Loading ...
Sorry, an error occurred while loading the content.

Re: How do you choose your programming language ?

Expand Messages
  • ik_5
    ... I m sorry to say that I do not know O Cmal or Haskell to do any comparison with them. ... Because Pascal took a long road since the first design of Wirth,
    Message 1 of 44 , Nov 21, 2005
    • 0 Attachment
      --- In hackers-il@yahoogroups.com, Omer Zak <omerz@a...> wrote:
      > On Mon, 2005-11-21 at 13:28 +0000, ik_5 wrote:
      > > You are missing then a lot about Pascal :)
      > > This is not a new Pascal, but rather the development of the language.
      > > If you would have worked with Borland Pascal (Not Turbo Pascal), this
      > > changes where less different then what you think on them now.
      > First of all, I am hereby officially registering my profound objection
      > to the failure of this mailing list to host a violent and highly
      > interesting religious war over the relative merits of languages like
      > Haskell and O'Caml, which brought to the table novel concepts, which are
      > missing from languages like C++ and Pascal.

      I'm sorry to say that I do not know O'Cmal or Haskell to do any
      comparison with them.

      > Now, if we are discussing Pascal, why not argue about more recent
      > languages from Niklaus Wirth's heritage, such as Modula-2, Modula-3 (not
      > designed by Wirth himself) or Oberon-2?

      Because Pascal took a long road since the first design of Wirth, and
      now it's not a student based language, but a strong language that
      stands on it's own... Modula was created as far as I know, to be
      mathematical based programming language with the idea of Fortran but
      developer friendlier ... and with better syntax :P

      > Another principle more worthy of arguing about:
      > Which language scales better for giant projects (and I mean projects
      > with 10^7 lines of source code)

      Well from experiences of coding huge projects with Delphi, VB and few
      other languages, I must say that Delphi was the best tool for
      maintaining huge projects. One of the reason is the IDE, and the
      second is the ease to make your code part of the IDE on design and run
      time... And thats before actually using the project handling tools and
      many other features inside the IDE (tracking about some deceleration
      up to the source file, with line, to start, or the ability to flexible
      the use of many units and libraries using one IDE.
      Integrated Custom Help files with the help system, tools to create
      "Delphi Docs", and many third party components also ease the job.

      BTW, Lazarus seems to have this same features more and more, and
      Lazarus have also stand alone tools.

      And speaking of Syntax usage. How do you declare today something in a
      header and use the same declaration in the source file (like .c) ? I
      guess Copy paste. On Delphi you just wrote the "header" part and
      pressed on key combination to have the right deceleration works for
      you on the code section (with the use of class namespaces etc...).

      And I can continue if you really want to hear more (Code templates in
      the IDE and replace it with the right paramters, so you will not need
      to write

      if boop then beep else baap


      if boop then
      else begin


      >and why? I am fed up with sandbox level
      > quibbles about the relative merits of
      > if (boop)
      > then begin; beep; end;
      > else begin; baap; end;
      > endif;
      > versus
      > if (boop) { beep; } else { baap; }
      > or
      > if (boop)
      > beep
      > else
      > baap
      > After you have spent time typing the above stuff, it no longer matters
      > which is easier to write. What matters is the difficulty to maintain
      > those pieces of smelly brown matter. And this is the bane of large
      > scale projects.
      > Let's argue like grown-up persons having wicked fun in the most naughty
      > nightclubs while trying to conquer the wilderness over there instead of
      > toddlers who play in the sandbox under the protective eyes of nannies
      > from the neighborhood nunnery.
      > --- Omer
      > --
      > Sent from a PC running a top secret test version of Windows 97.
      > My own blog is at http://www.livejournal.com/users/tddpirate/
      > My opinions, as expressed in this E-mail message, are mine alone.
      > They do not represent the official policy of any organization with which
      > I may be affiliated in any way.
      > WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html
    • Shlomi Fish
      Hi Ido! I should note that your mailer breaks the quoted parts of the message, and puts some of the quoted parts in separate lines. Please use a different
      Message 44 of 44 , Dec 15, 2005
      • 0 Attachment
        Hi Ido!

        I should note that your mailer breaks the quoted parts of the message, and
        puts some of the quoted parts in separate lines. Please use a different
        mailer for Hackers-IL.

        Also of note is that you have many spelling and grammatical mistakes in your
        English. It makes them inredibly annoying for everyone to read. Please take
        care to check and double-check your messages.

        See below for my response.

        On Sunday 11 December 2005 19:13, ik_5 wrote:
        > --- In hackers-il@yahoogroups.com, Shlomi Fish <shlomif@i...> wrote:
        > > Hi Ido!
        > >
        > > It seems that your mailer (gmail ?) does not keep the "Re:" in the
        > subject
        > > line and also heavily mangles the quoted texts. Please fix it, or
        > switch to a
        > > different mailer for this list's communication.
        > >
        > > On Saturday 10 December 2005 18:25, ik_5 wrote:
        > > > --- In hackers-il@yahoogroups.com, "Nadav Har'El" <nyh@m...> wrote:
        > > > > On Sat, Dec 10, 2005, Shlomi Fish wrote about "Your (Annoying) Raves
        > > >
        > > > about Pascal [was Re: [hackers-il] Re: How do you choose your
        > > >
        > > > programming language ?]":
        > > > > > On my
        > > > > > system, there's not a single program written in Pascal, yet almost
        > > >
        > > > everything
        > > >
        > > > > > is written in C or in something that is written in C. And I like
        > > >
        > > > it that way.
        > > >
        > > > > Actually, there is one program written in Pascal: TeX.
        > > > >
        > > > > But, the exception sort of prove the rules. If Pascal is so
        > great, how
        > > > > come that 15 years after I first used it, still none of the
        > programs I
        > > > > use (or justo one) is written in it? Makes me wonder too.
        > > >
        > > > If you have used Mac (Before Mac OSX), you would have encounter a lot
        > > > more programs written in Pascal (The official compiler that came with
        > > > Mac).
        > > >
        > > > As a sokoban player, this is a really known issue (some of the best
        > > > sokoban programs written for Mac):
        > > > http://members.aol.com/SokobanMac/
        > >
        > > Many DOS programs were written in Turbo Pascal, and later on many
        > Windows
        > > programs have been written in Delphi. But we are talking here about
        > Linux and
        > > UNIX in general, not of programs of mostly historical interest or of
        > programs
        > > that you'll now run under an emulation or virtualisation.
        > Well, First of all I'm using FPC that is written with Pascal :)
        > I don't want to give a full list of programs that written in Pascal
        > and can run on Linux as well (I'm using few of them), so I'll give you
        > something else:
        > http://www.kanzelsberger.com/pixel/?page_id=12
        > (http://en.wikipedia.org/wiki/Pixel32). BTW some of the code of this
        > program was donated back to FPC ! for example they wrote the compiler
        > port for BeOS, the handle of Image headers also avilable togather with
        > FPC etc..
        > I also play this very addictive game:
        > http://lwars.sourceforge.net/index.php
        > And I also use lazarus.freepascal.org.
        > I will not continue giving additional programs because I see the you
        > all starting arguing about it and blaiming me again.

        I didn't say that there were absolutely no programs written in Pascal for
        Linux or other UNIXes. It's just that for every program written in Pascal
        there are a hundred programs written in C and/or C++. (to say nothing of
        Perl, PHP, Python, etc.). And as I noted - I, and most other people, were not
        using any of them.

        > > > This tools should answers problems on any side, and will be the reason
        > > > for others to move to an open source slution rather then a closed
        > > > source one.
        > > >
        > > > And therefor I belive that if we will not create a port of this tools
        > > > to any closed source OS, and will control what can be ported, we can
        > > > have a real competition that everyone will benifit from it.
        > >
        > > [Later Text including Raves about Pascal Snipped]
        > Funny, When people bashing Java on a conversation about Pascal it's
        > ok, but when there is a conversation about Pascal, it's a rave ...

        First of all, your original conversation about Pascal was labeled "How do you
        choose your programming language", so naturally it would divert into
        conversations about Java. (Plus, many discussions divert into other topics -
        it's natural).

        Secondly, may I ask what is wrong with calling what you said "raves"? A rave
        is the opposite of rant:

        Quoting from http://www.m-w.com/cgi-bin/dictionary?book=Dictionary&va=rave :

        c : to talk with extreme enthusiasm <raved about its beauty>

        > > > > > Another annoying issue is the fact that you support the claim that
        > > >
        > > > developers
        > > >
        > > > > > should not port their programs to Windows. I and many other people
        > > >
        > > > have
        > > >
        > > > > I actually agree with this :-) Stuff I write (say, Hspell) can
        > be ported
        > > > > to Windows if someone cares enough to, but I'll be damned if I
        > do this
        > > > > porting myself. I do not use Windows, I don't plan to ever
        > switch to it,
        > > > > and I am not going to spend a minute on porting my work to a
        > system I'll
        > > > > never use.
        > > >
        > > > Actually I belive that if we wish to create a real competition with
        > > > Windows, Mac and every non open source OS, we must create tools that
        > > > will be helpfull and unique to Linux, or any other open source OS.
        > >
        > > I see several problems with this philosophy:
        > I see several problems with your presepsion of freedom.

        That was a non-sequitor. I'll still answer to it.

        > For example you do not like that I'm talking about Pascal, so you try
        > to shout me up. Isn't that something that hurt *my* freedom ? So why
        > not limiting the freedom on this subject as well ?!

        I didn't try to shut you up. I just tried to inform you that I (and probably
        many other people) found this pro-Pascal propaganda annoying and useless. You
        are free to say these things, but I warned you that you are only going to
        make me (and probably many other people) annoyed.

        And as part of my freedom I hold my right, not to *read* what you wrote or
        simply skip it, because I know in advance it will annoy me and will not
        contribute any useful insights.

        > See the rest of the response inline.
        > > 1. The very free software definition claims that the user should
        > have "the
        > > freedom to run the program, for any purpose". This includes running
        > it on
        > > proprietary operating systems. The Open Source Definition has a similar
        > > clause.
        > But if by doing so, this freedom is used aginst you.

        How so?

        > For example
        > Saffari. As long as Apple needed KHTML as-is, they used it and gave
        > back to the "community", but when they could leave KHtml and use their
        > own based KHTML engine, they left the project alone with many excuses.

        So? The KHTML people could still benefit from what Apple contributed while it
        did so? And isn't Apple still using a derivative of KHTML in their products?
        (while respecting its GPL and LGPL licenses?).

        > So now, all of the working hours of open source was given to
        > commercial use, that spit back to the community.

        What do you propose instead? Making KHTML non-free? I can think of similar
        risks for free software even if it only runs on Linux/BSDs.

        > Tomarrow Apple can
        > register the engine as a patent and prevent KDE of using their own
        > engine. So what can you do about that ?

        Patents are something else that is not related here. Tomorrow I can register a
        patent on the "Hello World" program and sue GNU for GNU hello. And how does
        restricting a software to run only on free operating systems will help you
        against software patents? You can issue a patent on that too.

        > > 2. The issue is not Linux vs. Windows. (or any other proprietary
        > operating
        > > system) The issue is free software vs. proprietary software. Making
        > a free
        > > program available under Windows, is better than making the Windows
        > users use
        > > an equivalent non-free program. That way, their freedom will be reduced.
        > If people saw "open source" and not "free pgorams" (I don't see Open
        > source as free!) you may be right, but people see "amm... I can get
        > this without the need to pay" instead of "can you copy the cd you just
        > bought ?"

        First of all let me say that I think you evaded the point here.

        So? Many people see the same thing when using Linux, which all they know about
        is that it is a piece of costless software. When I first started seriously
        using UNIXes (FreeBSD, SunOS 4.1.3, and IRIX) at my work in a web-design
        shop, all I knew about GNU was that I could FTP into prep.ai.mit.edu and
        download lots of useful software which I compiled for source. I did not know
        about the four Stallmanian freedoms, or about the GPL, copyleft, etc. It was
        just gratis sourceware for me.

        Many people who install Linux or who are made to use it by someone, are also
        not aware of this.

        Windows users are still better off using free software instead of its
        proprietary alternatives, because that way they may become better aware of
        the free software ideology as well as be more willing to try Linux. And they
        can also contribute bug reports, bug fixes, etc. A free software developer
        should aim to have the largest number of "passive" (i.e:
        not-particularly-contributing) users, and porting the software to Windows
        help that a lot.

        > > 3. How exactly are you going to prevent people from porting your
        > software to
        > > non-open-source OSes? Are you going to make them binary only? Are
        > you going
        > > to make their licensing restrict running them on non-free OSes?
        > (which will
        > > in turn make *that* software non-free). Are you going to use APIs
        > that are
        > > specific to free operating systems or even a particular one?
        > >
        > > I believe you agree that the first two options are not in accordance
        > with the
        > > FOSS ideology. The third option, while often necessary, can easily be
        > > overcome by a couple of conditionals, and if abused, will cause your
        > software
        > > to be ruled out of integration with the portable mainstream
        > programs. Plus,
        > > portable programming is a virtue not a vice, and one should not write
        > > non-portable programs on purpose.
        > License is one way, another is to avoid placing any code that can be
        > converted easly to non open source/free based OS.
        > > 4. It is easier to tell someone to switch to Linux after a large
        > amount of the
        > > programs he's using are available there too. The best way we can
        > ensure that
        > > is by making sure FOSS programs are portable to Windows as well as
        > Linux.
        > "If I have all of this tools on Windows why should I move to a new OS
        > ?" is one of the answers I got regarding this approach.
        > The PC history teaches us, that Electronic sheets was the reason at
        > the 80's to work with a PC with IBM/Microsoft DOS. It far better then
        > do it by hand with too many papers.
        > Mac teaches us, that if you give something good for graphics including
        > the programs for it built-in, graphics will be made with Mac.
        > Microsoft teaches us, that when you can place enough needed product
        > into one type of OS only, then people will use that OS only !
        > So why not to make linux/BSD/whatever OS the same thing ???

        That's because Linux has many inherent advantages over Windows. In Linux you
        can expect the same programs to work faster, be more stable, be better, etc.
        You can also install them more easily, hack on them more easily and better,
        etc. Linux also has some stability, maintenance, feature-set, security and
        other advantages over Windows. I was told that at Microsoft conferences,
        people get psyched over new Windows features that Linux and other UNIXes has
        had for years. Linux is also free as in speech and thus, is constantly
        enhanced by the community, and also has a much better public image than
        Windows at the moment.

        Porting software from UNIX to Windows is not going to make Linux less
        advantageous than Windows. But it does going to make FOSS more popular.

        > > 5. There are many places where people work on both Windows and
        > Linux, and need
        > > support for both. They are not going to introduce a Linux only solution
        > > there.
        > There are many people that uses Windows only and does not know what
        > Linux is. And there are few people (like my father) that only knows
        > Linux because thats the only OS I'm willing to support, so he does not
        > need Windows at all.
        > This are your audiences. The Windows only users, and of curse the
        > Linux only users. Your audience is not the dual users. Because they
        > will find it on their own.

        You are evading. I said that if you want your software to be popular you have
        to make sure it runs on both Linux and Windows, or else people who have to
        use both, will probably not use it.

        > > 6. It is possible to run Linux on top or side-by-side with Windows.
        > This can
        > > be achieved using Xen, VMWare, CoLinux, UserModeLinux (possibly in
        > the future
        > > when it will be ported to run on top of Win32), and many other cool
        > hacks
        > > like this. So what exactly have you acheived by purposely making your
        > > programs Linux-only?
        > So ?! I can have a super computer by making grids ... does it mean
        > that I really going to do it ? What about organisation that does not
        > allow to do such things. And I think that many emulators consider to
        > be illeagel ...

        You misunderstood me. What I said was that many (most?) people would be able
        to run Linux on top of Windows using these Emulators. Thus, they can enjoy
        the benefits of your Linux programs running on top of their Windows, whether
        you explicitly port it to Windows or not.



        Shlomi Fish

        Shlomi Fish shlomif@...
        Homepage: http://www.shlomifish.org/

        95% of the programmers consider 95% of the code they did not write, in the
        bottom 5%.
      Your message has been successfully submitted and would be delivered to recipients shortly.