RE: SIP is not Takt Time

Hi Kent,

This thread has examples of why a software person can be expected to have a 
difficult time seeing the bookkeeping image. It also tells me that no words 
brought forth from my domain is going to help until you see the concept of 
binary reasoning that drives all of bookkeeping in your domain. Once you 
see the pattern of binary relations in the issues in this thread, I believe 
bookkeeping's image will begin to form for you.

>I can see two things going on here. One is the definitions of the
>various measures in manufacturing. The other are the appropriate
>measures for use in software development.

Think of the two things as a binary relation that make up a whole, just as 
forces versus solution make up a context.

> >I finally got around to reading the SIP paper you posted here in
> >December. I'm afraid I cannot agree with your definition of SIP.
>
> >You define it as the time taken for working code to emerge from the
> >value stream. This is known in manufacturing as Takt Time or in
> >plainer English, Lead Time or Cycle Time.
>
>" from http://www.vmec.org/lean_corner/takttime.html
>
>Takt time is the goal. It must be reached to satisfy demand.

There is a passage in this paper that says: "Cycle time is a measured 
value, not a calculated value as takt time is. In other words, you must go 
out to the floor, and measure the time it actually does take to manufacture 
the product."

Takt time as David defines it in his message, which makes sense to me, has 
a relation to cycle time as parts have to a whole. Using that image the 
quote above might be improved by saying "Cycle time is an expression of 
potential...." where potential is an accumulated image of a complete set of 
takt measures that make up a cycle time. The binary aspect, is an 
isomorphic relationship between parts versus whole. A physical measurement 
of takt values is an expression of a completed time for a cycle.

>Cycle time is a measured value, not a calculated value as takt time is.
>In other words, you must go out to the floor, and measure the time it
>actually does take to manufacture the product."

This is the precise pattern that takes place in bookkeeping. The takt 
values are debits; cycle time is credit. There is a list of one or more 
debits for each expression of credit, just as you have a list of takt 
values for each expression of a cycle. The one and only one credit 
expression is made clear in this example because credit, or cycle, 
expresses an image of a complete set of measurements.

>I agree Value is #1, but I don't know how to measure it directly. SIP is
>production lead time, not Takt time (takt in XP is always one week, we
>just vary the size of the unit of production.)

The place to measure software's value is in the cycle time itself, and not 
as a measure of takt value contributed to the cycle. If software can 
shorten cycles, for whatever reason, the benefits can be used in other 
places. Cost accounting reveals that cycles form a recursive compositional 
image of any entity as a whole. IOW the company as system behaves in 
cyclical images. The top cycle, the company itself as a whole, is an 
expression called "equity," which gets disbursed to equity owners, as a 
measure of value returned after all of the cycles that make up the company 
in a particular time period.

Dan