The perils of the home-made framework

Oh my, as a consultant, I have seen the results of the home-made frameworks so many time.

I think the process is like this:

There is an engineer who might know the business topic of the company well (e.g. some specific part of biology or chemistry), but maybe does not have the experience of many frameworks and definitely never had to deal with legacy code.

They think:

"None of the 1,000 web frameworks are good enough for us and we don't want people to need to learn JavaScript or a JavaScript based framework so I will write one that uses Python, because they already know Python."

"I'll create an in-house framework where people can write Python and HTML combined in a single file (because I have not see CGI scripts and PHP from the 90s) and my framework will convert that to some JavaScript framework and that will be perfect."

"I don't write tests, because my code just works and besides this is only an in-house application. And I don't really know how to write tests."

"I don't need to write documentation either because it is obvious to everyone, even people who have never seen my framework how to use it. After all it is just more Python and the backend is written in Python anyway."

"Then I leave the company."

I have to admit, there is some joy in helping companies get out of such mess. However it also frustrates me to no end. I always feel that it was such a waste of time and money for the companies that allowed this to happen to them.

Published on 2021-08-04 by Gabor Szabo