On a professional level, I try very hard to give everyone the benefit of the doubt and reject categorizations of people’s competence in their jobs unless there is literally no way to characterize someone as anything other than incompetent. Then again, there are simply some people you encounter whose continued presence among the living constitutes a negative net benefit to the continued existence of the human race.
Today, I had the unavoidable need to enter the following comment into code I was writing that interacts with someone else’s system. I had to really consider the fact that this person might one day read this comment, so I did not make it lightly or without forethought. In the end, I chose to leave it in the code as a necessary explanation for the mentally deranged-seeming coding choices I was forced to make, so that future engineers could maintain the code without losing their minds…
-
‘Prtcnt table definition formatting, naming convention, and numbering are all
‘hosed, because someone doesn’t understand what a standard is. Fixing this
‘in code would require manipulating the msREGEX object from Basic Control
‘Language (cimBasic), which is a massive task not worth the effort.
‘Therefore, in this case, we have to hard-code column names to get
‘access to the data.
‘[Addendum]Oh, also, as it turns out, defective coder forgot to even
‘specify the identifiers for the last three columns of the table in the
‘table definition, so code couldn’t fix it anyway! Divining column functions
‘via labor-intensive iterative process henceforth known as “plug and pray”…
‘[Addendum – 2]On further analysis, it seems that defective coder added one
‘record to the table that has different column names and purposes from the
‘rest of the rows in the table, effectively shoehorning two tables into one
‘file with no table definition or schema at all provided for one of them.
‘Divining secondary table purpose and definition via “plug and pray” methodology…
My good Mister Hartley, I feel your pain.
I am a software engineer who graduated RPI in 89. I have struggled with this daily. Today – literally today – I am writing performance-killing code in a real-time application because another coder will not accept that their backend process is writing duplicate rows and I am forced to weed them out. These duplicate rows are happening because the coder is trying to use metadata that does not change (often) to determine whether the process already wrote instance data, instead of actually looking at the instance data! The coder in question doesn’t get the difference, even when shown the duplicate rows. The irony – I wrote them a performant filtered search function that they stubbornly refuse to use to see if “they already wrote a row like that”.
The word incompetent has come to mind more than once. The struggle is real, keep the faith.