home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.databases.theory
- Path: sparky!uunet!sun-barr!decwrl!csus.edu!csusac!unify!Unify.com!sean
- From: sean@Unify.com (Sean P. Curley)
- Subject: Re: What constitutes a 4GL?
- Message-ID: <blzeo58@Unify.Com>
- Sender: news@Unify.Com (news admin)
- Organization: Unify Corporation (Sacramento)
- References: <5188@airs.com> <1992Aug31.045102.25682@cc.uow.edu.au>
- Date: Mon, 31 Aug 92 18:07:42 GMT
- Lines: 45
-
- In article <1992Aug31.045102.25682@cc.uow.edu.au>, u9048979@cc.uow.edu.au (glen eastment) writes:
- > andrew@airs.com (Andrew Evans) writes:
- >
- > >At the risk of asking a seemingly trivial question compared to most of
- > >the other discussion that goes on around here, I'd like an
- > >authoritative description of what a fourth-generation language really
- > >is. In the scope of relational databases, is it simply a programming
- > >language with a database-oriented syntax? Or are there more specific
- > >attributes that a system must have in order to really be a 4GL?
- > >And who says that this is what a 4GL must be?
- > >--
- > >Andrew Evans (andrew@airs.com) - Infinity Development Systems - Waltham, MA
- >
- > I've always believed that a 4GL is where the programmer can specify *WHAT*
- > he/she wants, without having to specify *HOW* to do it. Sadly, many
- > 4GL still don't come close to this - there is still an amount of *HOW*, and
- > probably always will be. 4GL's these day are usually bundeled with tools
- > that take much work out of the *HOW* part - such as screen genrators. As
- > far as PC's go - I would include applications generating languages such
- > as DBASE IV as getting pretty close to 4GL's as far as database applications
- > go (In fact I would include the entire XBase language in this statement).
- >
- > Anyway, these are my opinions.
- >
- > Cheers!
- >
- > <GE> a.k.a. u9048979@wampyr.cc.uow.edu.au
-
- I think Douglas was close in his comments about 4GLs. And the slams were deserved.
-
- You might consider the name "4th Generation language" - it comes from a series of
- language classifications with 1st being machine language, 2nd being assembly and 3rd
- being C and other "more human-oriented" languages. I was taught that these levels
- of languages represent productivity for programmers. It was calculated at some
- point that an assembly language programmer could work more than 10 times faster than
- a machine language programmer (a magnitude or generation) and Cobol/Fortran/C/Pascal/etc.
- programmers could work more than 10 times faster than assembly programemrs, etc.....
-
- 4GLs therefore should be at least 10 time more productive than 3GLs.
-
- I have heard a lot of other defenitions about 4GLs, but most were Marketing Hype that
- companies use to disqualify competiters.
-
- Sean
-
-