Decoder ring

[David Pollak <fe...@gmail.com>]

Folks,

Just to make sure that everybody has the DPP decoder ring, my comments and
this list and on my blog should be read to mean:

   - "maybe not for long" -- I expect that my harsh comments may result in
   me getting booted from the ASF or ESME project.  I do not intend to leave
   unless asked to and I do feel very passionately about seeing ESME succeed.
   - I think Anne and Dick are just the right people to head up ESME.  I
   cannot think of a better person or team to lead the project.  While I think
   they have things to learn, I think that armed with the learning, they are
   going to rock... and I think they already rock!
   - My high regard for Anne and Dick does not take away from my high regard
   for the other participants in ESME, past and present.  I would name names,
   but then I'd forget someone and that might be considered a negative.  So,
   let me just say that everyone who has committed to this project has moved it
   forward.
   - I believe in burning down code.  Mesa for NextStep was my first
   spreadsheet, code started in 1991.  In the entire time that I ran the Mesa
   for NextStep project, there were 3 "crash" defects and 1 "incorrect
   recalculation" defect reported among our 50K+ user base (there were other
   bugs, but data loss and incorrect recalc bugs were top priority).  For a
   bunch of years in the 90s, every Chrysler lease and loan was done through
   Mesa, billions of dollars were traded through Mesa on trading floors every
   day, national wireless companies did their financials using Mesa, etc.  The
   code base exists (more than 50% of the LoC is the same) today as an OS X
   product.  I was so disappointed with the Mesa codebase that when we started
   the OS/2 project, I made sure that none of the code came over.  So, my
   expectations and commentary about my code bases do not reflect the common
   perception about code bases.

Thanks,

David

-- 
Lift, the simply functional web framework http://liftweb.net
Beginning Scala http://www.apress.com/book/view/1430219890
Follow me: http://twitter.com/dpp
Git some: http://github.com/dpp

Re: Decoder ring

[Daniel Kulp <dk...@apache.org>]

Wow..    What a set of threads/discussions to come back from vacation to see.    
:-)


On Sun May 3 2009 2:32:15 pm David Pollak wrote:
> Folks,
>
> Just to make sure that everybody has the DPP decoder ring, my comments and
> this list and on my blog should be read to mean:
>
>    - "maybe not for long" -- I expect that my harsh comments may result in
>    me getting booted from the ASF or ESME project.  I do not intend to
> leave unless asked to and I do feel very passionately about seeing ESME
> succeed.

I don't think you should need to worry about this at all.    You've earned 
your right to be on this project and thus have earned the merit to remain here 
unless you chose to leave.    

Seriously, if there is a difference between your expectations of what we, as 
mentors, should be be doing and what we are doing, you had every right to 
bring that up to make sure things are working correctly.    I've seen too many 
projects where the mentors "disappeared" and that obviously didn't go well, 
but I've also been involved in projects where the mentors yielded too heavy a 
hand and that also didn't go well. 

The main issue is that each project in Apache is different and thus stuff that 
worked great for one project might not work at all for another.    It's up to 
the project community itself to "figure out what works for them, within the 
limits set by the ASF".    The mentors are just there to enforced the last 
part and help answer questions regarding that.


> - I think Anne and Dick are just the right people to head up ESME.
>  I cannot think of a better person or team to lead the project.  While I
> think they have things to learn, I think that armed with the learning, they
> are going to rock... and I think they already rock!
>    - My high regard for Anne and Dick does not take away from my high
> regard for the other participants in ESME, past and present.  I would name
> names, but then I'd forget someone and that might be considered a negative.
>  So, let me just say that everyone who has committed to this project has
> moved it forward.

That's all perfect, providing they have the time and desire to step up and 
lead that way.   That may be part of the problem.   Right now, it doesn't look 
like anyone in the ESME community has enough time to really dedicate toward 
pushing the project forward.   That CAN be a problem, but it's also something 
that can change (in both directions) over time.


>    - I believe in burning down code.  Mesa for NextStep was my first
>    spreadsheet, code started in 1991.  In the entire time that I ran the
> Mesa for NextStep project, there were 3 "crash" defects and 1 "incorrect
> recalculation" defect reported among our 50K+ user base (there were other
> bugs, but data loss and incorrect recalc bugs were top priority).  For a
> bunch of years in the 90s, every Chrysler lease and loan was done through
> Mesa, billions of dollars were traded through Mesa on trading floors every
> day, national wireless companies did their financials using Mesa, etc.  The
> code base exists (more than 50% of the LoC is the same) today as an OS X
> product.  I was so disappointed with the Mesa codebase that when we started
> the OS/2 project, I made sure that none of the code came over.  So, my
> expectations and commentary about my code bases do not reflect the common
> perception about code bases.

And just to complete this bullet.....   when I then inherited the OS/2 code, 
one of the first things I did was "burn down" a LOT of the code, refactoring a 
lot of it and cleaning a lot of it up.   Reduced the line count by almost 20% 
while adding some features and fixing a bunch of issues.   So yes, even good 
code can benefit from a "burning down".     :-)

-- 
Daniel Kulp
dkulp@apache.org
http://www.dankulp.com/blog