Go BackGo Forward Index Home    

Project
Code
Meter


General Frequently Asked Questions


Is productivity measurements bad for programmers?

Most often the boss or client will undervalue the programmers work, causing unrealistically early deadlines, mental pressure, carelessness, personal conflicts, and dissatisfaction and detachment of the programmer, leading to low quality products and missed schedules (on top of bad feelings).
Being overvalued is dishonest, and leads to overpriced offers quoted by the company, losing appeal to clients, and ultimately cutting jobs.
Productivity measurements help programmers being valued, (not overvalued nor undervalued) which is a good thing.

Why not use cost estimation methods like COCOMO or COSYSMO?

These methods have some uses as tools at the hands of experts, since they will only produce a result as good as the input estimates they are given, thus require the user to know (or guess) the size, complexity and quantity of the source code sub components.
ProjectCodeMeter can be operated by a non-developer and usually produces more accurate results in a fraction of the time and effort.

What's wrong with counting Lines Of Code (SLOC / LLOC)?

Many cost estimation models indeed use LOC as input data, while this has some validity, it is a very inaccurate measurement unit.
in counting SLOC or LLOC, these two lines would have the same weight:
  i = 7;
  if ((i > 5) && (i < 10))  while(i > 0) ScreenArray[i][i--] = 0xFF;//draw diagonal line
While clearly they require very different effort to create.
As another example, a programmer could spend a day optimizing his source, thus reducing the size by 200 lines of code, does this mean the programmer had negative productivity? of course not.
ProjectCodeMeter uses a smart differential comparison which takes this into account.

Does WMFP replace traditional models such as COCOMO and COSYSMO?

Not in all cases. WMFP+APPW is specifically tailored to evaluate commercial software project development time (where management is relatively efficient), while COCOMO evaluates more factors such as design time, and COSYSMO can evaluate hardware projects too.
WMFP requires having a similar project (analogous), while COCOMO allows you to guess the size (in KLOC) of the software yourself. So in effect they are complementary.