Happy Birthday Extreme Programming! 
Author Message
 Happy Birthday Extreme Programming!

It was on this day 5 years ago that Kent Beck walked into a conference room
in an old Chrysler building and asked us to do the rain storm team building
thing.  It was a bit of a west coast kind of thing to do, but we all did it
anyway.  Then Kent began to explain to us how he wanted this project to be
different than other projects.  We had started the rewrite of the rewrite of
the Chrysler payroll system in Smalltalk.

The project was supposed to have a great impact on the software world.  It
was supposed to prove that object oriented languages like Smalltalk could be
used for applications previously thought of as Cobol on a main frame only.
In
retrospect, of course, we didn't show that at all.  That would be for some
other project to prove.  What we did show instead was that software projects
didn't have to have enormous and complex methodologies to be productive.

Don Wells

P.S. Go to the XP Universe conference!  www.xpuniverse.com



Sun, 24 Aug 2003 08:26:32 GMT  
 Happy Birthday Extreme Programming!
I'd be interested in hearing more of the history of C2.  I did read XP
explained, but I'm curious why you say "we didn't show that at all."

Quote:

> It was on this day 5 years ago that Kent Beck walked into a conference room
> in an old Chrysler building and asked us to do the rain storm team building
> thing.  It was a bit of a west coast kind of thing to do, but we all did it
> anyway.  Then Kent began to explain to us how he wanted this project to be
> different than other projects.  We had started the rewrite of the rewrite of
> the Chrysler payroll system in Smalltalk.

> The project was supposed to have a great impact on the software world.  It
> was supposed to prove that object oriented languages like Smalltalk could be
> used for applications previously thought of as Cobol on a main frame only.
> In
> retrospect, of course, we didn't show that at all.  That would be for some
> other project to prove.  What we did show instead was that software projects
> didn't have to have enormous and complex methodologies to be productive.

> Don Wells

> P.S. Go to the XP Universe conference!  www.xpuniverse.com

--

765-496-1185
"In matters of style, swim with the current;
in matters of principle, stand like a rock."
- Thomas Jefferson


Sun, 24 Aug 2003 23:37:37 GMT  
 Happy Birthday Extreme Programming!
Ya'know comp.software.extreme-programming doesn't get enough on-topic
posts. Why dontcha post there?
=====


Quote:

> I'd be interested in hearing more of the history of C2.  I did read XP
> explained, but I'm curious why you say "we didn't show that at all."


> > It was on this day 5 years ago that Kent Beck walked into a conference
> > room
> > in an old Chrysler building and asked us to do the rain storm team
> > building
> > thing.  It was a bit of a west coast kind of thing to do, but we all
> > did it
> > anyway.  Then Kent began to explain to us how he wanted this project to
> > be
> > different than other projects.  We had started the rewrite of the
> > rewrite of
> > the Chrysler payroll system in Smalltalk.

> > The project was supposed to have a great impact on the software world.  
> > It
> > was supposed to prove that object oriented languages like Smalltalk
> > could be
> > used for applications previously thought of as Cobol on a main frame
> > only.
> > In
> > retrospect, of course, we didn't show that at all.  That would be for
> > some
> > other project to prove.  What we did show instead was that software
> > projects
> > didn't have to have enormous and complex methodologies to be
> > productive.

> > Don Wells

> > P.S. Go to the XP Universe conference!  www.xpuniverse.com

> --

> 765-496-1185
> "In matters of style, swim with the current;
> in matters of principle, stand like a rock."
> - Thomas Jefferson

--
       <http://homepage.mac.com/keithray/resume.html>


Tue, 26 Aug 2003 00:21:28 GMT  
 Happy Birthday Extreme Programming!

Quote:

> It was on this day 5 years ago that Kent Beck walked into a conference room
> in an old Chrysler building and asked us to do the rain storm team building
> thing....

...and made an attempt to move 20 years backward good practices of software ENGINERING
(yes, ENGINEERING) by promoting hacking and anarchy to the range of METHODOLOGY.

A.L.



Thu, 28 Aug 2003 06:47:24 GMT  
 Happy Birthday Extreme Programming!

Andrzej Lewandowski schrieb:

Quote:

> > It was on this day 5 years ago that Kent Beck walked into a conference room
> > in an old Chrysler building and asked us to do the rain storm team building
> > thing....

> ...and made an attempt to move 20 years backward good practices of software ENGINERING
> (yes, ENGINEERING) by promoting hacking and anarchy to the range of METHODOLOGY.

> A.L.

Maybe you have missed it,but one of XP's rules says: "refactor mercilessly" . If that's what you are doing

during a hack, everything is fine.

All of us who do software engineering in the real world know that 20 years of good practices did not solve
the old problem, that more software projects fail in some sense than do succeed.

I feel it's about time for a little bit of anarchy !

regards
G.

P.S.:
These people back in 1776 who brought us democracy, first were regarded as anarchists aren't they ?

  Gerald.Zincke.vcf
< 1K Download


Thu, 28 Aug 2003 06:29:29 GMT  
 Happy Birthday Extreme Programming!
On Sat, 10 Mar 2001 16:47:24 -0600, "Andrzej Lewandowski"

Quote:


>> It was on this day 5 years ago that Kent Beck walked into a conference room
>> in an old Chrysler building and asked us to do the rain storm team building
>> thing....

>...and made an attempt to move 20 years backward good practices of software ENGINERING
>(yes, ENGINEERING) by promoting hacking and anarchy to the range of METHODOLOGY.

The practices of XP are grounded in the hallowed methodologies of
which you speak, but represent a subset of them that are performed all
the time, i.e. constant testing, refactoring, integration all in
extremely short cycles.  I don't see anything anarchistic about it.

Rob



Thu, 28 Aug 2003 09:51:02 GMT  
 Happy Birthday Extreme Programming!

Quote:



> > It was on this day 5 years ago that Kent Beck walked into a conference
> > room
> > in an old Chrysler building and asked us to do the rain storm team
> > building
> > thing....

> ...and made an attempt to move 20 years backward good practices of
> software ENGINERING
> (yes, ENGINEERING) by promoting hacking and anarchy to the range of
> METHODOLOGY.

> A.L.

I've seen one person reject XP because it is a collection of known best
practices that have been proven successful over the last 20 years... she
rejected it because there wasn't anything "new" about it.

Maybe you thinking XP is "hacking" because of its name.

Would "a lightweight implementation of RUP called dX" offend you as much
as "Extreme Programming" does?

Check out <http://www.objectmentor.com/publications/RUPvsXP.pdf>

--
       <http://homepage.mac.com/keithray/resume.html>



Thu, 28 Aug 2003 12:37:36 GMT  
 Happy Birthday Extreme Programming!

Quote:

>I'd be interested in hearing more of the history of C2.  I did read XP
>explained, but I'm curious why you say "we didn't show that at all."

The original idea was to build an enterprise payroll system using Smalltalk
instead of Cobol.  It was supposed to have fancy bells and whistles.
Instead it got mired down in interfacing with legacy systems, then politics
became a problem, then soon the project was canceled.  It is currently used
to pay 5 people.  All the other employees were moved back into the Cobol
system.  Soon it will not be used at all.

The project did not have to fail of course.  I think everyone here knows
that, but it did, so we did not prove what we started out to prove:  the
viability of Smalltalk for large MIS systems.

Don Wells
----
Go to the XP Universe conference!  www.xpuniverse.com



Thu, 04 Sep 2003 13:14:00 GMT  
 Happy Birthday Extreme Programming!
I hope this isn't just spin to smooth over the failure of *the*
seminal XP project.  

Would be interesting to see the survival and longevity rate of XP
projects over the years.  Has anyone done a study on this?

Bill Cole
JumpStart Systems

On Sun, 18 Mar 2001 00:14:00 -0500, "Don Wells"

Quote:


>>I'd be interested in hearing more of the history of C2.  I did read XP
>>explained, but I'm curious why you say "we didn't show that at all."

>The original idea was to build an enterprise payroll system using Smalltalk
>instead of Cobol.  It was supposed to have fancy bells and whistles.
>Instead it got mired down in interfacing with legacy systems, then politics
>became a problem, then soon the project was canceled.  It is currently used
>to pay 5 people.  All the other employees were moved back into the Cobol
>system.  Soon it will not be used at all.

>The project did not have to fail of course.  I think everyone here knows
>that, but it did, so we did not prove what we started out to prove:  the
>viability of Smalltalk for large MIS systems.

>Don Wells
>----
>Go to the XP Universe conference!  www.xpuniverse.com



Thu, 04 Sep 2003 23:44:29 GMT  
 Happy Birthday Extreme Programming!
G'day,

Here at ThoughtWorks we've had a development project going for almost 3
years.  It's been XP since Jan 2000.  The total team has 75 working in
Chicago and 8 here in Melbourne (Australia, not Florida - that's right,
internationally remote XP!).  The development iterations started out at 4
week per cycles and this has been wound back to  2 weeks per cycle.  It's
not pure XP in that it's far above the recommended 10-20 person team, it's
running a split site development and it's been going for a long time (but
been making regular deliveries) and we use customer proxies rather than real
customers.

Good things:
   Martin Fowler, Chet Hendrickson and Ward Cunningham are on the team.
    Team and customer m{*filter*}is very high - we constantly get good feed back
from the customer.
    The quality of the delivered software is visibly improving.
    More of the developers know more about the system - a function of
swaping roles & responsiblities each iteration.

Bad things:
    If you're doing remote development (XP or any sort) get a highspeed
line.  56k modems just don't do it.
    It's Java, not Smalltalk.
    The swap to XP does involve a big mind swap.  It's hard for developers
and analysts to get into the "use the simplest solution" mindset - there is
a reluctance to throw away code/design and to code for today, not for 6
months away

Things to watch:
    When moving from traditional development to XP you probably (pr > 99%)
won't have enough test cases.  Get these written early.
    Be prepared to throw away (refactor out of existence) lots of code.

Ross Bradley
ThoughtWorks


Quote:
> I hope this isn't just spin to smooth over the failure of *the*
> seminal XP project.

> Would be interesting to see the survival and longevity rate of XP
> projects over the years.  Has anyone done a study on this?

> Bill Cole
> JumpStart Systems

> On Sun, 18 Mar 2001 00:14:00 -0500, "Don Wells"


> >>I'd be interested in hearing more of the history of C2.  I did read XP
> >>explained, but I'm curious why you say "we didn't show that at all."

> >The original idea was to build an enterprise payroll system using
Smalltalk
> >instead of Cobol.  It was supposed to have fancy bells and whistles.
> >Instead it got mired down in interfacing with legacy systems, then
politics
> >became a problem, then soon the project was canceled.  It is currently
used
> >to pay 5 people.  All the other employees were moved back into the Cobol
> >system.  Soon it will not be used at all.

> >The project did not have to fail of course.  I think everyone here knows
> >that, but it did, so we did not prove what we started out to prove:  the
> >viability of Smalltalk for large MIS systems.

> >Don Wells
> >----
> >Go to the XP Universe conference!  www.xpuniverse.com



Fri, 05 Sep 2003 13:26:14 GMT  
 Happy Birthday Extreme Programming!

Quote:

> Bad things:
>     If you're doing remote development (XP or any sort) get a highspeed
> line.  56k modems just don't do it.
>     It's Java, not Smalltalk.
>     The swap to XP does involve a big mind swap.  It's hard for developers
> and analysts to get into the "use the simplest solution" mindset - there
is
> a reluctance to throw away code/design and to code for today, not for 6
> months away

How do you see the obvious negative points interacting ?  Did the use of
Java help, or exacerbate, or not affect, the split continent (incontinent ?)
development ?

    -- chris



Sat, 06 Sep 2003 07:08:29 GMT  
 Happy Birthday Extreme Programming!
Don,

Quote:
>> the viability of Smalltalk for large MIS systems.

don't wory - that was proven by others (not only by us)
we got a whole family of ST systems in that arena ..

Have a look at  http://www.agens.com/webnew/phoenix.nsf/pages/start.htm
to get an impression. In the meantime we've got two life insurance
systems running in ST plus a commission system on the way to production.

There are two problems we had with that ..
* the high performance parts on the Host side were later rewritten in
   C and C++ - so Smalltalk Batch did not really prove effective
* for the internet stuff we had to switch to a Java Servlet / HTML client
   architecture

But if something has to be built really fast and a fat client does we still
use smalltalk

Cheers

Wolfgang

--------------------------------------------------------------

http://www.objectarchitects.de/ObjectArchitects/



Sat, 06 Sep 2003 15:46:23 GMT  
 Happy Birthday Extreme Programming!

Quote:

> How do you see the obvious negative points interacting ?  Did the use of
> Java help, or exacerbate, or not affect, the split continent (incontinent ?)
> development ?

>     -- chris

I have a hard time seeing how Java helped with anything. (of course,
this is a Smalltalk newsgroup :-)

Nevin



Sat, 06 Sep 2003 22:52:52 GMT  
 Happy Birthday Extreme Programming!

Quote:

> There are two problems we had with that ..
> * the high performance parts on the Host side were later rewritten in
>    C and C++ - so Smalltalk Batch did not really prove effective

Sounds like a Smalltalk batch approach was tried and discarded?  Do you
have comparative info? (speed of one vs. the other, info on exactly what
it was trying to do, etc.)

Quote:
> * for the internet stuff we had to switch to a Java Servlet / HTML client
>    architecture

You switched away from what architecture?  And why?

I'm not criticising-- I'm just trying to better understand the issues
you faced.  And I appreciate your input.

Nevin Pratt



Sat, 06 Sep 2003 22:57:57 GMT  
 Happy Birthday Extreme Programming!
G'day,

To some extent the use of Java may have helped the distributed development.
The fact that it's essentially based on lots of small files rather than a
monolithic image can mean that distributed development is easier, and the
inability to modify the behaviour of base system classes can mean a more
stable base to build on; however I believe that these issues can be
addressed by a good code repository and reasonable partitioning of
development (so that developers don't clash too often).

I used VW & Envy for about 4 years and although it was great for local
development it's remote features weren't easy to use (this experience is a
couple of years old now, so please correct me if the world has changed since
then).  From what I remember, to do remote development using Envy (we had
development occuring in Zurich and London) the repository was cloned,
development went on at both sites for a while and then the repositories were
merged.  I seem to remember the merge process as being a "challenge".  I
think that an XP style development with constant integration would remove
the "merge hell" phase, but, as I said in the original post,  for
distributed development the network must be fast.

When I listen to the other developers here and discuss the problems they are
trying to solve, I realize how mature Smalltalk is.  It seems that many of
the problems that Java is addressing now have been solved in Smalltalk
years, possibly decades ago.



Quote:

> > Bad things:
> >     If you're doing remote development (XP or any sort) get a highspeed
> > line.  56k modems just don't do it.
> >     It's Java, not Smalltalk.
> >     The swap to XP does involve a big mind swap.  It's hard for
developers
> > and analysts to get into the "use the simplest solution" mindset - there
> is
> > a reluctance to throw away code/design and to code for today, not for 6
> > months away

> How do you see the obvious negative points interacting ?  Did the use of
> Java help, or exacerbate, or not affect, the split continent (incontinent
?)
> development ?

>     -- chris



Sun, 07 Sep 2003 07:03:31 GMT  
 
 [ 18 post ]  Go to page: [1] [2]

 Relevant Pages 

1. TASUG - Extreme Programming - Experience an Extreme Hour

2. Happy 20th Birthday, Smalltalk-80!

3. Happy Birthday disk drive!

4. Happy Birthday, Arnor

5. Happy Birthday & Mail Problem

6. FW: Happy Birthday from Madison South Dakota USA

7. Happy birthday from Siberia !

8. Happy Birthday

9. Happy Birthday Song in Clipper

10. Happy birthday, FORTRAN!

11. Happy Birthday - Rexx

12. Happy Birthday, Ruby, and an announcement....

 

 
Powered by phpBB® Forum Software