There's an interesting post made by Philip Greenspun on his weblog (http://blogs.law.harvard.edu/philg/2003/09/20#a1762). It is based on his experience in teaching seniors and graduate students in MIT's computer science program and he compares programming in Java to driving an SUV:
"A project done in Java will cost 5 times as much, take twice as long, and be harder to maintain than a project done in a scripting language such as PHP or Perl. People who are serious about getting the job done on time and under budget will use tools such as Visual Basic (controlled all the machines that decoded the human genome). But the programmers and managers using Java will feel good about themselves because they are using a tool that, in theory, has a lot of power for handling problems of tremendous complexity. Just like the suburbanite who drives his SUV to the 7-11 on a paved road but feels good because in theory he could climb a 45-degree dirt slope. If a programmer is attacking a truly difficult problem he or she will generally have to use a language with systems programming and dynamic type extension capability, such as Lisp. This corresponds to the situation in which my friend, the proud owner of an original-style Hummer, got stuck in the sand on his first off-road excursion; an SUV can't handle a true off-road adventure for which a tracked vehicle is required."
So, does this mean that Lisp is the Bradley Fighting Vehicle of programming tools?
bill_clement...@yahoo.com (Bill Clementson) writes: > capability, such as Lisp. This corresponds to the situation in which > my friend, the proud owner of an original-style Hummer, got stuck in > the sand on his first off-road excursion; an SUV can't handle a true > off-road adventure for which a tracked vehicle is required.
This is might be offtopic, but:
Maybe the person did not know how to drive the Hummer in off-road conditions? I remember once reading an article, where the offroad capabilities of Hummer were described -- the test driver could not pass some obstacle (a ditch). With a little help from more experienced driver (or was it a teacher in Hummer's drivings school, I don't remember) he could pass that very obstacle without problems. The trick was to get all-wheel driving to work -- using both accelerator and brake simultaneously. I guess this same trick would help to get the poor man out of the sand, too.
This relates to Lisp and programming in general because good programmers should know their tools. And the ones who do will get better results with tools they know than with some presumably better tools they do not know.
"... my personal preference is a much darker, uglier, and more shameful secret: Common Lisp, CLOS, plus an ML-like type inferencing compiler/error checker (with some things done in a sublanguage with Haskell semantics and Lisp syntax). Common Lisp dates from around 1982 and ML from 1984. I try to keep this preference concealed from young people who've been raised on a diet of C, Java, C#, Perl, etc. They just wouldn't find it credible that 20-year-old systems and ideas are actually better than the latest and greatest from Microsoft and Sun. Imagine my delight in running into a friend yesterday. She's a 23-year-old graduate student in computer science at Harvard. Conversation rolled around to programming tools. Unprompted she said 'What I think would be best is Common Lisp Object System with a modern type system'. I was stunned. I thought it was only dinosaurs like me that clung to Lisp. I had a second ephiphany for the week... Believing that Lisp circa 1982 plus some mid-1980s ML tricks thrown in is better than all of the new programming tools (C#, Java) that have been built since then is sort of like being a Holocaust denier."
> "... my personal preference is a much darker, uglier, and more > shameful secret: Common Lisp, CLOS, plus an ML-like type inferencing > compiler/error checker (with some things done in a sublanguage with > Haskell semantics and Lisp syntax). Common Lisp dates from around 1982 > and ML from 1984. > I try to keep this preference concealed from young people who've been > raised on a diet of C, Java, C#, Perl, etc. They just wouldn't find it > credible that 20-year-old systems and ideas are actually better than > the latest and greatest from Microsoft and Sun. > Imagine my delight in running into a friend yesterday. She's a > 23-year-old graduate student in computer science at Harvard. > Conversation rolled around to programming tools. Unprompted she said > 'What I think would be best is Common Lisp Object System with a modern > type system'. I was stunned. I thought it was only dinosaurs like me > that clung to Lisp. > I had a second ephiphany for the week... Believing that Lisp circa > 1982 plus some mid-1980s ML tricks thrown in is better than all of the > new programming tools (C#, Java) that have been built since then is > sort of like being a Holocaust denier."
I would bet that LISP will still be there in 20 or 30 years, while Java and C++ will have long been forgotten (ie. reached a bellow FORTRAN and COBOL level). Perhaps old C++ programmers will have a last flare in 2037.
Pascal> I would bet that LISP will still be there in 20 or 30 years, while Pascal> Java and C++ will have long been forgotten (ie. reached a bellow Pascal> FORTRAN and COBOL level).
That doesn't really say that much because there are still tons of COBOL and respective amounts of FORTRAN out there.
I have the theory that no language who has ever made it out of the lab will ever truly die. There are probably still professional programmers doing stuff in APL, PL/1, RPG and FORTH even if it is only on a small scale.
I do agree however that Lisp will only grow stronger over the years. New languages will flame up and sweep over the masses and then fade back into near nothingness, but each time they will add a bit more of Lisp to mainstream programming and in the end the masses will have matured to a level where they will be ready to embrace Lisp.
C++ added object orientation, Java has added the concept of garbage collection, Extreme Programming is popularizing dynamic languages and some of the stuff going on in the OO community (such Aspect Oriented Programming) will add more sophisticated OO technology such as multiple inheritance and generic functions.
It kind of feel goods to already be there, doesn't it :-)
------------------------+-------------------------------------------------- --- Christian Lynbech | christian #\@ defun #\. dk ------------------------+-------------------------------------------------- --- Hit the philistines three times over the head with the Elisp reference manual. - peto...@hal.com (Michael A. Petonic)
> I would bet that LISP will still be there in 20 or 30 years, while > Java and C++ will have long been forgotten (ie. reached a bellow > FORTRAN and COBOL level). Perhaps old C++ programmers will have a > last flare in 2037.
You say that is if C++ is some kind of fad. C++ has been widely used for nearly 20 years now, and doesn't show any signs of slowing down. Now, whether C++ will be around for another 20 years really depends on what kind of programming you're considering. The vast majority of software is composed of "back-end" applications. These are the ones following the trends of whatever language is most hyped at the moment. However, if you look at the apps that most users (games, office suites, productivity tools, etc) interact with on a dialy basis, they're all still written in C or C++. The major application developers (aside from Microsoft, which is forced to eat its own dog-food) haven't embraced Java/C#, and show no signs of doing so.
So, the merits of the language aside, its unfair to group C++ into the same boat as Java. It is clearly not just a fad. Unless you're one of those people who considers this whole Unix/C thing to be one big 3 decade long fad that's bound to end Real Soon Now (TM)...
> The major application > developers (aside from Microsoft, which is forced to eat its own > dog-food) haven't embraced Java/C#, and show no signs of doing so.
Do you have any statistics to back this? There *are* indeed lots of Java developers around, at least that's my impression from the publicity and the amount of web space devoted to Java.
>>>>> On Tue, 23 Sep 2003 09:47:53 +0200, Christian Lynbech ("Christian") writes:
Christian> There are probably still professional programmers doing Christian> stuff in APL, PL/1, RPG and FORTH even if it is only on a Christian> small scale.
There's a fair amount of APL and FORTH going on; I wouldn't call it a "small scale". Probably a lot more than Common Lisp!
> "Rayiner Hashem" <heli...@mindspring.com> schrieb > > The major application > > developers (aside from Microsoft, which is forced to eat its own > > dog-food) haven't embraced Java/C#, and show no signs of doing so.
> Do you have any statistics to back this? There *are* indeed lots of Java > developers around, at least that's my impression from the publicity and > the amount of web space devoted to Java.
Sure, they need lots of Java programmers to do what one Common Lisp programmer can do!
Ahem, sorry, couldn't resist... (BTW it's true) OK back to boxes and glue...
heli...@mindspring.com (Rayiner Hashem) writes: > So, the merits of the language aside, its unfair to group C++ into the > same boat as Java. It is clearly not just a fad. Unless you're one of > those people who considers this whole Unix/C thing to be one big 3 > decade long fad that's bound to end Real Soon Now (TM)...
Rayiner> You say that is if C++ is some kind of fad. C++ has been widely used Rayiner> for nearly 20 years now, and doesn't show any signs of slowing down.
That is probably true. On the other hand I think[1] that C++ is in a disappointing state of affairs. If C++ had been the success it alluded to be, Java would never have become the big hot thing it seems to have become.
Java is perhaps only on par with C++ right know, but given C++'s 10-15 years headstart plus it being the natural successor to the previous big thing (that being C) one may wonder what went wrong for the C++ guys.
[1] and this is really just thinking, I haven't any statistics and I am not a C++ programmer so I do not know whats happening in the C++ community
------------------------+-------------------------------------------------- --- Christian Lynbech | email: christian #\@ defun #\. dk ------------------------+-------------------------------------------------- --- Hit the philistines three times over the head with the Elisp reference manual. - peto...@hal.com (Michael A. Petonic)
Christian Lynbech <christian.lynb...@ericsson.com> writes: > There are probably still professional programmers doing stuff in > APL, PL/1, RPG and FORTH even if it is only on a small scale.
In addition to all the Say It Loud uses of Common Lisp, there are quiet ones, too. Same goes for at least some of the above .. wanna play with Forth? Get a Mac, and hold down Open-Apple+O+F when it boots.
(What's up with calling Open-Apple "Command"? It still looks like an open apple to me, and PC users understand "Open Apple", but "Command"? How do either of those symbols mean that?)
-- /|_ .-----------------------. ,' .\ / | No to Imperialist war | ,--' _,' | Wage class war! | / / `-----------------------' ( -. | | ) | (`-. '--.) `. )----'
"Marc Battyani" <Marc.Batty...@fractalconcept.com> writes: > "Robert Klemme" <bob.n...@gmx.net> wrote
> > "Rayiner Hashem" <heli...@mindspring.com> schrieb > > > The major application > > > developers (aside from Microsoft, which is forced to eat its own > > > dog-food) haven't embraced Java/C#, and show no signs of doing so.
> > Do you have any statistics to back this? There *are* indeed lots of Java > > developers around, at least that's my impression from the publicity and > > the amount of web space devoted to Java.
> Sure, they need lots of Java programmers to do what one Common Lisp > programmer can do!
> Ahem, sorry, couldn't resist... (BTW it's true) > OK back to boxes and glue...
Or, y'know, one Pascal programmer.
PS You couldn't pay me enough money to reimplement TeX in Pascal. As for CL, well, double my rent would do...
-- /|_ .-----------------------. ,' .\ / | No to Imperialist war | ,--' _,' | Wage class war! | / / `-----------------------' ( -. | | ) | (`-. '--.) `. )----'
> Do you have any statistics to back this? There *are* indeed lots of Java > developers around, at least that's my impression from the publicity and > the amount of web space devoted to Java.
There are tons of Java developers, yes. But they work on different types of apps from what I'm talking about. I made it clear that I was referring to the types of applications that people buy in shrink-wrapped packages from CompUSA, like Photoshop, Office, Quake III, etc. These are still written mostly in C and C++.
> That is probably true. On the other hand I think[1] that C++ is in a > disappointing state of affairs. If C++ had been the success it alluded > to be, Java would never have become the big hot thing it seems to have > become.
It was inevitable. C++ is a systems programming language. The scope of applications for which C++ is suited is much narrower than the scope of applications for which Java (or more appropriately, a real high level language) is suited. The reason Java has become so much of a success is because it embraced a new direction for computer languages --- giant class libraries and network computing.
> Java is perhaps only on par with C++ right know, but given C++'s 10-15 > years headstart plus it being the natural successor to the previous > big thing (that being C) one may wonder what went wrong for the C++ > guys.
Oh, Java will undoubtedly become much larger, in numbers, than C++. But I doubt Java will have longevity. C++ will be like C --- Doom IX will still be programmed in it.
t...@famine.OCF.Berkeley.EDU (Thomas F. Burdick) writes:
> play with Forth? Get a Mac, and hold down Open-Apple+O+F when it > boots.
Or a Sun (since Open Firmware a.k.a. Openboot is a Sun invention).
> (What's up with calling Open-Apple "Command"? It still looks like an > open apple to me, and PC users understand "Open Apple", but "Command"? > How do either of those symbols mean that?)
Because you issue 'commands' with it, I guess. I call it "The Museum Key". -- (espen)
> > Do you have any statistics to back this? There *are* indeed lots of Java > > developers around, at least that's my impression from the publicity and > > the amount of web space devoted to Java. > There are tons of Java developers, yes. But they work on different > types of apps from what I'm talking about. I made it clear that I was > referring to the types of applications that people buy in > shrink-wrapped packages from CompUSA, like Photoshop, Office, Quake > III, etc. These are still written mostly in C and C++.
I didn't read it that way. Thanks for clarification!
heli...@mindspring.com (Rayiner Hashem) wrote in message <news:a3995c0d.0309231942.24617bfc@posting.google.com>... > Oh, Java will undoubtedly become much larger, in numbers, than C++. > But I doubt Java will have longevity. C++ will be like C --- Doom IX > will still be programmed in it.
My thoughts on this aren't totally on topic, but I think when talking about the future of languages we also need to look at where things are headed.
Of Java's longetivity, time will tell. It's well situated now and has a huge user base.
On C/C++ I think you are right in that the shrink wrapped applications (Doom, Quark, etc.) all use it now and may very well continue to do so in the future. But if any language were to be tied solely to that wagon, then it will _not_ have much of a future, simply because the future for those types of applications is limited.
The whole application/document split is slowly but surely disappearing. More and more users don't understand the difference between running an application and using a web page. And more and more applications are some sort of web page or web service. I'm not sure of the timeframe, but there may be a day when C/C++ will be used for systems programming _only_ because the shrink wrapped applications of today have transmogrified to become some sort of web thingy.
This trend is in part why so many new languages are being developed and old ones (like Lisp) are being unearthed. You can write great web server stuff with Lisp. As applications move online you run into tricky problems involving concurrency and scope that don't fit well to traditional programming methods. And thus people are looking for more flexible languages. Maybe later, when the problem space is better understood some programmers will prefer less flexible languages that are better mapped to what they are doing.
Espen Vestre <espen@*do-not-spam-me*.vestre.net> writes: > t...@famine.OCF.Berkeley.EDU (Thomas F. Burdick) writes:
> > play with Forth? Get a Mac, and hold down Open-Apple+O+F when it > > boots.
> Or a Sun (since Open Firmware a.k.a. Openboot is a Sun invention).
One of my favorite quotations from the FreeBSD crowd:
Jordan Hubbard: We have a crash bug. It needs to be fixed. We DO NOT need to know how to print 3000 spaces in 11 different languages! :-) Daniel Sobral: I concur. But if anyone wants to do it with loader, : 3kbl 3000 0 do bl emit loop ; 3kbl will do the trick.
Seems that lots of people like Forth for loaders.
-- Fred Gilham gil...@csl.sri.com I was storing data in every conceivable way, including keeping a chain of sound waves running between the speaker and the microphone. There was no more memory left to be had....
On 24 Sep 2003 13:31:36 -0700, cperk...@medialab.com (Chris Perkins) wrote:
>The whole application/document split is slowly but surely >disappearing. More and more users don't understand the difference >between running an application and using a web page. And more and more >applications are some sort of web page or web service. I'm not sure >of the timeframe, but there may be a day when C/C++ will be used for >systems programming _only_ because the shrink wrapped applications of >today have transmogrified to become some sort of web thingy.
I don't believe this for one second. The only users who can't tell the difference between an app and a web page are ones who haven't used either. New users pick up on the difference very quickly.
I also don't think the convergence of look and feel is necessarily a good thing - IMO users should always be aware of whether they are crossing machine boundaries - for security and privacy reasons if for nothing else. Such awareness also explains perceived performance differences.
WRT personal software going web: imagine you are on deadline, sitting at your (budget friendly) web terminal, just finishing your job critical report using the MSWebWord service that your company leases (at an irresistable price) from some local ISV. Now imagine a moth flies into the server ...
> I don't believe this for one second. The only users who can't tell > the difference between an app and a web page are ones who haven't used > either. New users pick up on the difference very quickly.
The point I'm trying to make is that users don't care or need to make the distinction. For example, last year I saw a newspaper article listing the "Top 10 Utilities" for computers. All 10 had URLs, but 5 of them were applications that you downloaded/installed/ran, and 5 were web pages you visited and used. They were just listed among each other because they were all useful or interesting. The distinction as app vs. web app was not noticed by the writer and presumably the readers.
> I also don't think the convergence of look and feel is necessarily a > good thing - IMO users should always be aware of whether they are > crossing machine boundaries - for security and privacy reasons if for > nothing else. Such awareness also explains perceived performance > differences.
I tend to agree. Windows XP, for example, is really just Internet Explorer on steroids. It uses the IE HTML renderer to display nearly everything that the OS shows the user (directory listings, control panels, etc.). Its vaguely unsettling (and vaguely slow).
> WRT personal software going web: imagine you are on deadline, sitting > at your (budget friendly) web terminal, just finishing your job > critical report using the MSWebWord service that your company leases > (at an irresistable price) from some local ISV. Now imagine a moth > flies into the server ...
I think you're not making the point you think. You are describing a "single point of failure" problem - very risky. But, which scenario really fits that description? Using files on a desktop machine or using a web machine as a front to files that are "on the net"? We can't really know, but I'd be inclined to say that the second scenario at least sounds like it has less of a chance of suffereing the single point of failure problem. If nothing else you hopefully have a local copy and a copy on a server someplace. If my web service company is professional, there may be several copies all around the world.
Anyway, now we are even further off topic.
For good or bad, many applications are becoming web applications and most new applications written are web applications. As this trend occurs people are moving away from some established languages like C/C++ and using either languages that are supposedly better positioned for web app work (Java/C#) or to more flexible languages (Python/Lisp) or to certain domain specific ones (Perl/Ruby).
Chris Perkins Media Lab, Inc.
[ More interesting (to me) is that as this happens the now dominant paradigm of a split between document vs. application is slowly beginning to wane. And, my crystal ball says that files as they exist now (pathnames, extensions, custom data streamed to disk) are the next domino to fall, though it'll take a long long time and they'll never go away 100% - just marginalized away from user experience. But this is not the place to discuss this. I'll go post a write up on comp.predictions.stupid and discuss it with the denizens there. -CP]
Espen Vestre <espen@*do-not-spam-me*.vestre.net> writes: > t...@famine.OCF.Berkeley.EDU (Thomas F. Burdick) writes:
> > PS You couldn't pay me enough money to reimplement TeX in Pascal.
> TeX _is_ written in Pascal, so I wouldn't pay you a dime for doing > that ;-)
Well, web, iirc, which is a format that can be run through a preprocessor to either output pascal, or a book. (I tried cweb for a while, a variant that works with C, but found that I am not particularly talented at writing a meaningful book while writing a program...) -- I'm awfully glad I'm a Beta, because I don't work so hard.
> This trend is in part why so many new languages are being developed > and old ones (like Lisp) are being unearthed. You can write great web > server stuff with Lisp.
You've got a good point there. C++ is not a great language for network programming. The low-level memory model makes it irritating to use in a networked environment, where you want to be able to ship objects over the link without fussing about memory layout and where the vtables are and whatnot. Note the gymnastics stuff like CORBA and COM have to go through to present a distributed programming model to C++ code. In theory a language that abstracts the underlying memory model could present some *very* clean distributed programming interfaces. I don't know if any Lisps let you just transparently ship objects over a link, but it would certainly be doable.
