I Like a Good Quote

Murphy´s Laws of Computer Programming
1. Any given program, when running, is obsolete.
2. Any given program costs more and takes longer each time it is run.
3. If a program is useful, it will have to be changed.
4. If a program is useless, it will have to be documented.
5. Any given program will expand to fill all the available memory.
6. The value of a program is inversely proportional to the weight of its output.
7. Program complexity grows until it exceeds the capability of the programmer who must maintain it.

Top | New | Old

SW-User

that sounds like laws of users, programming would be more

- the optimizer always has bugs in it
- someone else's code always uses different indentation rules from you
- code you write on Friday afternoon will not be understandable on Monday morning
- moving to a new compiler will break something
- comments only document the parts that are obvious
- API calling conventions changed AGAIN
- users want something that seems easy to them but is actually difficult or impossible
- everything gets messier and uglier until eventually it has to be rewritten
- you have 2 days to learn a new language (bonus if it's Haskell)
- bug that disappears when you run it with the debugger

View 8 more replies »

RealtaReoite · 61-69, M

@SW-User My life...

And right now, I'm really trying to understand why one particular serial port on a microcontroller gets overrun errors (and locks up, ignoring any and all attempts to clear the error) at seemingly random times, despite the fact that the logic analyzer seems to claim that no overrun can possibly be occurring -- and the problem only shows up when the debugger is NOT connected.

Fortunately, I'm old enough now that all I can do is laugh. And wait for inspiration to arrive.

RealtaReoite · 61-69, M

Thanks, @CharlieZ. I've been doing this for a lot of years and "exorcism" is pretty much on the mark some days.

Somewhere in a dark cranny in my brain is a vague memory of some old Motorola (I think) debugger appropriately called an Exorciser. Back in the days that there was still truth in advertising.

Damn, I sound like a curmudgeon.

CharlieZ · 70-79, M

@RealtaReoite It´s both a bit scaring and a kind of hope that two old venerable sources of experience are still somehow valid.

One of them is and will always be "No Silver Bullet – Essence and Accident in Software Engineering" by Fred Brooks (1986), a must read.

The other one is (sadly) the history of the Sabre software, when they upgraded the (otherwise working)inherited code.
A what "not to do" strong example.
In other words, to try to fix "oportunistic code" with more "oportunistic code" is the best récipe for disaster.
Eveen when that oportunism once was as good as icecream, it´s always also a one shoot hunting shotgun.

Add a comment...