Hello,

An open source pascal operating system could be cool.

Would translating/porting linux to pascal be possible ?

Bye,
Skybuck.

As with many things, it only depends of the time you have. But as you grow older you get a different idea of life expectancy and the use of your remaing time.

Regards,

Gerard.

What is the use anyhow?

2008/12/4 Skybuck Flying

> An open source pascal operating system could be cool.

There are already such projects (including mine :-)). Search google code and sourceforge.

> Would translating/porting linux to pascal be possible ?

Yes, but who would? And if he/she would, does he/she have the time?

Would translating/porting linux to pascal be possible ?

Possible? Yes it is. Worth of...? I'm afraid not.

By the way, Linux is good as it is now and Pascal isn't the best option to create an operating system.

Guillermo "Ñuño" Martínez

Guillermo Martínez Jiménez wrote

By the way, Linux is good as it is now and Pascal isn't the best option to create an operating system.

Are you sure? doesn't older MacOS's versions where written in Object Pascal?

I think the problem here (again) is not the language, it's the critical mass of users of the language. Using C for Linux was a good bet, not because the language is good (Pascal is way better for me), but because C has a wider user base who can fix/add features.

Leonardo M. Ramé
http://leonardorame.blogspot.com

Yes, and there's nothing wrong with Pascal for an OS. It would be excellent.

I think the problem here (again) is not the language, it's the critical mass of users of the language. Using C for Linux was a good bet, not because the language is good (Pascal is way better for me), but because C has a wider user base who can fix/add features.

Using C for Linux was the only way, because there was a free C compiler (GCC) and none for Pascal or other comparable languages. Things have changed since then, but much of the software industry is on a path decided from the situation 20 years ago. The industry took the C route since there was no cross-platform Pascal, while Linux made the choice from available free software. Now everything must have inherited design flaws from C just because of what was available in the 80's. That is, unless the tide changes. It can happen.

Now, that sounds like one real problem, related to the porting issue, which we can address. And not that doesn't need as much work.

So the KDE and Gnome projects are language-locked? (How? Surely users don't select software by language? Or are the distros incomplete, FPC-wise? Hard to recompile?) Anyway. that sounds bad! And that suggests a much shorter path: Adapt a distribution to "Pascal Gnome" support, with full Pascal interfaces and code, or whatever is missing. Port Gnome examples, if it isn't already done, and show that Pascal (FPC) is not only an option but a *better* solution. Easier than C/C++, faster than Java, much faster than Python... I don't know if an FPC-tuned distro would make any difference, stronger FPC support in, say, Ubuntu, would make a bigger impact. Acceptance of FPC sounds like a good goal to me.

Who isn't distributing? Gnome? That isn't a distribution. I don't quite understand what you mean.

The "FPC OS" doesn't have to be 100% written in FPC, not even 10%. No hurry. The C code is ugly but it works as long as you don't take GCC out. Port what really counts, making it possible and well supported to write *new* programs in FPC.

/Ingemar

Ingemar Ragnemalm wrote

So the KDE and Gnome projects are language-locked?

Yes, they are.

(How? Surely users don't select software by language?

Users don't care, but they usually use what comes with their distro and window manager, so if KDE comes with a bunch of software they will use it instead of looking for alternatives.

Or are the distros incomplete, FPC-wise?

distributions are not the same thing as window manager.

Distributions are much more flexible language-wise, but they tend to package whatever comes with the window manager.

It would also be possible to build a distribution instead of a window manager, but I'm usually into writing code, as opposed to setting things up, so building a distribution doesn't seam a lot of fun to me =)

Ingemar Ragnemalm wrote

Now everything must have inherited design flaws from C just because of what was available in the 80's.

This is one strong argument against porting Lunix to Pascal, isn't it? Porting means sort of translating old design flaws to another language.

That is, unless the tide changes. It can happen.

It has happened in the past. There has been OS/2 for example, with it's superior OO design. RIP.

There have been many OSses with new appealing features, BEOS coming to my mind.

And there have been some people starting to write an OS in pascal, but I've never heard of something coming out as ... say at least an alpha version or some sort of prototype.

SCNR,
Marc

Are you sure?

Yes, I am.

doesn't older MacOS's versions where written in Object Pascal?

Yes, it does, but as I said I think it wasn't the better options.

I think the problem here (again) is not the language, it's the critical mass of users of the language. Using C for Linux was a good bet, not because the language is good (Pascal is way better for me), but because C has a wider user base who can fix/add features.

I disagree. C is better for write operating systems *by definition*: C was created to write UNIX, Pascal was created to learn good programming techniques. C is low/mid-level language, Pascal is high-level (and Object Pascal is even higher): OS are the lowest software level.

I'm not saying it's impossible: here you have MacOS and Toro. I'm just saying that _I think_ it isn't the best option. Of course a better option is to write the kernel in C and Assembler and the utilities in Pascal and Object Pascal.

Guillermo "Ñuño" Martínez

Hello

I don't mean to start a culture war here, but I think there is a bit of overstatement in this post that 'C' is somehow a better language for writing an OS in than say Pascal or OP or anything else.

In point of fact, the only 'best' language for any processor is its own machine code which is always binary.
And the first operating system software written was done so in the assembly language which is written to use mnemoics
that when 'assembled' convert directly into the binary machine language on a one to one basis.

Now EVERY compiled language, including C, must be processed from its syntactical representation into assembly language or machine code. In fact, if you download the GCC complier source code and read it, you'll see almost immediately that it actually a 'language' front-end connected to a 'assembler machine code' back end.

When the 'C' language was designed at AT&T it used 'assembler' like constructs and syntax because it's author wanted
to stop writing programs in a language called BCPL..

So aside from a 'historical accident' that the first AT&T OS ..UNIX.. was being written at the same time C was being developed, there is no other reason for any program, including an OS, to be written in C. To suggest that C is better is rather like suggesting that Spanish is a better language than English for writing a novel.

That said, if someone wished to write a modern OS in FP, which has nearly every programming construct that 'C' has (ints, doubles, floats, bytes, and bits) and you were willing to put in the time to fine tune and modify the FP compiler, you could produce a OS in FP that would be every bit as effective and perform as well or better as one written in C. And as far as the processor running the OS couldn't tell what language the OS was written in.

If anyone wants to convince themselves on this its simple, you'll find Ritchie (the creator of the C language) explained these points in the preface to his first book on C. There is also a brief mention of this in the Wikipedia article on the UNIX operating system. (see the article footnotes)

Prince

On Fri, Dec 5, 2008 at 6:59 PM, Guillermo Martínez Jiménez <

Here is a direct reference from the Wikipedia article I referenced in my prior post ...

In 1973, Unix was rewritten in the C programming
language<http://en.wikipedia.org/wiki/C_%28programming_language%29>, contrary to the general notion at the time "that something as complex as an operating system, which must deal with time-critical events, had to be written exclusively in assembly
language".[4]<http://en.wikipedia.org/wiki/UNIX#cite_note-Stallings-3>The migration from assembly
language <http://en.wikipedia.org/wiki/Assembly_Language> to the higher-level language <http://en.wikipedia.org/wiki/High-level_programming_language> C resulted in much more
portable<http://en.wikipedia.org/wiki/Software_quality#Portability>software, requiring only a relatively small amount of machine-dependent code to be replaced when porting Unix to other computing platforms<http://en.wikipedia.org/wiki/Platform_%28computing%29> .

Prince

Link to the article about the AT&T UNIX OS and C ....

http://en.wikipedia.org/wiki/UNIX

Dear All ,

In relation to programming language to be used to program an operating sytem , the Burroughs Corporation is a very good example .

( I am NOT saying that porting Linux to Free Pascal is a good idea . There is MINIX3 , porting Free Pascal to MINIX3 could be a very good job .
For 'why' , please see ( http://www.minix3.org/ )
" ... Single-chip, small-RAM, low-power, $100 laptops for Third-World children ..."
)

I worked on the Burroughs systems
( B3500 since 1974 ... later , B4700 , B6700 ) .

They designed the COBOL compiler , then the B2000 (?) but I am sure that B3500 was in that form :

Burroughs mid-sized computers were at least 10 times faster than equivalent other main-frames on data processing jobs because COBOL statements were translated directly to machine code , i.e. , its machine codes were NOT like their contemporary main-frames .

(
http://en.wikipedia.org/wiki/Burroughs_B2000
" ... The architecture was built to support COBOL programming in the most efficient way possible ..."
)

Later , they designed their Algol-like language ESPOL , and then B5??? series . In the B6700 main-frame the machine language was the ESPOL , i.e. , the computer was executing ESPOL directly . During development of this series , design team did not say to upper management that "... are designing a computer that it will NOT have machine language ( assembler ) but a 'new machine' ." because acceptance of a design not having a machine language was very unlikely .

(
http://en.wikipedia.org/wiki/Burroughs_Corporation http://en.wikipedia.org/wiki/Burroughs_large_systems http://en.wikipedia.org/wiki/MCP_(Burroughs_Large_Systems) http://en.wikipedia.org/wiki/ESPOL
http://bitsavers.org/pdf/burroughs/B6500_6700/5000094_B6700_ESPOL_Jun72.pdf http://en.wikipedia.org/wiki/ALGOL_60
http://en.wikipedia.org/wiki/NEWP
)

Thank you very much ,

Mehmet Erol Sanliturk

Guillermo Martínez Jiménez wrote

By the way, Linux is good as it is now and Pascal isn't the best option to create an operating system.

Other than Object Pascal not being as popular/mainstream as C... is there any technical limitations in the language and the reason you say it's not a good choice?

Regards,
- Graeme -

_______________________________________________
fpGUI - a cross-platform Free Pascal GUI toolkit
http://opensoft.homeip.net/fpgui/

Marc Santhoff wrote

It has happened in the past. There has been OS/2 for example, with it's superior OO design. RIP.

Ah yes... I surely miss OS/2! The Workplace Shell was awesome. Drag-and-Drop and Customization was awesome!

Regards,
- Graeme -

_______________________________________________
fpGUI - a cross-platform Free Pascal GUI toolkit
http://opensoft.homeip.net/fpgui/

In our previous episode, Guillermo Mart�nez Jim�nez said:

I don't see original Pascal started and ended a bit higher. But this is a Free Pascal list, and Free Pascal and Delphi can get down and dirty too.

There sure isn't much in C that FPC can not do. And the few bits that miss (if any) would probably be added soon when major OS development would start.

I think it is more a matter of FPC being geared towards apps development as a compiler than a matter of language.

I'm not saying it's impossible: here you have MacOS and Toro. I'm just saying that _I think_ it isn't the best option. Of course a better option is to write the kernel in C and Assembler and the utilities in Pascal and Object Pascal.

Well, it is a pity that there is so much routine discussion in this thread that seems to boil down to a dogmatic kneejerk "C is better, C has always been better, because Linux/Unix was programmed in it", and so little real funded argumentation why this is really the case.