home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.lang.c:16602 comp.software-eng:4318 alt.folklore.computers:16296
- Newsgroups: comp.lang.c,comp.software-eng,alt.folklore.computers
- Path: sparky!uunet!snorkelwacker.mit.edu!bloom-picayune.mit.edu!news
- From: scs@adam.mit.edu (Steve Summit)
- Subject: Re: Will we keep ignoring this productivity issue?
- Message-ID: <1992Nov16.160026.7548@athena.mit.edu>
- Sender: news@athena.mit.edu (News system)
- Nntp-Posting-Host: adam.mit.edu
- Organization: none, at the moment
- References: <1992Nov1.132750.9856@vax.oxford.ac.uk> <1776@aviary.Stars.Reston.Unisys.COM> <1992Nov11.055130@eklektix.com>
- Date: Mon, 16 Nov 1992 16:00:26 GMT
- Lines: 76
-
- In article <1992Nov11.055130@eklektix.com>, rcd@raven.eklektix.com (Dick Dunn) writes:
- > This was in the "productivity of a C programmer" thread, about a week ago.
- > I haven't seen anyone picking up on it... Bob Munck (munck@stars.reston.
- > unisys.com) wrote:
- >
- >> ...However, it is a fairly-well-established rule of thumb that
- >> very good programmers can be an order of magnitude or more productive than
- >> the average and do a good job...
- >
- > We toss this number around a lot. A couple years ago I tossed it at a
- > friend... But
- > his retort came from a direction I hadn't expected: he said "It's more like
- > *two* orders of magnitude!
-
- Actually, I had been tempted to comment on Bob Munch's earlier
- remark, to point out that the infamous Jargon File (a.k.a. The
- New Hacker's Dictionary), for what it's worth, has this to say
- about productivity:
-
- ...Productivity can vary from one programmer to another
- by three orders of magnitude. For example, one
- programmer might be able to write an average of 3 lines
- of working code in one day, while another, with the
- proper tools, might be able to write 3,000. This range
- is astonishing; it is matched in very few other areas of
- human endeavor.
-
- This note appears in the entry on "superprogrammer," a term which
- the File notes "is more commonly used within such places as IBM
- than in the hacker community."
-
- You might reasonably suspect that the several appearances of the
- term "hacker" in such close proximity to the above-quoted
- statement dilute the credibility of its three-orders-of-magnitude
- claim, or relegate such code to the undocumented & unmaintainable &
- hence worthless category. The File's discussion on productivity
- continues, however, with the observation that
-
- [Such terms tend] to stress naive measures of
- productivity and to underweight creativity, ingenuity,
- and getting the job *done* -- and to sidestep the
- question of whether the 3,000 lines of code do more or
- less useful work than three lines that do the {Right
- Thing}.
-
- It doesn't really matter whether you believe that the factor is
- 20 or 100 or 1000; the point is that the factor is definitely
- large. And the suggestion that apples and oranges are being
- compared -- that the prodigy can code thousands of <choose your
- metric> per day only for certain kinds of projects or only by
- neglecting all documentation or only by being antisocial and not
- interacting with others -- is sort of dodging the issue: perhaps
- it's worth contemplating for a moment the qualities of programs
- and the methodologies which do permit such productivity, and
- trying to recast recalcitrant projects in more favorable terms.
- I love it when some intractable programming task (ugly, grisly,
- special-cased grunge code that's impossible and no fun to write
- and is guaranteed to have lots of nasty bugs) can, by means of
- some change in methodology or avenue of attack or mere outlook,
- be transformed into blissfully clean code that practically writes
- itself.
-
- I agree completely with Dick that it would be valuable to better
- understand the factors (whatever they are; internal or external
- or otherwise) which allow some programmers to be so much more
- productive than others. That there is a productivity
- differential is not unique to programming, but the reasons behind
- the differential are more likely to be.
-
- I've neglected to honor Dick's Followup-To and have added a third
- crosspost; please redirect followups appropriately (software
- engineering to comp.software-eng, Jargon File to
- alt.folklore.computers).
-
- Steve Summit
- scs@adam.mit.edu
-