Cobol - What matters most when getting a job done?

Author

What matters most when getting a job done?

Pete Dashwood

2007-10-03, 9:55 pm

I think many people here will have thoughts on this topic and may want to
vote in the poll. Be quick, it closes soon...

http://www.codeproject.com/#poll

(Scroll to the bottom of the page)

Pete.
--
"I used to write COBOL...now I can do anything."

Robert

2007-10-03, 9:55 pm

On Thu, 4 Oct 2007 13:19:01 +1300, "Pete Dashwood" <dashwood@removethis.enternet.co.nz>
wrote:

>I think many people here will have thoughts on this topic and may want to
>vote in the poll. Be quick, it closes soon...
>
>http://www.codeproject.com/#poll

Respondants are short sighted. A year hence, no one will care whether the client was happy
or it was delivered on time. They will judge it on quality.

2007-10-03, 9:55 pm

In article <5miprnFdk5d6U1@mid.individual.net>,
Pete Dashwood <dashwood@removethis.enternet.co.nz> wrote:
>I think many people here will have thoughts on this topic and may want to
>vote in the poll. Be quick, it closes soon...
>
>http://www.codeproject.com/#poll
>
>(Scroll to the bottom of the page)

Hmmmmm... 1290 votes before my viewing, the second-highest is 'The job is
error-free' - interesting to consider in light of 'the only bugs in code
are the ones you ain't found yet' - and head-and-shoulders above is 'The
client/boss etc (sic) is happy'.

Seems like I'm not the only one who says that a well-functioning system,
by definition, is one which causes the person resposible for signing the
checks for it to smile. Maybe I should find a better circle with which to
associated.

DD

William M. Klein

2007-10-04, 3:55 am

Choke, gasp, cringe

You have GOT to be kidding!!!

--
Bill Klein
wmklein <at> ix.netcom.com
"Robert" <no@e.mail> wrote in message
news:3li8g39823e64hddtjdkm6r6iq8oihjv0a@
4ax.com...
> On Thu, 4 Oct 2007 13:19:01 +1300, "Pete Dashwood"
> <dashwood@removethis.enternet.co.nz>
> wrote:
>
>
> Respondants are short sighted. A year hence, no one will care whether the
> client was happy
> or it was delivered on time. They will judge it on quality.

Pete Dashwood

2007-10-04, 7:55 am

"William M. Klein" <wmklein@nospam.netcom.com> wrote in message
news:x6%Mi.27132$vo5.18702@fe04.news.easynews.com...
> Choke, gasp, cringe
>
> You have GOT to be kidding!!!
>
That was a bit like my reaction when I first saw it. Are you talking about
the results or the quiz? :-)

Or Robert's comment about Quality... :-)

Pete.
> --
> Bill Klein
> wmklein <at> ix.netcom.com
> "Robert" <no@e.mail> wrote in message
> news:3li8g39823e64hddtjdkm6r6iq8oihjv0a@
4ax.com...
>
>

Pete Dashwood

2007-10-04, 7:55 am

<docdwarf@panix.com> wrote in message news:fe1ihd$hes$1@reader1.panix.com...
> In article <5miprnFdk5d6U1@mid.individual.net>,
> Pete Dashwood <dashwood@removethis.enternet.co.nz> wrote:
>
> Hmmmmm... 1290 votes before my viewing, the second-highest is 'The job is
> error-free' - interesting to consider in light of 'the only bugs in code
> are the ones you ain't found yet' - and head-and-shoulders above is 'The
> client/boss etc (sic) is happy'.

I had a feeling you'd relate, Doc :-)
>
> Seems like I'm not the only one who says that a well-functioning system,
> by definition, is one which causes the person resposible for signing the
> checks for it to smile. Maybe I should find a better circle with which to
> associated.

Not necessarily "better", but definitely like-minded it seems :-)

Pete.
--
"I used to write COBOL...now I can do anything."

Bill Gunshannon

2007-10-04, 7:55 am

In article <5mju82Fdq10aU1@mid.individual.net>,
"Pete Dashwood" <dashwood@removethis.enternet.co.nz> writes:
>
>
> <docdwarf@panix.com> wrote in message news:fe1ihd$hes$1@reader1.panix.com...
>
> I had a feeling you'd relate, Doc :-)
>
> Not necessarily "better", but definitely like-minded it seems :-)
>

On visiting the site, I read the survey questions and then just
walked away shaking my head. No, I didn't respond as I think
the whole thing was rather silly.

bill

--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>

Howard Brazee

2007-10-04, 6:55 pm

On Wed, 03 Oct 2007 21:07:55 -0500, Robert <no@e.mail> wrote:

>Respondants are short sighted. A year hence, no one will care whether the client was happy
>or it was delivered on time. They will judge it on quality.

"Quality" depends on what was needed and wanted. All those survey
items were measures of quality.

I remember reading an article about school buildings - the author was
proposing that the big, expensive showcase schools were mistakes. They
were designed to last forever - but didn't adjust well to new
demographics and technology.

A few years ago hospitals were being designed around all the new bulky
electronics that were now available. Now the monitors in the
operating room are flat, and the electronics in the patients rooms
roll in and out.

And companies with well designed, well written CoBOL batch programs
are asking to have real-time answers that they can play with on their
handheld computers. So we have to change our definition of
"quality".

(This is nothing new - people have *always* complained about the new
guys not being interested in quality. The new people have different
definitions of what quality should be).

Howard Brazee

2007-10-04, 6:55 pm

On 4 Oct 2007 12:47:42 GMT, bill@cs.uofs.edu (Bill Gunshannon) wrote:

>On visiting the site, I read the survey questions and then just
>walked away shaking my head. No, I didn't respond as I think
>the whole thing was rather silly.

Ditto.

William M. Klein

2007-10-04, 6:55 pm

commenting on "quality" being the ONLY thing remembered a year later and NOT
whether the project was delivered on time or whether the customers were happy.

It was my experience that a year later, "end users" (or customers) remember:

A) is it working today?
B) how many "fixes" have been needed since it was delivered
C) Was getting it in the first place a "good" experience (i.e. on time and what
was requested).

Now, if the definition of "quality" (in RW's note)means "is it working correctly
today" - then I suppose that is true (and first place in MY list) - but I don't
think I remember cases of "working today" AND NOT "users were happy when
delivered" AND users actually happy today (because to get from "not happy
originally" to "working today" means a lot of pain during the year in-between)

--
Bill Klein
wmklein <at> ix.netcom.com
"Pete Dashwood" <dashwood@removethis.enternet.co.nz> wrote in message
news:5mju3cFdu18oU1@mid.individual.net...
>
>
> "William M. Klein" <wmklein@nospam.netcom.com> wrote in message
> news:x6%Mi.27132$vo5.18702@fe04.news.easynews.com...
> That was a bit like my reaction when I first saw it. Are you talking about the
> results or the quiz? :-)
>
> Or Robert's comment about Quality... :-)
>
> Pete.
>
>

Robert

2007-10-04, 6:55 pm

On Thu, 4 Oct 2007 23:37:30 +1300, "Pete Dashwood" <dashwood@removethis.enternet.co.nz>
wrote:
[color=darkred]
>"William M. Klein" <wmklein@nospam.netcom.com> wrote in message
>news:x6%Mi.27132$vo5.18702@fe04.news.easynews.com...
>That was a bit like my reaction when I first saw it. Are you talking about
>the results or the quiz? :-)
>
>Or Robert's comment about Quality... :-)
>
>Pete.

Let me clarify. Deadlines are too often used as an exuse for poor quality. They say, 'We
don't have time to do it right. It's due today. Just get it done as quickly as possible.'
I respond, 'We'll get it done today AND it will be done right.'

I keep reading that a high percentage of IT projects miss their deadlines. My experience
must be atypical, because I've never worked on a project that missed its deadline. At
least, I can't recall one in the last twenty years.

Frank Swarbrick

2007-10-04, 6:55 pm

>>> On 10/3/2007 at 6:19 PM, in message
<5miprnFdk5d6U1@mid.individual.net>,
Pete Dashwood<dashwood@removethis.enternet.co.nz> wrote:
> I think many people here will have thoughts on this topic and may want to

>
> vote in the poll. Be quick, it closes soon...
>
> http://www.codeproject.com/#poll

I probably won't vote because I'm not sure which of the following they are
looking for:
1) What *I* believe.
2) What I believe my *management* believes.

Frank

HeyBub

2007-10-04, 6:55 pm

Pete Dashwood wrote:
> I think many people here will have thoughts on this topic and may
> want to vote in the poll. Be quick, it closes soon...
>
> http://www.codeproject.com/#poll
>
> (Scroll to the bottom of the page)

Why isn't "Whether you will get paid" on the list?

Alistair

2007-10-04, 6:55 pm

On 4 Oct, 18:06, Robert <n...@e.mail> wrote:
> On Thu, 4 Oct 2007 23:37:30 +1300, "Pete Dashwood" <dashw...@removethis.enternet.co.nz>
> wrote:
>
>
>
>
>
>
>
>
>
>
>
>
>
> Let me clarify. Deadlines are too often used as an exuse for poor quality. They say, 'We
> don't have time to do it right. It's due today. Just get it done as quickly as possible.'
> I respond, 'We'll get it done today AND it will be done right.'
>
> I keep reading that a high percentage of IT projects miss their deadlines. My experience
> must be atypical, because I've never worked on a project that missed its deadline. At
> least, I can't recall one in the last twenty years.- Hide quoted text -
>
> - Show quoted text -

I think Doc's phrase about 'what value of...' missed it's deadline? is
applicable. All projects that I have worked on (including the one that
I think should have been postponed) made their implementation dates.
Some bits arrived a bit later but that is a different story.

Pete Dashwood

2007-10-04, 6:55 pm

"Robert" <no@e.mail> wrote in message
news:fu6ag3l4us8vl4bg4uoqvnf0u83b6ebfug@
4ax.com...
> On Thu, 4 Oct 2007 23:37:30 +1300, "Pete Dashwood"
> <dashwood@removethis.enternet.co.nz>
> wrote:
>
>
> Let me clarify. Deadlines are too often used as an exuse for poor quality.
> They say, 'We
> don't have time to do it right. It's due today. Just get it done as
> quickly as possible.'
> I respond, 'We'll get it done today AND it will be done right.'
>
> I keep reading that a high percentage of IT projects miss their deadlines.
> My experience
> must be atypical, because I've never worked on a project that missed its
> deadline. At
> least, I can't recall one in the last twenty years.
>
I agree with your sentiments that getting quality AND getting it on time are
the goals to go for.

The only way I have ever managed to achieve this is by timeboxed, iterative,
interactive (evolutionary) development.

It CAN be done, but doing so usually requires some major re-thinking in the
development strategy. It also means establishing enough credibility that
they will let you depart from their usual approaches :-)

I wish I could share your experience of things being done on time but I
can't. Certainly, over the last 15 years I have built a good track record of
delivering good stuff on time, but it is only with timeboxed development and
a DSDM based approach that this has happened. I still see other projects
coming in late and I've had long discussions with other project managers who
are interested in the methods I use.

(There is a half written book about this sitting on my hard drive; it is one
of my back-burner projects. Hopefully, I'll finish it next year :-))

Pete.
--
"I used to write COBOL...now I can do anything."

Pete Dashwood

2007-10-04, 6:55 pm

"Bill Gunshannon" <bill@cs.uofs.edu> wrote in message
news:5mk5neFdoipaU1@mid.individual.net...
> In article <5mju82Fdq10aU1@mid.individual.net>,
> "Pete Dashwood" <dashwood@removethis.enternet.co.nz> writes:
>
> On visiting the site, I read the survey questions and then just
> walked away shaking my head. No, I didn't respond as I think
> the whole thing was rather silly.
>

Fair enough .

A tolerance for silliness is a definite prerequisite for hanging around CLC
:-)

It is very important not to take things too seriously.

It is only computer programming; not life and death... :-)

So, Bill, do you have no opinion about what matters most when getting a job
done?

Pete.
--
"I used to write COBOL...now I can do anything."

Pete Dashwood

2007-10-04, 6:55 pm

"Howard Brazee" <howard@brazee.net> wrote in message
news:v00ag35g8f9ijbhqqaed5fe9k9tbqcu0hg@
4ax.com...
> On 4 Oct 2007 12:47:42 GMT, bill@cs.uofs.edu (Bill Gunshannon) wrote:
>
>
> Ditto.

Same question as I asked Bill: Do you have no opinion on what matters most
when getting a job done, Howard?

I would agree the quiz may be frivolous, but I don't think it's silly. The
results surprised me.

Pete.
--
"I used to write COBOL...now I can do anything."

Pete Dashwood

2007-10-04, 6:55 pm

"Frank Swarbrick" <Frank.Swarbrick@efirstbank.com> wrote in message
news:4704CB91.6F0F.0085.0@efirstbank.com...
> <5miprnFdk5d6U1@mid.individual.net>,
> Pete Dashwood<dashwood@removethis.enternet.co.nz> wrote:
>
>
> I probably won't vote because I'm not sure which of the following they are
> looking for:
> 1) What *I* believe.
> 2) What I believe my *management* believes.
>

The Code Project is an excellent group of developers in various languages.
They run these polls every month (I thought it was w

ly, but I see I was
wrong about that). I don't think they take them too seriously and sometimes
(as in this case) the results can be surprising. There are over half a
million people in this group, working at the coal face in many and varied
industries. What they report through these polls is usually very
interesting. Typically there will be around 4000 respondents, so it is a
reasonable sample size.

In terms of your dilemma above Frank, they aren't "looking for" anything. It
is really just a bit of light relief from the trials of cutting code. (I'm
sure that many here can relate to that :-))

As you can only vote once (I tried doing it twice and they were onto me :-))
you can elect either of the scenarios you mentioned above, but not both of
them :-)

Finally, ANYONE can suggest a poll and they will probably run it. If people
here think this one is "silly" why not suggest a better one?

Pete.
--
"I used to write COBOL...now I can do anything."

Pete Dashwood

2007-10-04, 6:55 pm

"HeyBub" <heybubNOSPAM@gmail.com> wrote in message
news:13gaf0ai5cjh67f@news.supernews.com...
> Pete Dashwood wrote:
>
> Why isn't "Whether you will get paid" on the list?
>
Probably implicit in "Is your Boss/Client happy with the result"... :-)

Pete.
--
"I used to write COBOL...now I can do anything."

Bill Gunshannon

2007-10-04, 6:55 pm

In article <5ml7jdFe5sajU1@mid.individual.net>,
"Pete Dashwood" <dashwood@removethis.enternet.co.nz> writes:
>
>
> "Howard Brazee" <howard@brazee.net> wrote in message
> news:v00ag35g8f9ijbhqqaed5fe9k9tbqcu0hg@
4ax.com...
>
> Same question as I asked Bill:

Somehow I missed where you asked me that, I guess.

> Do you have no opinion on what matters most
> when getting a job done, Howard?

Of course I have an opinion but that survey was more along the lines of:
"Answer Yes or NO, have you stopped beating your wife?"

>
> I would agree the quiz may be frivolous, but I don't think it's silly. The
> results surprised me.

The results from a meaningless question where other answers are just
as possible is equally meaningless. Given the apparent desire to keep
it simple a better way would have been to have people rank those options
as to their importance. And maybe included a place to write in something
else that the voter considered of greater importance. And then, use all
that input to create a more meaningful survey.

But that's just my opinion. :-)

bill

--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>

Bill Gunshannon

2007-10-04, 6:55 pm

In article <5ml7emFdus5lU1@mid.individual.net>,
"Pete Dashwood" <dashwood@removethis.enternet.co.nz> writes:
>
>
> "Bill Gunshannon" <bill@cs.uofs.edu> wrote in message
> news:5mk5neFdoipaU1@mid.individual.net...
>
> Fair enough .
>
> A tolerance for silliness is a definite prerequisite for hanging around CLC
>:-)
>
> It is very important not to take things too seriously.
>
> It is only computer programming; not life and death... :-)
>
> So, Bill, do you have no opinion about what matters most when getting a job
> done?
>

Oh, here's the one where you asked me. Guess I just saw them out of
order. In any event, I hope I answered to your satisfaction. :-)

bill

--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bill@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>

Frank Swarbrick

2007-10-04, 6:55 pm

>>> On 10/4/2007 at 4:36 PM, in message
<5ml872Fe337qU1@mid.individual.net>,
Pete Dashwood<dashwood@removethis.enternet.co.nz> wrote:

>
> "Frank Swarbrick" <Frank.Swarbrick@efirstbank.com> wrote in message
> news:4704CB91.6F0F.0085.0@efirstbank.com...
> to
> are
>
> The Code Project is an excellent group of developers in various
> languages.
> They run these polls every month (I thought it was w

ly, but I see I
> was
> wrong about that). I don't think they take them too seriously and
> sometimes
> (as in this case) the results can be surprising. There are over half a
> million people in this group, working at the coal face in many and
> varied
> industries. What they report through these polls is usually very
> interesting. Typically there will be around 4000 respondents, so it is a
>
> reasonable sample size.
>
> In terms of your dilemma above Frank, they aren't "looking for"
> anything. It
> is really just a bit of light relief from the trials of cutting code.
> (I'm
> sure that many here can relate to that :-))
>
> As you can only vote once (I tried doing it twice and they were onto me
> :-))
> you can elect either of the scenarios you mentioned above, but not both
> of
> them :-)
>
> Finally, ANYONE can suggest a poll and they will probably run it. If
> people
> here think this one is "silly" why not suggest a better one?

I don't think it was silly, necessarily, just not very well worded.

Frank

Pete Dashwood

2007-10-04, 9:55 pm

"Bill Gunshannon" <bill@cs.uofs.edu> wrote in message
news:5ml8ufFdu0njU1@mid.individual.net...
> In article <5ml7jdFe5sajU1@mid.individual.net>,
> "Pete Dashwood" <dashwood@removethis.enternet.co.nz> writes:
>
> Somehow I missed where you asked me that, I guess.
>
>
> Of course I have an opinion but that survey was more along the lines of:
> "Answer Yes or NO, have you stopped beating your wife?"
>

That would be YES or NO, on a scale of 1 to 5 :-)

>
> The results from a meaningless question where other answers are just
> as possible is equally meaningless. Given the apparent desire to keep
> it simple a better way would have been to have people rank those options
> as to their importance.

That is exactly what it did. Rank them from 1 through 5 with a description
of what each ranking meant and how strong it was.

> And maybe included a place to write in something
> else that the voter considered of greater importance. And then, use all
> that input to create a more meaningful survey.
>
That would be a nice touch but it is unlikely anyone would have time or
inclination to go through 4,000 results reading the comments...

Pete.
--
"I used to write COBOL...now I can do anything."

Pete Dashwood

2007-10-04, 9:55 pm

"Bill Gunshannon" <bill@cs.uofs.edu> wrote in message
news:5ml91hFdu0njU2@mid.individual.net...
> In article <5ml7emFdus5lU1@mid.individual.net>,
> "Pete Dashwood" <dashwood@removethis.enternet.co.nz> writes:
>
> Oh, here's the one where you asked me. Guess I just saw them out of
> order. In any event, I hope I answered to your satisfaction. :-)
>
Yes, thanks. While I appreciate you responding, I would have liked it more
if you had actually looked at the quiz and read the instructions :-)

Never mind... next time :-)

Pete.
--
"I used to write COBOL...now I can do anything."

donald tees

2007-10-04, 9:55 pm

Pete Dashwood wrote:
> "Howard Brazee" <howard@brazee.net> wrote in message
> news:v00ag35g8f9ijbhqqaed5fe9k9tbqcu0hg@
4ax.com...
>
> Same question as I asked Bill: Do you have no opinion on what matters most
> when getting a job done, Howard?
>
> I would agree the quiz may be frivolous, but I don't think it's silly. The
> results surprised me.
>
> Pete.

Getting paid?

Donald

donald tees

2007-10-04, 9:55 pm

HeyBub wrote:
> Pete Dashwood wrote:
>
> Why isn't "Whether you will get paid" on the list?
>
>
I agree. I suspect that 99% of all jobs would not be finished if the
offer of payment were withdrawn.

Donald

donald tees

2007-10-04, 9:55 pm

Pete Dashwood wrote:
> "HeyBub" <heybubNOSPAM@gmail.com> wrote in message
> news:13gaf0ai5cjh67f@news.supernews.com...
> Probably implicit in "Is your Boss/Client happy with the result"... :-)
>
> Pete.

The two are not always synonymous. Would you work for somebody that
could not afford to pay you, but expressed great delight in your work?

Donald

2007-10-04, 9:55 pm

In article <5ml71qFdmf1rU1@mid.individual.net>,
Pete Dashwood <dashwood@removethis.enternet.co.nz> wrote:

[snip]

>(There is a half written book about this sitting on my hard drive; it is one
>of my back-burner projects. Hopefully, I'll finish it next year :-))

Oh, I *cannot* resist...

.... next year's fine, as long as it meets its milestones.

DD

Pete Dashwood

2007-10-05, 3:55 am

"donald tees" <donaldtees@execulink.com> wrote in message
news:8_adncO7krDiD5janZ2dnUVZ_uninZ2d@go
lden.net...
> Pete Dashwood wrote:
>
> The two are not always synonymous. Would you work for somebody that could
> not afford to pay you, but expressed great delight in your work?
>
I have done.. :-)

What price can you put on friendship? :-) (And a little charity now and then
does no harm..what goes around comes around)

Having said that, it would not be my preference, no... :-)

Pete.

Robert

2007-10-05, 3:55 am

On Thu, 04 Oct 2007 16:13:46 GMT, "William M. Klein" <wmklein@nospam.netcom.com> wrote:

>commenting on "quality" being the ONLY thing remembered a year later and NOT
>whether the project was delivered on time or whether the customers were happy.
>
>It was my experience that a year later, "end users" (or customers) remember:
>
>A) is it working today?
>B) how many "fixes" have been needed since it was delivered
>C) Was getting it in the first place a "good" experience (i.e. on time and what
>was requested).
>
>
>Now, if the definition of "quality" (in RW's note)means "is it working correctly
>today" - then I suppose that is true (and first place in MY list) - but I don't
>think I remember cases of "working today" AND NOT "users were happy when
>delivered" AND users actually happy today (because to get from "not happy
>originally" to "working today" means a lot of pain during the year in-between)

It's like buying a car. Users can't tell what's under the hood, all they know is the
sticker price. They find out about quality when the odometer reads 100,000, nothing has
gone wrong and they're still on the original brakes.

Howard Brazee

2007-10-05, 6:55 pm

On Fri, 05 Oct 2007 00:27:39 -0500, Robert <no@e.mail> wrote:

>It's like buying a car. Users can't tell what's under the hood, all they know is the
>sticker price. They find out about quality when the odometer reads 100,000, nothing has
>gone wrong and they're still on the original brakes.

But does that make a superior car to the one next to it that replaced
its brakes at 90,000 miles?

Howard Brazee

2007-10-05, 6:55 pm

On Fri, 5 Oct 2007 11:16:24 +1300, "Pete Dashwood"
<dashwood@removethis.enternet.co.nz> wrote:

>I agree with your sentiments that getting quality AND getting it on time are
>the goals to go for.
>
>The only way I have ever managed to achieve this is by timeboxed, iterative,
>interactive (evolutionary) development.

Interactive (evolutionary) development is often the problem, as it
tends to be accompanied by interactive (evolutionary) requirements.

Howard Brazee

2007-10-05, 6:55 pm

On Fri, 5 Oct 2007 11:25:47 +1300, "Pete Dashwood"
<dashwood@removethis.enternet.co.nz> wrote:

>Same question as I asked Bill: Do you have no opinion on what matters most
>when getting a job done, Howard?
>
>I would agree the quiz may be frivolous, but I don't think it's silly. The
>results surprised me.

My answers to that poll are "it depends".

Michael Mattias

2007-10-05, 6:55 pm

"Howard Brazee" <howard@brazee.net> wrote in message
news:vcjcg3hj274j402h99fjgs2vs23o24i5i9@
4ax.com...
> On Fri, 05 Oct 2007 00:27:39 -0500, Robert <no@e.mail> wrote:
>
>
> But does that make a superior car to the one next to it that replaced
> its brakes at 90,000 miles?

Well, no. Yes. Maybe.

Regardless, just as in auto manufacturing it takes GENUINE SKILL to ensure
software bugs don't show up until the warranty has expired and the repair
becomes billable.

MCM

2007-10-05, 6:55 pm

In article <vcjcg3hj274j402h99fjgs2vs23o24i5i9@4ax.com>,
Howard Brazee <howard@brazee.net> wrote:
>On Fri, 05 Oct 2007 00:27:39 -0500, Robert <no@e.mail> wrote:
>
>
>But does that make a superior car to the one next to it that replaced
>its brakes at 90,000 miles?

I would say that depends, Mr Brazee, on the criteria used to evaluate
superiority. Given that a quality is often assessed in the context of a
task ('It is a good refrigerator for keeping things cold but since it is
small it is not a good refrigerator for holding more than (amount) of
food') the aspect of 'superior for... ' might need to be considered.

'The car's crash-worthiness is more like crash-worthlessness... but hey,
the brakes last 100K miles!'

DD

Howard Brazee

2007-10-05, 6:55 pm

On Fri, 5 Oct 2007 09:55:44 -0500, "Michael Mattias"
<mmattias@talsystems.com> wrote:

>
>Well, no. Yes. Maybe.
>
>Regardless, just as in auto manufacturing it takes GENUINE SKILL to ensure
>software bugs don't show up until the warranty has expired and the repair
>becomes billable.

Consumers often call this "planned obsolescence"

Howard Brazee

2007-10-05, 6:55 pm

On Fri, 5 Oct 2007 17:30:46 +0000 (UTC), docdwarf@panix.com () wrote:

>
>I would say that depends, Mr Brazee, on the criteria used to evaluate
>superiority. Given that a quality is often assessed in the context of a
>task ('It is a good refrigerator for keeping things cold but since it is
>small it is not a good refrigerator for holding more than (amount) of
>food') the aspect of 'superior for... ' might need to be considered.
>
>'The car's crash-worthiness is more like crash-worthlessness... but hey,
>the brakes last 100K miles!'

I like the "solution" to the problem of bad brakes being a louder
horn.

Pete Dashwood

2007-10-05, 6:55 pm

"Howard Brazee" <howard@brazee.net> wrote in message
news:2gjcg39g5f3gpfj50cdd32t6b4jaf0r64k@
4ax.com...
> On Fri, 5 Oct 2007 11:16:24 +1300, "Pete Dashwood"
> <dashwood@removethis.enternet.co.nz> wrote:
>
>
> Interactive (evolutionary) development is often the problem, as it
> tends to be accompanied by interactive (evolutionary) requirements.

That is exactly the point Howard, and I've covered it here so often.

Evolutionary requirements are exactly what makes a system responsive and
flexible. There is NO problem with continuously changing requirements
PROVIDED JAD and timeboxing are utilised, and the "customer" KNOWS that
changing requirements MAY mean some functionality will not be delivered. It
is my experience that when this is clear, and the final priorities are down
to the customer, not IT, everything is fine. When IT works WITH peole
(instead of "us" and "them") it becmes a true joint venture and all
concerned realise that quarts don't fit into pint pots. What is delievered
will be what is most important to the people using it, (even though
priorities and requirements change in the real world) and that is good for
IT and the Business.

The fundamental concepts are:

1. Don't build everything right away. You are going to change it anyway, so
build what is perceived as most important NOW. (Use Cases decide the
functionality, Object modelling decides the components, the Business decides
the priorities.)

2. Timebox development and iterations.

3. Use JAD workshops to review what is being developed, to prototype user
interfaces in the context of the Use Case, and to re-prioritise if
necessary.

Amazing results in terms of useful and timely functionality delivered, and
satisfaction in both IT and the Business can be achieved with this approach.

Pete.
--
"I used to write COBOL...now I can do anything."

Tom

2007-10-06, 3:55 am

On Oct 3, 8:19 pm, "Pete Dashwood"
<dashw...@removethis.enternet.co.nz> wrote:
> I think many people here will have thoughts on this topic and may want to
> vote in the poll. Be quick, it closes soon...
>
> http://www.codeproject.com/#poll
>
> (Scroll to the bottom of the page)
>
> Pete.
> --
> "I used to write COBOL...now I can do anything."

What SHOULD matter the most was did the project meet its objectives.
Was the business able to function the way they thought would when the
envisioned the project. In reality, people seem to lose sight of
this, and concentrate most on whether the project is implemented on
time. The error free aspect is usually sacrificed for the sake of
timeliness.

Howard Brazee

2007-10-08, 6:55 pm

On Sat, 6 Oct 2007 11:33:18 +1300, "Pete Dashwood"
<dashwood@removethis.enternet.co.nz> wrote:

>1. Don't build everything right away. You are going to change it anyway, so
>build what is perceived as most important NOW. (Use Cases decide the
>functionality, Object modelling decides the components, the Business decides
>the priorities.)
>
>2. Timebox development and iterations.

Evolutionary requirements are very useful - as long as those making
them have the authority to sign off on evolutionary costs - and as
long as the big wigs don't look at the final cost and define the
project as a failure.

We keep hearing about tremendous failures in our industry. These are
often because huge projects have too much inertia to turn around
quickly to adjust to evolving requirements. On the other hand, it's
easy to plan on a couple of extra days for a 40 hour project need to
adjust to a different requirement. (Those days might take a couple
of hours of programming - but requirements changes often need
sign-offs).

Howard Brazee

2007-10-08, 6:55 pm

On Fri, 05 Oct 2007 20:35:45 -0700, Tom <thasselb@gmail.com> wrote:

>What SHOULD matter the most was did the project meet its objectives.
>Was the business able to function the way they thought would when the
>envisioned the project. In reality, people seem to lose sight of
>this, and concentrate most on whether the project is implemented on
>time. The error free aspect is usually sacrificed for the sake of
>timeliness.

And "error free" is often designed by "does it do what the
requirements asked", as opposed to your stated goal above.

Charles Hottel

2007-10-09, 3:56 am

"Pete Dashwood" <dashwood@removethis.enternet.co.nz> wrote in message
news:5ml71qFdmf1rU1@mid.individual.net...
>
>
> "Robert" <no@e.mail> wrote in message
> news:fu6ag3l4us8vl4bg4uoqvnf0u83b6ebfug@
4ax.com...
> I agree with your sentiments that getting quality AND getting it on time
> are the goals to go for.
>
> The only way I have ever managed to achieve this is by timeboxed,
> iterative, interactive (evolutionary) development.
>
> It CAN be done, but doing so usually requires some major re-thinking in
> the development strategy. It also means establishing enough credibility
> that they will let you depart from their usual approaches :-)
>
> I wish I could share your experience of things being done on time but I
> can't. Certainly, over the last 15 years I have built a good track record
> of delivering good stuff on time, but it is only with timeboxed
> development and a DSDM based approach that this has happened. I still see
> other projects coming in late and I've had long discussions with other
> project managers who are interested in the methods I use.
>
> (There is a half written book about this sitting on my hard drive; it is
> one of my back-burner projects. Hopefully, I'll finish it next year :-))
>
> Pete.
> --
> "I used to write COBOL...now I can do anything."
>
>
>

Now I see that it is you, Pete Dashwood, that is responsible for the demise
of COBOL and the COBOL community! All along you have been ahead of the curve
on relational , OO, and systems development methods. You could have saved us
all many times from our ignorance with a couple timely and easily written
books, but no, you left them half-written on yout hard drive:-)

Pete Dashwood

2007-10-09, 6:55 pm

"Charles Hottel" <chottel@earthlink.net> wrote in message
news:13gltchrkh3g203@corp.supernews.com...
>
> "Pete Dashwood" <dashwood@removethis.enternet.co.nz> wrote in message
> news:5ml71qFdmf1rU1@mid.individual.net...
>
> Now I see that it is you, Pete Dashwood, that is responsible for the
> demise of COBOL and the COBOL community! All along you have been ahead of
> the curve on relational , OO, and systems development methods. You could
> have saved us all many times from our ignorance with a couple timely and
> easily written books, but no, you left them half-written on yout hard
> drive:-)

Ah, Charlie, I have never been one to sow seed on stony ground... :-)

Besides, the total amount I have contributed here would be equivalent to
several books.

I was going to write a book on OO COBOL (way back when I first got the hang
of it) but Wilson Price very kindly gave me a copy of his, and it was so
good I decided I had nothing to add.

I tried to persuade the local Polytechnic to pick it up as a COBOL text, and
almost succeeded, but in the end they decided to drop COBOL.

The Book that is currently on my hard drive will get written (I had another
look at it and it is nearer 70% :-)) because it addresses a wider audience
than just the COBOL community. It is about Project Management rather than
programming, and has some quite radical ideas which I have found work very
well in practice. When I was first approached to write it, I was kind of
reluctant (too busy) and they wouldn't pay any advance, so I said, "OK, it
gets done when it gets done..." I have writing contacts who have POD (Print
on Demand) publishing facilities and I'll probably hand it to them when its
done.

I haven't done any writing for ages and I need to get back to it.

Pete.
--
"I used to write COBOL...now I can do anything."

>

Sponsored Links