NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/comp/lang/cplus/16527

Text File | 1992-11-18 | 2.0 KB | 44 lines

Newsgroups: comp.lang.c++
Path: sparky!uunet!charon.amdahl.com!pacbell.com!sgiblab!sti.com!dgreen
From: [e-mail] (Dan R. Greening)
Subject: Re: New Employee Test
Message-ID: <[e-mail]>
Sender: [e-mail]
Nntp-Posting-Host: smurf
Organization: Software Transformation, Inc.
References: <[e-mail]>
Date: Wed, 18 Nov 1992 19:39:38 GMT
Lines: 31

[e-mail] (ASSIST Coordination Site) wrote:

>i am in the process of hiring 2 C++ programmers.  This Wednesday 11/18/92
>we are conducting the final interviews with 4 candidates.

The problem is interesting, and after a minute or two I figured it out.
However, I'm not sure whether you will get what you want by doing a
programming test.

First, when I first looked at it, I wasn't sure that I could figure it
out, but I have written reference-counting envelope/letter classes (whose
envelope class fits in 4-bytes), a flexible and efficient AVL tree
template, etc.  Isn't that the kind of creativity you would seek, and not
whether someone knows about the default copy constructor?

Second, I get kind of annoyed when people run me through a test
like that: it makes me think that I'm working with someone who can't tell
whether I know C++ through some other means.  I once had a recruiter from
Goodyear tell me that he would be giving me "the Brown test", whatever that
is, to determine how good a programmer I was (you would think my 92 percentile
on the CS GRE exam would have been sufficient).  I never called the
Goodyear rep back, because I don't think I want to work in an environment
like that (untrusting, incompetent, etc.).

Third, whether someone understands the picayune details of a programming
language tells me nothing of their productivity, and their ability to work
with others.  These are things that must be ferreted out.
--
____
\  /Dan Greening    Software Transformation   1601 Saratoga-Sunnyvale Rd, #100
 \/[e-mail]   [phone] x313       Cupertino, CA 95014