Re: Death by analogy Part 2 (was Re: is ada dead?)

[Al Christians <alc@PublicPropertySoftware.com>]

Sure, the 2% example is low, but my analogy was 150% payback.  If the
math works, why wouldn't  people go for it?  Risk. Nothing is sure. 
I didn't say that it didn't pay off, but most won't make the investment.
Too much risk. Programming language is a hot-button issue for very
many.  
But it is way overblown.  Management is the number one factor related  
to software productivity.  The change in productivity from changing 
language is unlikely in most cases to exceed 30% or so, according to 
what I have read.  The variability of management is huge in comparison.
Investing in big budget software tools when you don't know what crazy
thing your PHB is going to do next is equally crazy.

What I was trying to illustrate by analogy is the difference between
value-based pricing (which is what the Ada software vendors cling to)
and cost-based pricing with costs spread over many more customers. 
Customers like cost-based pricing with costs spread over many more 
customers.  Most people will join the herd to gain these cost 
advantages. What is the value of food, water, or air?  Aren't we glad
that no one expects us to pay for these things what they are worth?

Consider the firm that is developing software and surviving without 
using Ada.  For example, may they use Java, Cobol, or Fortran to do 
things that are not too difficult with Java, Cobol, or Fortran, and
they are getting by ok.  Why risk a big investment to change language?
No reason.  Along comes a much bigger and more challenging development
opportunity, and they realize that Java, Cobol, or Fortran is not the
right way to go for taking on this bull by the tail.  Would they now
be wise to invest $100k/year+ in Ada tools so that they can rise to 
the occasion?  If they do, the odds are stacked against them.  Moving 
up to the next magnitude of difficulty or different problem domain and
changing languages simultaneous combines to be a big risk.  Not
advisable in the least.  Better they should have developed some good
experience in Ada by doing some minor projects with it over the years
before using it to try for the home run.  But who is going to spend
big bucks annually to license products for minor projects only?  Not
I when my budget gets reviewed.  

No need to argue with my crazy logic.  Just show me a bunch of 
companies developing applicationsoutside of the military and embedded 
systems markets in the 10-to-100-developer range that switched from 
anything else to Ada and succeeded and are still committed to it today.  
I'd be interested to know why and how.

Programming languages mostly propagate organically using low-level
contagion.  This is hindered by high $ price tag.


Al

"Michael P. Card" wrote:
> 
> Hello CLA-
> 
> I considered the implied cost/benefit analogy for Ada here to be incorrect, and I addressed that in "Part 1".
> 
> As a separate topic, I would also say that RE: development tools for C/C++, the old adage "you get what you pay for" is still true. I have been on
> jobs which used the freebie "gcc" tools that come from the RTOS vendor, and I have compared the kind of multi-threaded debugging/etc support that
> you get with these tools to those available from third party compiler vendors.
> 
> I have heard people make arguments similar to this wire-cutter analogy, e.g. "Yeah, but we get a C/C++ compiler FOR FREE with the operating
> system; these other tools cost thousands of dollars per seat! We can't afford that! How much difference could there be?"
> 
> It is unfortunate that these kinds of decisions are made at the beginning of a project before their full impact can be known. Only later do
> problems like excessive memory usage, poor debugging support, erroneous or inefficient code generation, lack of exception handling etc. become
> apparent, and by then it's too late to switch. Money gets wasted in lost productivity every day, and in the end you end up spending far more than
> you would have to buy a better tool at the outset.
> 
> So, I would say again that even beyond Ada vs. C/C++/Java/C#/fad du jour arguments, there is a "cheap toolset vs. expensive toolset" argument
> which is similar in that the "best" answer depends on what kind of job you are doing.
> 
> - Mike
> 
> Al Christians wrote:
> 
> > Jerry Petrey wrote:
> > >
> > > This is certainly not nonsense.  But don't feel bad.  Many people in
> > > the industry are unable to understand the true cost of developing
> > > software and only look at the up-front coding costs, tool costs, etc.  > That is one of the main reasons most software is over budget and of
> > > poor quality or not even ever delivered.
> > >
> >
> > Suppose you are an electrician and you hear about a new kind of
> > wirecutter.  There are studies that say this wirecutter improves
> > average productivity by 2%.  If you do the math, you can figure that
> > this is worth $2,000 to you over the expected 5 year life of the
> > wirecutters.  You go to the store and see $1,295 wirecutter on sale
> > next to all the others at $11.  Which pair do you buy?  Which toolmaker
> > has biggest market share and good cash flow to finance ways to improve
> > their product?
> >
> > For $1,284 most can think up a reason why they are not average.
> >
> > Al