Re: Test environments (was: upgrade) -- MIDRANGE-L

I've been biting my tongue, but I can't hold off any more!

The subject is testing and I feel that I have a valid reason for promoting - 
sorry, discussing - our range of automated testing
tools!

I'll try and be brief -


Automation of testing, like any other repetitive task, surely makes sense if 
you can do other work while the automated work is going
on in the background and will even highlight any action you need to take when 
it is completed.  In essence you are doing two lots of
work in the time it would take you to do one.  So the old tests have been 
recorded and can be played back whilst you record new
ones, and then these get played back when you record the next set.

All the time you are building up a test library and just manually creating new 
tests.  One of the main reasons that a lot of work
doesn't get tested thoroughly is because it is mundane and boring and the 
program worked the previous 'n' times so why bother to
check it this time.... I won't bore you with stories we have all heard before 
as I am sure most of us know of errors that have crept
through.

One of the main problems with testing or running parallel with the old and the 
new system is the extra amount of time needed to
check the test results.  Again it is down to people to go through the sheet of 
paper or the pages of screens to actually identify if
there is an error or not and after a long time checking through seeing the same 
results (i.e. the programs are running properly and
there aren't any errors) they may well fail to spot the errors.

Once an error has been spotted the next difficulty is providing the developer 
with the actual scenario that led to the error being
created so that it can be replicated and then fixed - not always an easy task - 
because us users never really know if we pressed
"enter" or "space" !!

Finally - and I am sorry that this is taking so long, but I would much rather 
lay my wares on the table than send a load of Spam
mail to the list (meant a joke - sorry English sense of humo(u)r!  - Why not 
use a subset of data for testing, and use that same
data each time so it is easy to roll back AND spot errors, instead of taking a 
copy of the live database where the data is in
different stages and who knows if the different results are due to an error in 
the program or just a different field value.

All the above are possible with our automated testing tool - TestBench - first 
time I mentioned the product!!

Please email me if you would like details or a demo. (and thanks for being 
patient and reading this far - I tried to be as
non-marketing as possible - AND not pretending to be other than a vendor)

Jamie Coles
jcoles@origsoft.com