Re: Is CM important? (was: Use pragma INLINE or not?)

Re: Is CM important? (was: Use pragma INLINE or not?)

[Greg Harvey]

In <15645@encore.Encore.COM> jcallen@Encore.COM (Jerry Callen) writes:

[previous letter stuff edited out here.]
>Don't drop it, but please make sure that you DO find time to teach something
>about CM. How many disasters could have been avoided if:

>- All of the source had been under CM?

>- The CM system could have easily recreated a particular collection of
>  source modules?

>- A good makefile existed?

>- Some regression testing had been done?

>Having a brilliantly crafted design doesn't help if the design disappears
>into the bit bucket because of an error with an editor, or a system crash,
>or some other problem that computer systems and users are all too prone to.

>It's hard to appreciate the value of a good CM system until you have used one.
>No, I don't mean RCS and make; they're useful tools, but they are just 
>pieces of what should be a larger system. 

Personal experience time:

I walked into a company (who shall remain nameless) where a group of
programmers were working on a postscript language intepreter/printer.
I considered each of the programmers on that team to be significantly
more competent in computer science than I was.  My main contribution
to the effort was to help establish configuration management and
help establish a thorough going test suite for the system.  I believe
that these additions help increase the value of THEIR efforts
several fold, but of course these changes would have been of no
value if these programmers had not been competent.

BTW...the customer that was contracting this work was so pleased
with these changes that they swallowed without comment a doubling
of the rates we were charging, just because of the enhanced value
of the additional management.  The comment I treasure the most,
though, was when one of the lead programmers (who co-owned the
company, by the way) made a comment later about how much smoother
things went under the new system than under the old approach.

I know this sounds like horn tooting, and I'm sorry about that.  I
was very excited at the added benefit just a little CM gave us.

[more stuff deleted]

>-- Jerry Callen
>   jcallen@encore.com


--
If you get the impression I'm not qualified to speak for my company, it's
because I ain't, I can't, I don't, I won't, and I don'wanna.
Greg Harvey                    --Temporarily without mail service
Lockheed, Houston Texas        --Hope to have a PSCNI route soon! 

Re: Is CM important? (was: Use pragma INLINE or not?)

[Michael Feldman]

In article <15645@encore.Encore.COM> jcallen@encore.Com (Jerry Callen) writes:
>
>Don't drop it, but please make sure that you DO find time to teach something
>about CM. How many disasters could have been avoided if:
>
You are certainly correct here.

This is all getting away of the Ada purpose of the group, but one more $0.02:

I'm a tad oversensitive these days to the current fashion to bash computer 
science as producing "irrelevant" industry folks, 'cause we don't teach enough 
"software engineering" stuff. When I ask them - in all sincerity - which 
undergraduate courses we should drop (to make room in a 4-year program), the 
answer is almost always

- operating systems
- compilers
- algorithm theory
- formal languages and automata

Would you like a software engineer who had never studied:

- enough (1 course) on OS resource allocation, process scheduling, and
  memory management to understand something of concurrency, deadlock,
  and garbage collection;

- enough (1 course) on parser construction to be able to read/write a simple
  parser for a nontrivial command language? Parsers are everywhere:
  prettyprinters, style checkers, even editors that allow search and
  replace by giving an expression that needs to be analyzed;

- why no amount of bit-fiddling will improve an O(N**3) algorithm (and
  other such considerations);

- why many problems that seem easy to express in words cannot be solved at
  all in tractably finite time (such as "why do we need to test - why can't
  we just prove a program correct?")

True, there are other things about the software life-cycle we may be doing 
less than full justice to, but arguing that we should kill off the very
foundations of ALL programming, leaving the software engineer with no
bedrock fundamentals, is going rather too far, dontcha think?

I'd be happy to continue this by e-mail with all comers. My ears and
mind are open to suggestions. Is this a "too many notes" problem? Exactly
which subjects do we trade for more SE? There's no free lunch, sadly.

Mike

Re: Is CM important? (was: Use pragma INLINE or not?)

[Jim Showalter]

jcallen@Encore.COM (Jerry Callen) writes:

>Real programmers make mistakes: syntactic errors, letter transpositions,
>even (horrors!) logic errors. I like to check in after every edit, so
>all those errors are visible. So what? If it "taxes" your CM system to
>have a lot of check-ins then you need a new CM system.

>Or would you rather see an afternoon of work go down the tubes because of
>a power flicker?

This makes no sense to me. If I have saved a file to disk, it is saved
whether it is checked in or not. I don't check in stuff I'm not yet happy
with--why would I want to store line differentials on a piece of crud?
The only reason I can see for checking something in is to have a record
of milestone to which I would WANT to return should I decide I want to
revert later on. I hardly think I'm going to want to revert to something
shot full of syntax errors, letter transpositions, and logic errors.
-- 
*** LIMITLESS SOFTWARE, Inc: Jim Showalter, jls@netcom.com, (408) 243-0630 ****
*Proven solutions to software problems. Consulting and training on all aspects*
*of software development. Management/process/methodology. Architecture/design/*
*reuse. Quality/productivity. Risk reduction. EFFECTIVE OO usage. Ada/C++.    *

Re: Is CM important? (was: Use pragma INLINE or not?)

[Jim Showalter]

Mike asks what should be dropped from a computer science curriculum in
order to make room for more software engineering courses.

General ed.

Yes, yes--I've heard all the talk about how we want to turn out fully
formed humans with wide ranging interests in things as diverse as
foreign languages, Shakespeare, etc.

Bull.

What we OUGHT to be turning out is crack software engineers with a sound
background in fundamental computer science. If Shakespeare has to take
a backseat to that, so be it. The rationale for forcing someone who is
not interested in, say, 14th century history to learn it anyway has always
eluded me--it's not as if the person comes away from a course on it that
he/she was forced to take and that interfered with their actual area of
study any wiser, happier, or more inclined to love history. Let those
who have wide-ranging interests pursue them, by all means--but let those
who are techo-weenies and who will always BE techno-weenies at least
excel at it.
-- 
*** LIMITLESS SOFTWARE, Inc: Jim Showalter, jls@netcom.com, (408) 243-0630 ****
*Proven solutions to software problems. Consulting and training on all aspects*
*of software development. Management/process/methodology. Architecture/design/*
*reuse. Quality/productivity. Risk reduction. EFFECTIVE OO usage. Ada/C++.    *

Re: Is CM important? (was: Use pragma INLINE or not?)

[Greg Eisenhauer]

In article <1991Jul3.233045.11174@netcom.COM>, jls@netcom.COM (Jim Showalter) w
rites:
|> Mike asks what should be dropped from a computer science curriculum in
|> order to make room for more software engineering courses.
|> 
|> General ed.
|> 
|> Yes, yes--I've heard all the talk about how we want to turn out fully
|> formed humans with wide ranging interests in things as diverse as
|> foreign languages, Shakespeare, etc.
|> 
|> Bull.
|> 

Jim, this has nothing to do with Ada.  I'm sure that there must be some
newsgroup out there where people are interested in your opinions about
how to turn out techno-droids who are as interesting as their keyboards,
but comp.lang.ada is probably not the best place to start.  Can you try
to get your posting enthusiasm under control?  

Re: Is CM important? (was: Use pragma INLINE or not?)

[Jim Showalter]

eisen@cc.gatech.edu (Greg Eisenhauer) writes:

>Jim, this has nothing to do with Ada.

You're right. Sorry. Completely forgot that software engineering and
software engineering education have nothing whatsoever to do with Ada.
Nor do maintenance, readability, tools, or many of the other topics
that have arisen in this group from time to time. Sorry. Really must
remember to keep the conversation totally narrowly focused on the one
or two questions per week that have to do with something like mutable
record semantics or compiler recommendations for DOS platforms.

>I'm sure that there must be some
>newsgroup out there where people are interested in your opinions about
>how to turn out techno-droids who are as interesting as their keyboards,

Better competent techno-droids with a complete grounding in their 
area of expertise than well-rounded semi-competents lacking an
education in even fundamental computer science. Besides, the idea
that you can "make" a person interesting is absurd. Interesting
people take classes in philosophy and psychology on their own. Dullards
remain dullards whether or not they have completed introductory
check-in-the-box general ed courses or not.
-- 
*** LIMITLESS SOFTWARE, Inc: Jim Showalter, jls@netcom.com, (408) 243-0630 ****
*Proven solutions to software problems. Consulting and training on all aspects*
*of software development. Management/process/methodology. Architecture/design/*
*reuse. Quality/productivity. Risk reduction. EFFECTIVE OO usage. Ada/C++.    *

Re: Is CM important? (was: Use pragma INLINE or not?)

[Michael Feldman]

In article <46820007@hpcupt3.cup.hp.com> edge@hpcupt3.cup.hp.com (Kathy Edge) w
rites:
>
>My suggestion would be to follow basic SE principles in teaching
>the other subjects.
>
I quite agree. In fact, my colleagues and I have been saying for years,
"software engineering is what we all do and teach, in all courses."

Trouble is, I keep hearing from industry folks that they don't see courses
titled SOFTWARE ENGINEERING on the transcripts of their interviewees. I
sometimes think they're just looking for an easy way to sift applicants.

Back in the old days, when "structured programming" was new, we made a big
deal out of it. Hardly anyone uses that term these days, because what we
used to call SP is now so much a part of our daily programming that it's
no longer a separate agenda item. In my opinion, this is what needs to
happen with SE. Indeed, it IS happening. If you look at the more recent
1st-level books - in whatever language - they are all doing pretty decent
SE, albeit on a very simple level for beginners, and not usually 
involving teams. But the term SOFTWARE ENGINEERING is hard to find in these
books. Instead of making a big deal out of it, they're pretending that
there is no other way to do it. This is exactly right!

There are really 2 kinds of SE - I'm indebted to Chuck Engle for teaching
me the terms "micro-SE" and "macro-SE." The micro kind is the stuff at the
program level, including modules (or whatever your favorite language calls
them), interface/implementation issues, etc. The macro kind is all the
system-development superstructure: documents, testing strategies, etc.

Surely micro-SE is being taught these days as an inherent part of any
reasonable CS course, yes even compilers (certainly I teach it there).
The macro kind ought to be also infused into everything. I observe
that indeed much prohress is being made here, to judge from what's
reported at SIGCSE conferences, etc. Maybe this'll give my industry 
friends reason for hope after all.

Mike

Re: Is CM important? (was: Use pragma INLINE or not?)

[Kennet h B. Hawks]

In article <15645@encore.Encore.COM> jcallen@encore.Com (Jerry Callen) writes:
>  --Stuff Deleted--
>
>Real programmers make mistakes: syntactic errors, letter transpositions,
>even (horrors!) logic errors. I like to check in after every edit, so
>all those errors are visible. So what? If it "taxes" your CM system to
>have a lot of check-ins then you need a new CM system.
>
>Or would you rather see an afternoon of work go down the tubes because of
>a power flicker?
>
>-- Jerry Callen
>   jcallen@encore.com

Jerry - does your _formal_ CM really capture and baseline all your key strokes?
If so, WOW is big brother watching you!!!

There should be a vast difference between the formal CM baseline and the
day to day output within an individual developers domain.  What I was pointing
out was the "hacker" tends to bash things out, say "OK, boss I'm done." 
Put their output in the master library, and then 'discover' they forgot some- 
thing important.  When this happens frequently/consistantly I have to conclude
they are not doing a complete and thorough job.

The points you raise are indicative of the CM system itself, (both procedural
and tool capability), not of the need and proper application of formal CM
controls.  Incidently, on projects that I have seen each keystroke captured
into the formal CM system, the CM people soon outnumbered the developers
and management had to abandon the approach.  The power surge gotcha is 
handled by individual discipline, (how often do you save your work), and system
management, (how often does the system create a journal file for you).

Kenneth B. Hawks                                   |\   /|   "Fox Forever"
Rome Laboratory, Griffiss AFB, NY                   ^o.o^         BSA
hawksk@lonex.radc.af.mil                            =(v)=
Disclaimer:  There is no one else here who thinks like I do; therefore....

Re: Is CM important? (was: Use pragma INLINE or not?)

[Matt Telles]

In article <1991Jul3.233045.11174@netcom.COM> jls@netcom.COM (Jim Showalter) wr
ites:
>Mike asks what should be dropped from a computer science curriculum in
>order to make room for more software engineering courses.
>
>General ed.
>
>Yes, yes--I've heard all the talk about how we want to turn out fully
>formed humans with wide ranging interests in things as diverse as
>foreign languages, Shakespeare, etc.
>
>Bull.
>
>What we OUGHT to be turning out is crack software engineers with a sound
>background in fundamental computer science. If Shakespeare has to take
>a backseat to that, so be it. The rationale for forcing someone who is
>not interested in, say, 14th century history to learn it anyway has always
>eluded me--it's not as if the person comes away from a course on it that
>he/she was forced to take and that interfered with their actual area of
>study any wiser, happier, or more inclined to love history. Let those
>who have wide-ranging interests pursue them, by all means--but let those
>who are techo-weenies and who will always BE techno-weenies at least
>excel at it.
>-- 
>*** LIMITLESS SOFTWARE, Inc: Jim Showalter, jls@netcom.com, (408) 243-0630 ***
*
>*Proven solutions to software problems. Consulting and training on all aspects
*
>*of software development. Management/process/methodology. Architecture/design/
*
>*reuse. Quality/productivity. Risk reduction. EFFECTIVE OO usage. Ada/C++.    
*


Jim,
   This really doesn't belong in comp.lang.ada (Maybe comp.software-eng??)  but
I will answer your comment anyway.  I am a programmer/system engineer/whatever
you want to call me.  I work with many other programmers, many of whom:

   * cannot write in English (or any other non-programming language).
   * cannot reasonably understand user requirements because they do not have an
y
common frame of reference.
   * cannot do the job because of the above two points.
 
   If you want to drop something from the Computer Science degree area, try the
following:

   - Calculus:  No one uses the stuff anyway and you can always look it up.
   - Algebra :  "" 
   - FORTRAN :  This is obvious ..

   - Ada     :  Hey! This way the rest of us can get jobs ..


:)'s for the  humor impaired .....

Matt


-- 
==============================================================================
Matt Telles 	                 mattel@auto-trol.COM
                                 {...}ncar!ico!auto-trol!mattel
Auto-trol Technology 12500 N Washington Denver, CO 80241-2404 (303)252-2874

Re: Is CM important? (was: Use pragma INLINE or not?)

[Robert I. Eachus]

In article <1991Jul8.154527.29136@auto-trol.com> mattel@auto-trol.com (Matt Tel
les) writes:

   I am a programmer/system engineer/whatever you want to call me.  I
   work with many other programmers, many of whom:

      * cannot write in English (or any other non-programming language).
      * cannot reasonably understand user requirements because they do
        not have any common frame of reference.
      * cannot do the job because of the above two points.

   Dikstra once said that the best predictor of success in the field
of programming was a person's ability to use their native language.


--

					Robert I. Eachus

with STANDARD_DISCLAIMER;
use  STANDARD_DISCLAIMER;
function MESSAGE (TEXT: in CLEVER_IDEAS) return BETTER_IDEAS is...