'd like to be able to do it using Scheme as a > scripting language one day! > ... I would like to code all the interpreters and compilers in both > Scheme and Javascript... > Or, maybe a Scheme that has more facilities could fit the bill. > Until then, I'll stick with Scheme: it's free and does just fine for my > purposes. > Hopefully with Rice getting Microsoft funding; perhaps they can come up with > a Scheme solution...
Dude, you're in the wrong newsgroup. Unsubscribe from comp.lang.lisp and subscribe to comp.lang.scheme -- these newsgroups are separated for a reason.
> 1) Many only use CL's object system for big applications. For smaller > applications don't a lot of people revert back to writing procedural code?
If you have to ask this, then clearly you don't know. Therefore, your statement that many use CLOS only for big applications is a statement made out of ignorance.
> "Alain Picard" <apicard+die-spammer-...@optushome.com.au> wrote in message > news:86isz5flxf.fsf@ibook.local.... >> "Andre van Meulebrouck" <vanme...@earthlink.net> writes: > 1) Many only use CL's object system for big applications. For smaller > applications don't a lot of people revert back to writing procedural code?
Yes. And so what? Not every application and/or part of an application is programmed best in an OO style. I speak from experience here as I have programmed in pure OO languages (Smalltalk, anyone?) and - surprise - even within the Smalltalk pure-OO world, a lot of the code looks suspiciously procedural.
> 2) Many Scheme programmers write procedurally.
And many write functionally. Again, so what?
> #2 is due to: if you don't have a pervasive, simple, built in object > #system > and you have to roll your own, that's often when happens (or people roll > their own in very different ways).
It doesn't matter. Most of the OO systems that have been rolled have similar capabilities. It's the thinking, not the standardization that matters.
> It seems like the best thing to add to a Windows network is another > Windows > machine if you want the least amount of headaches. That's the problem. > Integration!
Integration is a problem. But it's no insurmountable and only has to be done once. Most people don't realize that this "Integration" is only a small part of any particular application. Tying yourself down to a less productive language because a small amount of the task is simple to do in it is extraordinarily sub-optimal. In fact, it's the lazy and slow thing to do rather than the smart and fast thing.
> No, that's not what I meant. I meant that many LISP programmers look down > their nose at conventional technologies as if they are too far beneath the > dignity of LISP programmers and unworthy of the attention of LISP > programmers.
Well, yes. In most cases, the technologies are less than worthy because most of them have already been done in Lisp and either accepted or rejected for good reason. Why don't you consider that most of these languages are re-inventing wheels and ask why they waste their time doing that?
> Many LISP programmers come > across that way to people in the conventional world. That's what I'm > talking about.
Why? Because we refuse to fawn upon programmers who re-invent odd-shaped wheels and start cults of personality to promote them? thanks, but I'll stick with the round one I have.
> I know some LISP programmers that wouldn't dream of working with Microsoft > technology and other "inferior" technologies; or they do so but they > complain vehemently and constantly about how "brain dead" those "silly" > technologies are.
First of all, most of us "LISP programmers" have day jobs where we use "standard technology" - including Microsoft "technology" - and know full well the benefits and limitations therein. In fact, I'd venture to say that the knowledge about non-Lisp technology in the Lisp community is much higher than the knowledge of Lisp technology in the non-Lisp using community. Why do you believe that our arguments stem from ignorance of the technology?
> That just isn't going to open doors anywhere! That isn't being what I'd > call a good "ambassador for LISP".
Well, you're reinforcing the opinion of this Lisp community member that non-Lispers are a bunch of rubes that come in and spout off without really knowing the territory. I don't feel that I lose much by not knowing this kind of person. You're a lousy ambassador for your viewpoint, too. So there. Now can we stop the namecalling and talk about (a) why you think that Lisp needs to be promoted, (b) why you think that the current Perl-/VB-using person is a good target for evangelization, and (c) what technological items you believe that Lisp lacks and why we need them?
* Erik Naggum | Thank you for pointing out the misuse of "take" over "make", a remnant of | my native tongue which I now cannot imagine how slipped in there.
* Doug McNaught | AFAIK, "to take a decision" is perfectly standard British English. | Americans tend to use "make" instead.
FWIW, this did /not/ help. I pronounce my R's in the right place, and I'm damn proud of it!
-- Erik Naggum, Oslo, Norway
Act from reason, and failure makes you rethink and study harder. Act from faith, and failure makes you blame someone and push harder.
> My spelling of LISP is not telling at all. I spell it like that from a > linguistic perspective because it is an acronym and I believe it should be > capitalized regardless what the trends are (and they vary).
Whatever happened to your philosophy:
> When you see the herd is going Microsoft, target Microsoft technologies
What should you do when you see the herd is spelling it Lisp? Or do you pick and choose which trends you care to follow?
Doug McNaught wrote: > AFAIK, "to take a decision" is perfectly standard British English. > Americans tend to use "make" instead.
Hmmm. I'm not sure about the use of infinitive form as this sounds rather odd. However, in the context of the original message, this is within normal correct usage.
Andre van Meulebrouck wrote: > On the server side, the situation is even rosier for LISP via Microsoft. > (On a server, you can do anything you want!)
> .Net is not language centric: everything compiles to CLR and there is a > common type system. The common type system means you can catch exceptions > from a module written in a diffent language, and you can extend it (etc.).
> In .Net, you can use Scheme (it has a CLR compiler, created by > Northwestern > University). And ML, Haskell, and a lot of other languages are supported > (by 3rd party vendors).
So why don't you just go ahead and use and promote it? You seem to think that Lisp & Scheme are interchangeable.
Listen -- I've looked into compiling Common Lisp into the CLR. In the final analysis, the impedence mismatch is just too great (less than into the JVM, but still too great). The Lisp that you get out of it is so mutated that it doesn't look much like Common Lisp. How do you do keyword arguments? How do you deal with mapping betwwen CLR types like Integer32 and CL fixnums? How do you pass items to and from the Lisp world without making either the code in the non-Lisp world do tons of conversion or adding tons of type information to the Lisp world? It just doesn't work out. You'd be better off taking an approach like Corman Lisp does, exposing the Lisp environment as a large COM object and passing string objects back and forth (or talking over streams).
I guess that my biggest gripe about your approach is that a lot of folks like you with no real skin in the Lisp game keep coming around telling people who do what to do to make their life better. All of them "love Lisp and want to make it succeed", too. I know. I've been there. It's a stage that many of us go through. Luckily, for most of us, it passes.
If you want to actually DO something constructive, i have a few ideas... How about contributing to make the Lisp world a better place? Write some code that performs a useful function. Send SBCL or CMUCL or CLisp or Roger Corman some code to help their systems become more standards compliant. Start with one of the open source implementations and show us how to compile to and integrate the resulting code into the cLR. But for God's sake, shut up and do something useful, rather than prattling on.
Kenny Tilton <ktil...@nyc.rr.com> writes: > Your reflex reaction to tfb's quote blinded you to its import, viz, > that you are fixated so badly on the moment that you cannot see the > trend I keep pointing out: the "herd" is voting with their hooves > against the Fortran branch of evolution and for the Lisp branch, by > jumping all over Python, Ruby, Perl, and now thx to you I will add > JavaScript (well, I don't actually know if the herd digs that, but at > least its designers seem to have discovered Lisp).
I havwe to admit, I was quite impressed to see that the CL community (the free-CL-on-Unix subset of that community, even) is in such a great state (as perceived by JavaScript users) that they want to piggyback off our success :-)
Andre van Meulebrouck wrote: > For instance, .Net isn't language centric; how about targeting .Net with > CL?
Not language-centric? Where's the support for multiple inheritance? Where's the support for integers larger than 64 bits in size? How do I access a global symbol in the .NET CLR? How do I add a new type of method combination to my C# member function? The CLR may be *less* language-centric than the JVM, but it still demures to the tyranny of the (linguistically speaking) norm.
"Andre van Meulebrouck" <vanme...@earthlink.net> writes:
> In fact, I dare say Javascript is quite the thing to know: if something > horrible happens, people will want to get on the internet to find out all > that they can; and Javascript is the most prevalent language used on the > internet.
If something horrible happens, there won't _be_ an Internet.
(Some might argue that that in itself is not so horrible, but ...)
* Daniel Barlow wrote: > If something horrible happens, there won't _be_ an Internet.
There won't be a web anyway. I was in Denmark on the 11-Sep-2001, and I went back to my hotel and got the news from teletext and broadcast TV, because there were no reachable news websites at all that I could find.
Erik Naggum <e...@naggum.no> wrote: > * Erann Gat > | And "not altogether positive" is just a diplomatic way of saying that > | it's been a fucking disaster. > What remains to be explained, however, is why "Lisp" gets blamed.
Yes. It's certainly not clear from your (Erann's) summary that there was *any* problem caused by Lisp or the use of Lisp, or that "sending it into space" had much to do with the JPL brass opinion of the language.
If there's not some other side that we're missing here, it would seem that the decision not to use Lisp at JPL, was not the result of sending Lisp into space, or anything whatsoever to do with Lisp, the programers using it there, or the software they produced with it, but plain and simple ignorance and FUD.
Erik Naggum <e...@naggum.no> writes: > * Erik Naggum > | Thank you for pointing out the misuse of "take" over "make", a remnant of > | my native tongue which I now cannot imagine how slipped in there.
> * Doug McNaught > | AFAIK, "to take a decision" is perfectly standard British English. > | Americans tend to use "make" instead.
> FWIW, this did /not/ help. I pronounce my R's in the right place, and > I'm damn proud of it!
Hmm, although some Brits use perverse pronunciation, not all of them do. And South Asians speak English quite well, using essentially a British dialect.
-- /|_ .-----------------------. ,' .\ / | No to Imperialist war | ,--' _,' | Wage class war! | / / `-----------------------' ( -. | | ) | (`-. '--.) `. )----'
> .Net is not language centric: everything compiles to CLR and there is a > common type system. The common type system means you can catch exceptions > from a module written in a diffent language, and you can extend it (etc.).
Well, if it is not language centric (I would argue that it is), it certainly is bent on OO methodology.
> In .Net, you can use Scheme (it has a CLR compiler, created by Northwestern > University). And ML, Haskell, and a lot of other languages are supported > (by 3rd party vendors).
There should be a big disclaimer to this - both Northwestern and PLT (they put out an ad for a full-time two year position to port DrScheme to .Net a couple of months ago) received /very/ (well, compared to anything they usually receive) large grants from Microsoft to do just that. I suspect this is also the case for some of ther other .Net projects.
> .Net also features the best garbage collector ever created (according to the > person in charge of that department at Microsoft and I have no reason to > doubt him). LISP gurus were called in to help craft it (sorry, but I won't > dare drop any names here as I don't think it would be a good idea to do so!)
I think this claim is BS, especially so since it comes from the "person in charge of that _department_." At the very least it will make sure all languages run equally slowly. At the worst it will be biased towards VB and C#.
> So, it would appear that everything is in place for LISP to be very healthy > because of Microsoft.
Maybe Lisp will benefit slightly from some of the side-effects of .Net (I don't see how), but I think Microsoft is more interested in cramming VB and C# down everyone's throats (their IDEs are very expensive).
> I am pretty current on what's going on at Microsoft however; because that's > where I think the big break for LISP is most likely to happen. And on that > count, I'm very optimistic!
I'd keep my eye out more on Microsoft's .Net and Palladium shenanigans if I were you.
