Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Year 2000 computer problem

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Earlier this year I sent a letter to the Washington Post as I was
getting a bit perplexed at many comments in the press concerning the
so-called Y2K problem. I also discussed this with one of the Post
reporters. The letter, however, was never published. 
I have written to you in the past only to briefly provide other
information requested by one or more of your authors. This is a bit
longer than you probably would desire, but I believe it is relevant.
Let's see how many of you experienced engineers agree with my original
"Y2K" letter herewith:

Sophomoric articles and letters continue to appear in the Post regarding
the "Year 2000 Computer Bug". This is potentially a widespread computer
problem, but certainly not one requiring $50 billion to correct (as
quoted in recent Post articles). The ABC News Reports on the internet is
even estimating a cost of "up to $600 billion will be spent worldwide to
rewrite 250 billion lines of code". 
I am astonished at the increasing hysteria. I have put the majority of
people in one of two categories: those too primitive to know any better;
and those profiting either by dollars, or by laughter, at the first
group. 
I attended a meeting in Fairfax County, Virginia, recently of a "Y2K
Community Action Group". I thought I was at one of those meetings in old
British mystery films where, at any moment, I could expect an old tall
spinster in all black to come out and start yelling, "Repent!"   
For 40 years, I have been doing some type of programming, beginning in
the fifties before the introduction of such programming compilers as
Fortran, etc. and I believe, as do many others, that the technical issue
is grossly exaggerated. 
Ordinarily, this type of correction to alter input data formats (such as
the year) requires an original  "source" listing of the program. If that
is not available, there are means to obtain such a listing from the
current program in use. 
It is then a minimal effort to determine the entry for the date. It is
only necessary to change the 2-digit integer representation of the date
to a 4-digit integer to account for the "2000" problem. No other change
beyond that should be necessary. If further search is thought necessary,
there are several simple alternatives to quickly locate any operations
in the program involving the date.
With specific regard to COBOL programming (mentioned in several Post
articles on this subject), I would quote from a widely used computer
handbook: "COBOL programs are verbose but easy to read because most
commands resemble English. The programmer, therefore, hardly can help
documenting the program, and program maintenance and enhancement are
easy even if personnel change frequently. COBOL is the most widely used
programming language in corporate mainframe environments." 
As for the extent of the problem existing in non-programmable "chips" as
mentioned in some of these same articles, that may be a nuisance, but
the remedy would seem to be, in general, simple. Obsolescence is
obsolescence.
The problem is one of extent and inept management, past and present, not
technical procedures. Give the suspected programs to some clever high
school hackers and we will save a few billion dollars.