home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.c++
- Path: sparky!uunet!microsoft!hexnut!jimad
- From: jimad@microsoft.com (Jim Adcock)
- Subject: Re: feedback wanted on appropriate OOPL
- Message-ID: <1992Dec29.230713.9401@microsoft.com>
- Date: 29 Dec 92 23:07:13 GMT
- Organization: Microsoft Corporation
- References: <92358.192930COP80196@UCF1VM.BITNET> <1992Dec28.173620.14793@microsoft.com> <1992Dec29.011354.5929@informix.com>
- Lines: 80
-
- In article <1992Dec29.011354.5929@informix.com> cshaver@informix.com (Craig Shaver) writes:
- |In article <1992Dec28.173620.14793@microsoft.com> jimad@microsoft.com (Jim Adcock) writes:
- |>2) Having just done an interview with a interview candidate who has
- |>an MS in CS, who claimed C/C++ as a primary language, but who couldn't
- |>program simple problems in a straight-forward manner, using style
- |>appropriate for the C/C++ languages, I must again express frustration
- | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
- |
- |I have used both Smalltalk and C++ and do not agree with your conclusions
- |without concrete evidence to the contrary. Could you please elaborate
- |on what style is appropriate for c/c++? It would be a help to have
- |some references to material that clarifies the differences in OO style.
-
- I am expected to provide you with concreate evidence about an
- interview candidate ??? The kinds of problems I'm talking about
- are people who don't know the difference between pointers and
- references and pass-by-value, don't understand const, don't know
- how to program without GC, can't get pointers right, don't get
- their Ctors and Dtors right, who want to do everything on the heap, etc....
-
- |>at CS departments who don't teach students the skills they need to
- |>get a job. C is the mainstream programming language for the software
- |>industry. C++ is the mainstream OO programming language for the
- |>software industry. Non-mainstream languages are fine for PhD candidates
- |>who plan to remain in acedemia.
- |
- |I thought COBOL was the mainstream language?
-
- It was. So was mainframes.
-
- |Could you support this
- |set of statements with some examples? Evidence?
-
- I have done so many times and will again. Consider, for example one
- measure, namely the number of references in the Computer Select Database
- in 1992 of articles from the major computer magazines. Number of
- references to various languages we have been discussing:
-
- C 39498
- C++ 8148
- COBOL 6179
- Smalltalk 841
-
- Same source, number of software products listed referencing each
- above language:
-
- C 9748
- COBOL 3222
- C++ 572
- Smalltalk 39
-
- If you look at a large number of languages over time, you see C/C++
- gaining [C++ strongly] and leading the pack by a large margin. Then
- comes a middle grouping of a large number of historical languages --
- pascal, fortran, cobol, ... then bringing up the rear is a very large
- number of "esoterica" languages that never seem to reach critical
- mass for some reason or another -- typically because they never get
- even one stable efficient implementation on even one popular machine.
-
- Other measures are relatively easily available from a number of
- subscription "industry watch" services. Check out the number of
- compilers that get sold for what languages, and by who, for example.
-
- Other simple "measures" are to go to a good computer book store and
- count the number of texts available for different languages. C++
- has 100+ texts in print, for example.
-
- |>3) Neither 1) nor 2) above is as important as having students who
- |>know and can perform the necessary applied mathematics.
- |
- |This is a very broad statement. Could you give an example please?
-
- Sure. If you must write a bubble sort you ought to at least be able to
- quickly determine its O(n^2). You ought to be able to reduce
- standard sums, or at least know that they can be reduced. If you
- do engineering you should know complex numbers .... Most people
- need to know statistics so that they can handle at least the
- rudiments of quality control, or at least realize when someone
- else is screwing it up, etc.
-
-