PDA

View Full Version : Promoting Free Pascal in Schools



WILL
02-08-2005, 07:08 AM
Hmm... I wonder if there would be some validity to trying to rally support from schools and universities to adopt Pascal via FPC into their curriculum.

I mean... Pascal used to be the defaco learning language, why not now? C sucks for trying to learn concepts and you pay a high price for even minor beginner mistakes. I remember my first C program froze up and rebooted the computer when I ran it. It had only a simple code block... no code! :lol:

I'm sure there can be a strong case put together for the schools to pick it up. Many universities use GCC in their programs so why not another opensource compiler?

And Pascal is just a better language to learn programming concepts in. Object Pascal is a great way to learn about OOP too! Rather than your messy C/C++ conventions Borland's rendition of Object management is great. Why else would the industry still use it?

Borland teaches courses for Pascal for use with their Delphi so why can't Free Pascal have the same. It just needs support by some key universities and it's sure to catch on as an alternative to teaching Java or *gak* VB. :p

Any takers on this idea? FPK? Synopsis? Lightning?

Traveler
02-08-2005, 07:36 AM
When I went to school I got a 4 week Pascal course (as a introduction for the C course). I heard later that it was the last year they teached pascal, after that they replaced Pascal and C with Java. That was over 6 years ago, but I think its still the case.

Bart
02-08-2005, 07:48 AM
The main problem is that the companies don't use pascal. Delphi is an exception and it is regulary used in companies but I don't think Lazarus is an alternative as it is still under heavy development and it is not the best choise for database programming. Companies don't mind to spend money if they get quality for it.

My opinion is that the GTK 1 interface is too ugly if you want to develop a commercial program. Lazarus should support the Windows gui and qt support would be very nice for both Open Source and commercial programmers.

MikeS
02-08-2005, 02:34 PM
What I just don't understand, is why (game)companies in general don't use object pascal. I mean, we all know about how C++ came off of C when it was popular, and they had a huge userbase there, but why not use Object Pascal.

Object Pascal is...
-Clean
-OOP
-Faster to develop in
-Easier to develop in
-Powerful

Those five elements are the same you see in C, with the exception of it being clean(I'd really just say Object Pascal is bit cleaner than C).

Anyway...Object Pascal is a perfect candidate for learning. I kind of wish I'd have skipped BASIC(Not that basic is all that bad), and jumped into Pascal a few years ago.

8)

Bart
02-08-2005, 05:33 PM
Pascal has the reputation of being an ancient language. C++ is the most popular programming language these days and if you want to write games for a console, I don't think there are compilers available for other languages. Or am I wrong?

WILL
02-08-2005, 10:18 PM
Well those based on ARM processors are close to getting a Free Pascal compiler. ;) Well FPC 4 GBA pending... I beleive there is also a Visual Basic solution aswell. Then again... there is always the assemblers themselves, but I doubt anyone would make the next greatest console hit in pure assembly.


I've found out some more information on licencing for consoles <u>here (http://www.gamedev.net/community/forums/showfaq.asp?forum_id=5#console)</u>.

And <u>here (http://warioworld.com/apply/agb.html)</u> is the one that FPC 4 GBA promoters will take special interest in. I hope that future Pascal developers will be able to challange and get one of these bad boys some day. ;)


I think the key to Pascal's return to fame is the schools! It still has a reputation fobeing a teaching language so thats an in. And form there people will learn more about Pascal or namely the dilect of Object Pascal and learn of it's power, cleanliness, availability, stability and hopefully in the near future improved speed of it's compilers compiled & assembled code.

Raskolnikov
25-08-2005, 11:12 AM
I think Borland Pascal is taught at the local polytechnic before moving onto Delphi - i also think they teach game programming as part of the Bachelors degree..*i think*.


What I just don't understand, is why (game)companies in general don't use object pascal. I mean, we all know about how C++ came off of C when it was popular, and they had a huge userbase there, but why not use Object Pascal.

I dunno why C/C++ has been favoured by dev houses..maybe it's because the API's are released in those languages and no one wants to spend the time ( and no doubt money ) in using a 3rd party / in-house port..there's the issue of documentation, troubleshooting etc and ofcourse dont forget there are far more C/C++ game developers around than (Object) Pascal. I think Bioware ( Baldurs Gate, Neverwinter Nights, MDK2 etc ) uses Delphi to some extent for various tools and what not.

I dunno about you guys, but writing something like Quake 3 and having to type BEGIN all the time would drive me nuts.

Sly
25-08-2005, 12:24 PM
As a professional game developer myself, I can answer with some authority on this. C++ is the standard in games development because every commercial and the vast majority of free libraries out there for games are written with C or C++ in mind. A lot of game development grew out of the hacker scene, and in that scene they mostly used C because it was a high level language that still allowed the programmer easy access to the low level metal of the PC.

A lot of the stuff we do in game coding involves pointer manipulation, and as a long time Pascal and C programmer, I have to say that pointers are a whole lot easier to use in C.

C++ offers templates. Pascal does not have anything to rival templates yet. C++ offers full support of operator overloading, which only just now FPC supports in a limited manner and Delphi is gaining even more limited support for operator overloading.

Object Pascal may support classes, but the fact that every class inherits from TObject is one major drawback due to two things: every class instance incurs the cost of a pointer to the VMT because at least one method in TObject is declared virtual, and every class must be allocated individually. In C++, a class instance can be declared locally on the stack. Can't do that in Object Pascal. In C++, a class only incurs the cost of a VMT pointer if a method is declared virtual. Therefore
class TSimple
&#123;
public&#58;
int member;
&#125;;
only costs four bytes. In Object Pascal

type
TSimple = class
public
member&#58; Integer;
end;
is eight bytes in size. Plus you must allocate each class instance by calling Create, which calls several methods of TObject internally. Creating an instance of the same class for use within a function in C++ takes nothing more than just declaring a variable of type TSimple, which takes no more CPU time than it takes to decrement the stack pointer by four bytes.

Say you want a function to create a given sized array of TSimple instances and initialise each array element to 1. The code for this in C would be

TSimple *AllocArray&#40;int size&#41;
&#123;
TSimple *pArray = &#40;TSimple *&#41;malloc&#40;size * sizeof&#40;TSimple&#41;&#41;;
memset&#40;pArray, 1, size * sizeof&#40;TSimple&#41;&#41;;
&#125;
One memory allocation. One call to memset. Simple. Quick. Blindingly fast. Now for the equivalent in Object Pascal.

type
PSimple = ^TSimple;
PSimpleArray&#58; array &#91;0..High&#40;Integer&#41; div SizeOf&#40;TSimple&#41;&#93; of PSimple;

function AllocArray&#40;size&#58; Integer&#41;&#58; PSimpleArray;
var
pArray&#58; PSimpleArray;
i&#58; Integer;
begin
GetMem&#40;pArray, size * SizeOf&#40;PSimple&#41;&#41;;
for i &#58;= 0 to size - 1 do
begin
pArray&#91;i&#93; &#58;= TSimple.Create;
pArray&#91;i&#93;.member &#58;= 1;
end;
end;

That was a lot more complex. We first have to declare a new type that is a pointer to the type we want. Then we have to declare an array of the largest possible size of that pointer type to handle the unknown array size. Allocate an array of pointers to the class instances, then allocate and initialise each class instance individually. Each call to Create calls other methods of TObject, which all takes time per instance. TObject even uses dynamic strings, which on a console system are a big no-no. You allocate once and never reallocate. Unfortunately, dynamic strings reallocate memory almost every time you so much as sneeze at them.

I'll leave the cleaning up of the array to the reader. Hint: The C version is one statement.

Another benefit for C++ is the huge support of tools such as development tools and debuggers. The debugger in Delphi, even in the latest version, is woeful and horribly inadequate when compared to the debugging tools we have access to with C++.

This post may seem like it is bagging Pascal as a game development language. The intent was to point out some reasons why C and C++ is the preferred language in game development, and some of the reasons why. Pascal is a great language. Every language has its strong points and its bad points. In the world of professional game development, C and C++ have more benefits and a lot more support than Object Pascal could provide.

P.S. Bioware used to use C++ Builder for their tools. After Neverwinter Nights, they are looking for alternatives (still using C++). Quotes from http://www.gamasutra.com/gdc2003/features/20030306/brockington_03.htm

Last, but not least, we had problems with the toolset component of the project. The Aurora Toolset was built in Borland C++ Builder, based on the fact that we heard all sorts of rumors of a Linux and Macintosh version of Builder. However, this turned into a monstrous headache for development. C++ Builder would throw floating-point exceptions during the middle of the graphics engine, which would not show in the middle of the same library run through Visual C. We couldn't link the graphics library with Builder without going to a DLL, so we had to make the graphics library source tree work on C++ Builder as well. The Macintosh version of Builder vanished into the land of vaporware, and we were left holding the bag.
So, we are trying to duplicate the rules code in both C++ Builder and Visual C for the most important components. We are investigating alternatives to C++ Builder for future projects.
Even trying to maintain C++ code that compiles on both Visual C++ and C++ Builder is a major PITA. Believe me, I've been there and done that. Almost like trying to maintain code that compiles in Delphi and Free Pascal. That is a growing rift right there.

Hint: If they had masked floating point exceptions, they would not have had those errors. This is a pity though that the Visual C++ runtime masks floating point exceptions by default and Borland's runtime does not. Direct3D and OpenGL throw floating point exceptions all over the place.

Almindor
26-08-2005, 06:32 PM
Good points albeight a bit out of topic, question was pascal in schools.

IMHO pascal will not return into the mainsteam but it may survive with a smaller userbase if FPC and most importantly Lazarus teams can get it right. (the former is basicly "there" the latter has a lot to do IMHO).

Altho putting pascal into schools again might help it, it might also hurt it as it already did once. Most people don't remember the old school pascal days in a good way. Almost everyone switched to C or somesuch after school or at home because what's from school is not good.

WILL
27-08-2005, 05:47 PM
Hmm... well I think the 'whats at school is not good' syndrome came from the fact that alot of schools or school institutions all typically run off of budgets, many of them in North America(Canada and US specifically) lately being reduced quite signifigantly. And as such, expensive compilers like Delphi at X-hundred a pop are out of the question. Turbo Pascal was alright because the pricing was a bit mor reasonable, but as things go, Pascal got less affordable.

With the introduction of Free Pascal, it is possible that this situation with the schools can be averted. Now with GCC already having a rather strong hold in the Open Source 'market' people might argue that C is the better way to go, but if we all recall, Pascal was made a 'learning language'. Now it's come a long way in the many years, but if you sell it as a learning language to the schools with the key notes that it is;

A) Better at teaching PROPER object oriented programming;

B) Easier to read and understand; and

C) a thousand times more friendlier as a language becasue of it's excellent debugging and error catching capabilities.

These were the key points that teachers would talk about with utter glee in their voice when I asked about the C/Pascal question. And I've learned to believe it even today.

C may be the way that the market goes today, but markets change! And like it did before, I believe it is highly possible that it can again. But it starts with the next generation and what good things that they are taught in school today.

Ingemar
25-05-2006, 06:25 PM
I think Pascal is the best programming language that I have ever used, but it is severely backtalked by C lobbyists. They have been lying about it for years. In reality, it has only one severe drawback: portability. That's because Borland refused to make their dialect conform with standards. But today we have FPC, cross-platform and modern! :D

For a long time, using Pascal was a fantastic competitive advantage for me, it made me so much faster than my C-using competitors. In recent years, the advantage is smaller since Pascal support has been dropped just about everywhere, my old professional tools are discontinued, but I still hope to get back on track with Free Pascal.

Switching to C/C++ doesn't seem like a good move to me. Then I would have no advantage against the rest of the crowd, nothing that can make me better except hard work. Pascal makes my code more readable, easier to maintain, faster to write, and executes on par with C. If it is a small language, no problem. Its qualities can be our little secret. :wink:

Concerning teaching, today's habit of using C/Java is plain madness. We should not iron in a single solution in the students, we should widen their scope with alternatives. pascal is a great language for teaching as for anything else, but let them start with C and try Pascal later. They just might get the point better in that order. :roll:

fragle
26-05-2006, 09:34 PM
Good thing i don't have to promote anything - they've been using pascal for teaching for more than 10 years now, yay! (probably even more, cause i don't know when it was started actually ;) ) First we were taught with Turbo Pascal (i fall into that category), but around 2002-2003 they've switched to freepascal (i mean country-wide, not just my school). There is even some sort of IDE (http://ims.mii.lt/fps/en/about/index.html) developed to avoid using the default console editor :shock:
In universities they're kind of undecided though.. When i first came they've just switched from teaching pascal to c++/java hybrid, but apart from those classes, one can write his projects in almost anything he wants ;)

Ingemar
27-05-2006, 06:42 AM
When i first came they've just switched from teaching pascal to c++/java hybrid, but apart from those classes, one can write his projects in almost anything he wants ;)
I teach computer graphics, and I have, officially, the same principle for project work, but the students hardly dares to believe it. They even ask whether they can use C++ classes (which the C labs don't). When a student humbly asks if he/she can use Delphi, despite the fact that the labs are Java and C, I more than approve. I try to give them a hint about my opinion - use what makes them productive, and why shouldn't they use the best if they are among those who understand it?

But the computer support is a problem. I had a student this winter who had only used Turbo Pascal. Fine, I said, then he should be happy with FPC, right? Sure... if the support hadn't refused to install it. :x

WILL
27-05-2006, 06:54 AM
Ah... Pascal advocates after my own heart. :)

Well it's great to hear that in some parts of the world things are almost done right.

Great info on FPS, fragle! I'll have to post this one in the news this weekend.

Ingemar: What do you mean by 'the support had refused to install it'? Are we talking a person or the software?

Ingemar
27-05-2006, 07:45 AM
Ingemar: What do you mean by 'the support had refused to install it'? Are we talking a person or the software?
A person, the sysadm. I don't have rights to install anything, I must ask the computer support people, and he said that I had to be "very convincing" to make him do it. Of couse that means that he didn't want to. The same install took one minute on my own Linux computer. :roll:

I suppose his refusal is due to the risk of messing up a RedHatEL installation by installing new software (and I have done that myself some time), but personally I think it is his job to handle. If RedHatEL isn't good enough, he should find a better distribution. (Any suggestions?)

WILL
27-05-2006, 07:55 AM
(Any suggestions?)
Give the guy a swift kick in the arse and tell him not to be lazy? :lol: Oh wait, maybe not what you meant huh? ;)

IMHO RedHat sucks the big one! :P I'm serious though. Feldora Core is the result of the RedHat company --which Linus worked for for a time, btw-- trying to be all commercial like Microsoft. Linux and Microsoft are not good combinations. Hence the mess that is RedHat. Sure it's stable, great... but it's like some half breed thing that it's quite with the GNU standard anymore.

I recommend Debian, but thats me. It's like Slackware, but with a really dependable(the best out of all of them!) packaging system and it conforms to ALL the Linux standards. Plus it does a rather good job of keeping up to date with all the major things like Apache, PHP, mysql, gimp, kde, etc, etc...


If he *needs* a reason to install it, tell him it's a part of your course material and it is *required* for you to use in your classes. If you can pull that off with the head of your department, then I think you're set.

dmantione
27-05-2006, 07:56 AM
A person, the sysadm. I don't have rights to install anything, I must ask the computer support people, and he said that I had to be "very convincing" to make him do it. Of couse that means that he didn't want to. The same install took one minute on my own Linux computer. :roll:

I suppose his refusal is due to the risk of messing up a RedHatEL installation by installing new software (and I have done that myself some time), but personally I think it is his job to handle. If RedHatEL isn't good enough, he should find a better distribution. (Any suggestions?)

Yes, he can use our rpm without risk of messing up his distribution, since removing the rpm will remove all traces of fpc. It is a good opportunity to show one of the weapon facts of fpc: fpc programs do not depend on any libraries and are Linux distribution independent.

Therefore our rpm works on any distribution and he does not need a specific Red Hat rpm. Such a feature should get make any Linux sysadmin curious :)

Ingemar
27-05-2006, 10:49 AM
(Any suggestions?)
Give the guy a swift kick in the arse and tell him not to be lazy? :lol: Oh wait, maybe not what you meant huh? ;)

No, but it sure feels like the right thing to do. :roll:


IMHO RedHat sucks the big one! :P I'm serious though. Feldora Core is the result of the RedHat company --which Linus worked for for a time, btw-- trying to be all commercial like Microsoft. Linux and Microsoft are not good combinations. Hence the mess that is RedHat. Sure it's stable, great... but it's like some half breed thing that it's quite with the GNU standard anymore.

I kind of agree. I don't feel I have the best dist. Whenever I need to install anything that is not in the official dist, I get version problems, have to dig up rpm's elsewhere, cross my fingers and install. No, it doesn't always work.


I recommend Debian, but thats me. It's like Slackware, but with a really dependable(the best out of all of them!) packaging system and it conforms to ALL the Linux standards. Plus it does a rather good job of keeping up to date with all the major things like Apache, PHP, mysql, gimp, kde, etc, etc...

I had two installations of Debian dying on me when running updates. X died when I updated the NVidia drivers. Bad luck? I don't know.

BTW, I think we are off-topic now. Sorry about that.

WILL
27-05-2006, 11:18 AM
Well golden rule #1 of Linux... it it ain't broke, don't go trying to fix it. :P

Hardware drivers for Linux in general can be a tricky thing. But Debian, as far as I know has gotten a lot better at this. Plus they are often up to date with everything. AND they have a Lazarus package that I believe is up to date! :) (Laz ppl correct me if I'm wrong on this?)

At the very least installing Lazarus on any distrobution is not a problem anymore. Or so I'm told, I've not been in Linus for some time now.

At any rate, trying to get us back on topic (my fault really, so I forgive you! :D) getting Laz and FPC installed should be the least of any sysadmin's worries. So tell Mr. 'I don't wanna'-pants to get it in gear. :P

dmantione
27-05-2006, 03:03 PM
But Debian, as far as I know has gotten a lot better at this. Plus they are often up to date with everything. AND they have a Lazarus package that I believe is up to date! :) (Laz ppl correct me if I'm wrong on this?)


Debian and up to date?! They are still shipping fpc 1.9.4!

michalis
27-05-2006, 04:24 PM
But Debian, as far as I know has gotten a lot better at this. Plus they are often up to date with everything. AND they have a Lazarus package that I believe is up to date! :) (Laz ppl correct me if I'm wrong on this?)


Debian and up to date?! They are still shipping fpc 1.9.4!

(sorry for going off-topic again...)

It's not so bad --- It's 2.0.0 in etch (testing distribution). 1.9.4 is in sarge (stable). It's normal that stable distribution doesn't have most up-to-date software as it's frozen.

But actually it is bad --- why 2.0.0, not 2.0.2 ? I don't know, most probably the maintainer doesn't have enough time... I guess that someone who is familiar with creating Debian packages and has some time should just go and offer help, see http://packages.qa.debian.org/f/fpc.html.

WILL
27-05-2006, 08:08 PM
Sarge is actually old now. There has been 2 versions since. Stable is one and the next one which I forget the name to... anyone know?

Well yeah, I think that there really needs to be someone that promotes FPC and Lazarus on Debian (and all the other distros), as it's somewhat a part of the process of keeping Pascal in the eyes of Linux people.

I mean, look how much I bugged Florian about GameBoy Advance and now it's gonna be a part of the next version of FPC. (not quite official yet though)... oh and of course I probable wouldn't have gotten quite as far as we did wiothout the help of Mr. Legolas. ;) He really took that project with both hands!

Anyways, point of note; I think w should all actively push FPC/Lazarus in all these places. Schools, Linux distros, anywhere we can really...

Robert Kosek
27-05-2006, 08:14 PM
Semi-aside from the point, I think that we as a community and as pascal programmers need to work on the things like Sly mentioned that make folks think that Pascal is outdated. I don't think it is, and I love using it, so let's cure the perceived issue and improve our favored language.

I also think we need to push for Pascal in schools. It is an excellent language and I have learned much just doing it as a hobby. There is so much you can learn on your own with it, but I think that the perception must be changed for Pascal to be "reaccepted".

dmantione
27-05-2006, 08:22 PM
Well, somthing that makes a language appeal to young people is that you can make games in it :D How many schools are teaching people programming with the CRT unit? :think: That helps of course to the dated image Pascal has.

Robert Kosek
27-05-2006, 08:29 PM
Yes, I think the Pascal using curriculae need an update desperately. I think it's time to go VCL and DirectX/OpenGL. :D But then again, most folks have lost caring for the language somewhere.

WILL
27-05-2006, 09:28 PM
What if there was a group with a petition and an active goal of pushing Pascal to all these universities all over the world?

It seems to be the way to get something done. You form an organized group of people that believe something, get it in print, start talking to professionals and the institutions that will make it happen and push for results.

The new Development Company to branch from Borland would definately support it so they are an asset (you would need a bunch of these interested groups) amoung other companies and commercial developers that could back you. It would then be taken quite seriously as it has the support organizations to keep things rolling.


This group's goals would probably be:

:arrow: Establish a presence of the Pascal and Object Pascal languages in secondary and post secondary schools and institutions.

:arrow: Promote Object Pascal in the commercial markets.

:arrow: Establish Object Pascal as a set language standard.

:arrow: Cultivate interest in developer products that use Pascal and Object Pascal and gain these companies' support in the promotion of Object Pascal.


I think that is not too unreasonable. :) If such an organization exists, with a modivasted leadership, we'd see more possitive amount of interest in Object Pascal.

Talk to the big companies that push Object Pascal. (Borland, RemObjects, WinSoft, Indy, etc) There is a big list! They will want to help and strengthen the point that Object Pascal is very juicy as a commercial development language.

michalis
27-05-2006, 10:31 PM
Sarge is actually old now. There has been 2 versions since. Stable is one and the next one which I forget the name to... anyone know?


Just to clarify things:

Sarge (3.1) is the last stable release. There were 2 updates to sarge (3.1r1 and 3.1r2) but these are intentionally limited only to fix security and other serious bugs. Next release is codenamed etch --- currently that's the testing distribution, and it will become next stable release.

Ingemar
28-05-2006, 07:21 AM
Yes, I think the Pascal using curriculae need an update desperately. I think it's time to go VCL and DirectX/OpenGL. :D But then again, most folks have lost caring for the language somewhere.
I did the Pascal interfaces for OpenGL on the Mac myself. It works like charm. I move between Pascal and C a lot, and can easily feel the lower stress level when using OpenGL with Pascal.

C lobbyists keep flaming Pascal for being useless, slow and outdated, and they have been wrong all the time. They often don't even know that you can have more than one source file! :shock: And they don't know about Object Pascal, despite the fact that it predates C++ both as language and as a commericial application building tool. Or maybe they know but don't want to tell?

With the recent new features in implementations like FPC, I think Pascal is quite modern. I am currently spending some of my time learning FPC, and I can't wait until I can do my first FPC game. :wink:

AthenaOfDelphi
28-05-2006, 10:08 AM
The biggest stepping stone to encouraging more interest in Pascal at any level is industry.

Whilst industry sees Pascal as outdated, it will continue to insist that educational institutions teach C++, Java, C# etc.

Unfortunately, industry is in love with the 'next big thing' because its new, innovative... it must be better... right??? And of course, we musn't forget the millions of lines of code that already exist... industry will want maintainers for this, so again, they will want C++ and Java.

From my personal experiences, I've come to the conclusion that much of this is down to middle to upper managers, who understand only the latest buzz words. In other words, they don't actually have a clue about what may or may not actually be the best language. One project I worked on, the company had around 15 Delphi developers. We worked our butts off to build this system. It won them some big contracts and was in its own right a fantastic system... at that point, middle management stepped in and decided that the whole thing needed to be re-written in C++ (at the time, C++ was one of the buzz words)... as a consequence, nearly the whole of the development team were forced out or gotten rid of. That is the kind of stupid business decision that drives languages forward, and whilst that sort of thing is happening, an older language, like Pascal, stands little chance of making head way in industry.

Of course as has been pointed out, there are schools that are teaching with older versions of Pascal... now my personal take on this is that the world of windows programming can be more than a little daunting, even for an experienced developer. So I think teaching them the basics using older versions which don't have that overhead isn't such a bad thing. It wouldn't be very good to scare them off. There is also the point that only a small percentage of software engineers are games developers and so it is better to teach the kids stuff that is likely to be useful in a wide range of fields.

I don't actually think the problem is the choice of teaching tool, I think the problem is the kids themselves. They are bombarded with high tech and fancy graphics everywhere and so can't appreciate the simple wonder of writing simple programs, couple this with most kids apparent inability to sit still and concentrate for any longer than about 30 seconds and whatever is used isn't going to garner interest in it.

Personally, I think the key to encouraging Pascal adoption is to snag the people (especially kids) who like to think for themselves... hook them and prove to them that Pascal is a modern language that can compete with C++, Java, C# etc. and it may just start making some headway in industry... when it does that, it will start making headway elsewhere because industrys demand for Pascal programmers will rise.

Just my thoughts :-)

Ingemar
28-05-2006, 11:17 AM
One project I worked on, the company had around 15 Delphi developers. We worked our butts off to build this system. It won them some big contracts and was in its own right a fantastic system... at that point, middle management stepped in and decided that the whole thing needed to be re-written in C++ (at the time, C++ was one of the buzz words)... as a consequence, nearly the whole of the development team were forced out or gotten rid of. That is the kind of stupid business decision that drives languages forward, and whilst that sort of thing is happening, an older language, like Pascal, stands little chance of making head way in industry.
That's a horrible story! Really scary - and the worst part is that I do believe you! These things happen, big money is thrown away just because someone told a manager that they should follow a stupid buzz word.

Today, many young programmers go for languages like Python, since it clearly is better than C++, but then they, and their users, suffer from the pains of its poor performance. So there are people who are looking for alternatives.

dmantione
28-05-2006, 01:10 PM
While it is abviously true that educational institutions educate what the market demands, they are also have a certain resistance against the current programming hypes. I.e. the market once wanted them to educate Visual Basic developers, however most universities resisted the pressure.

For the same reason many schools and universities consider C/C++ an unacceptable language for beginners. A lot of them have embraced Java, which indeed solves some deficiencies of C, unfortunately it introduces others.

Without question, many universities start in C/C++, and not withotu reason, as in teh current world, most software is going to be written in it, so there is something to say to not to withold students this knowledge.

On the other hand, there are still a lot of arguments in favour of Pascal nowadays, and it never hurts to communicate that.

I think there is a network, schools have a influence what companies use, and companies have an influence what shools use. There is a third important group, and that is to which we belong: Amateurs, and this is often written with a big A to emphasize the role they played into the direction the computer industry took.

To make Pascal attractive for the enterprise, one way is to have a good amateur base: I.e. compare it to Linux or PHP, it got into the enterprise because amateurs used it, and slowly got developed to the level it was more attractive for enterprises than other solutions. In fact, it can also be said that Delphi became less attractive to amateurs over the years, and as a result it became harder for companies to recruit good Delphi programmers.

WILL
28-05-2006, 09:52 PM
So I think teaching them the basics using older versions which don't have that overhead isn't such a bad thing.

Well why not newer versions with little overhead? :) Whats everyone got against variation these days! :lol: Seriously though, I can see the demand on these developer tools Lazarus, Delphi and FPC becoming greater as the interest increases. And with that the need for different 'editions' this varying needs and skillsets. Borland used to get that, but sort of dropped the ball and has become some ALM monster. The FPC team, doesn't have all the manpower I think it needs to do that quite yet. And I couldn't begin to tell you about the Lazarus people, they haven't invited my into their 'click' yet. :P Yet... :twisted: (Hey, I'm damn nosy when I want something, ok?! :lol:)

But no response to my little group idea? Noone is interested in something like that? It would sort of help to map out all the issues and factors relating to the success of Pascal (and Object Pascal) in the marketplace and popularity over the net and in schools.

I think a Pascal Advocates network would help grow those amatures that Dani?īl is talking about. I often look at PHP and MySQL's success and think to myself, they got hat popular off of their own hard work... it's gotta be possible to do the same with something that is already as estabilished as Pascal. We just need to make enough noise.

Almindor
12-06-2006, 11:08 AM
How about just stop talking and start coding? :D

WILL
12-06-2006, 06:07 PM
Umm... I think thats how we got into this PR trouble in the first place. ;)