Shannon's Journal Discussion Thread, Vol. 5

[Shannon]

@Shannon

Hi how are you,

Why would there even be a chance that frm would not be included in USLM 4.2, let alone it not being updated to 4.9+?

There are always factors to consider that may or may not be known about.  The reason I use and rely on the models is because they seem to consistently be able to see the effects of variables and information I may not be aware of.  For that reason, I run through the models each decision, such as "Do I turn on FRM for this program?" and consider the answer.  

In some cases, the state of FRM being incomplete and less than 100% effective might overall damage the results more than help.  In those cases, the FRM is left out, turned off.  

If you take that into account, as well as the fact that I don't know how long FRM 4.9 will take to develop, it makes sense that FRM may not make it into USLM 4.2.  

Also if you were trying to test how effective it will be with the frm turned on or off- then why not include two versions one with it on and one with it off?

Two reasons.  

1. Having two versions makes the job of figuring out valid data from users experiences exponentially more difficult, because it's not consistent, and not everything is being reported.  I already get only a slim amount of useful data.  Cutting that in half isn't really helpful.

2. The amount of work and time to build those programs doubles, and that's not insignificant for either the amount of work or the time involved.  Once I have the script finished, it typically takes me 2-3 days to go from script to published.  I am already drowning in how long these programs now each take to make.  It was 2.5 months to script, build and publish MIR v3, and that was after 3 years of R&D.  Spending time and energy unnecessarily is not on my to-do list.  Especially on a program that most people will be getting as a free upgrade because they weren't happy with their purchase of v4 or 4.1!