What is your worst development experience ?
-
Using a computer with insufficient resources. A few years ago (more than I care to remember) I was developing a hardware graphics processing module using FPGA's. I had what I considered the final iteration and wanted to run one last round of simulation for the entire design. Just after lunch (1:00pm), I set up the simulator for 30 seconds of real-time processing (1500 frames) and started the run. At 6:00pm it was still processing (phew!) so I left everything running and went home. The next morning (7:00am) it was still running but around 9:00am the software reported the simulation was complete, so I clicked the [Continue] button and the computer then took another 2 hours post processing files until it finally generated an "Insufficient Memory" error message, deleted all the results and terminated the program. I had 32Gb of memory installed (was 8Gb) and the repeat simulation took just over 1 hour... (my original expectation) All that for a 30 second simulation. BTW the design was successful.
Live long and prosper
-
Using a computer with insufficient resources. A few years ago (more than I care to remember) I was developing a hardware graphics processing module using FPGA's. I had what I considered the final iteration and wanted to run one last round of simulation for the entire design. Just after lunch (1:00pm), I set up the simulator for 30 seconds of real-time processing (1500 frames) and started the run. At 6:00pm it was still processing (phew!) so I left everything running and went home. The next morning (7:00am) it was still running but around 9:00am the software reported the simulation was complete, so I clicked the [Continue] button and the computer then took another 2 hours post processing files until it finally generated an "Insufficient Memory" error message, deleted all the results and terminated the program. I had 32Gb of memory installed (was 8Gb) and the repeat simulation took just over 1 hour... (my original expectation) All that for a 30 second simulation. BTW the design was successful.
Live long and prosper
I'm not sure exactly what you're asking about. My worst *coding* experience? Probably one of the innumerable death marches I've been involved in. Pick one. (I used to consult, and was often involved in project rescue, so i've been involved in a lot of failed projects - it's not fun) In recent memory, strictly in debugging terms, I ran into an issue in my parser code with a particular grammar and the issue only cropped up after over a minute of lalr(1) table generation so each time i started the debugger i had to wait over a minute to get something useful to debug. it took me awhile to track down that bug. as i recall, there was nothing wrong with that section of code, it was something upstream that was causing the problem. :((
Real programmers use butterflies
-
Using a computer with insufficient resources. A few years ago (more than I care to remember) I was developing a hardware graphics processing module using FPGA's. I had what I considered the final iteration and wanted to run one last round of simulation for the entire design. Just after lunch (1:00pm), I set up the simulator for 30 seconds of real-time processing (1500 frames) and started the run. At 6:00pm it was still processing (phew!) so I left everything running and went home. The next morning (7:00am) it was still running but around 9:00am the software reported the simulation was complete, so I clicked the [Continue] button and the computer then took another 2 hours post processing files until it finally generated an "Insufficient Memory" error message, deleted all the results and terminated the program. I had 32Gb of memory installed (was 8Gb) and the repeat simulation took just over 1 hour... (my original expectation) All that for a 30 second simulation. BTW the design was successful.
Live long and prosper
Dealing with a test environment that took an unbecoming length of time to configure. It was such a PITA that I routinely checked in code untested, which was definitely frowned on if it caused problems. Thankfully it rarely did, and never in a way that couldn't easily be patched (the product had no-restart patching, even in deployed software, forty years ago). I had developed an in-house application framework, so it couldn't be tested without knowing how to configure the test environment for whichever applications might be affected by the latest changes. I could have asked that the test group put someone on call, but it was hard enough to convince some folks that I should work on an application framework, not customer features. So I simply inspected my code carefully, checked it in with fingers crossed, and took the flak when it caused sanity tests to fail. On balance, it ended up being far cheaper to do it that way, but it was such a violation of The Process that it would never have been officially countenanced.
-
Using a computer with insufficient resources. A few years ago (more than I care to remember) I was developing a hardware graphics processing module using FPGA's. I had what I considered the final iteration and wanted to run one last round of simulation for the entire design. Just after lunch (1:00pm), I set up the simulator for 30 seconds of real-time processing (1500 frames) and started the run. At 6:00pm it was still processing (phew!) so I left everything running and went home. The next morning (7:00am) it was still running but around 9:00am the software reported the simulation was complete, so I clicked the [Continue] button and the computer then took another 2 hours post processing files until it finally generated an "Insufficient Memory" error message, deleted all the results and terminated the program. I had 32Gb of memory installed (was 8Gb) and the repeat simulation took just over 1 hour... (my original expectation) All that for a 30 second simulation. BTW the design was successful.
Live long and prosper
Being told that all my SQL had to be on a single line, and not formatted. It would be as readable as formatted SQL, which is true if you just read out aloud the sentence. Wrote a VS plugin that would show a formatted SQL statement after selecting it and pressing a button. Wasn't allowed to copy/paste that formatted statement into code though, it had to be on a single line. Lots of scrolling just to read the statement wouldn't matter, that was preferred to people "wasting time" on formatting.
Bastard Programmer from Hell :suss: If you can't read my code, try converting it here[^] "If you just follow the bacon Eddy, wherever it leads you, then you won't have to think about politics." -- Some Bell.
-
Using a computer with insufficient resources. A few years ago (more than I care to remember) I was developing a hardware graphics processing module using FPGA's. I had what I considered the final iteration and wanted to run one last round of simulation for the entire design. Just after lunch (1:00pm), I set up the simulator for 30 seconds of real-time processing (1500 frames) and started the run. At 6:00pm it was still processing (phew!) so I left everything running and went home. The next morning (7:00am) it was still running but around 9:00am the software reported the simulation was complete, so I clicked the [Continue] button and the computer then took another 2 hours post processing files until it finally generated an "Insufficient Memory" error message, deleted all the results and terminated the program. I had 32Gb of memory installed (was 8Gb) and the repeat simulation took just over 1 hour... (my original expectation) All that for a 30 second simulation. BTW the design was successful.
Live long and prosper
The first time I went on a customer trip with my boss. I was an intern at the time. We were writing a control system for part of a manufacturing line. At one point, a part of the application I had written stopped working. I couldn't understand what was wrong, my boss was getting pissed, and the customer was looking unimpressed. We spent hours going over this and not getting anywhere. Finally, I went back and compared the code we were running to my original copy. They were different, and I hadn't made the changes. When I showed it to my boss, he admitted changing some things because he didn't like how I did part of it. When we put my code back, the application started working again. I didn't say more than a half-dozen words to the guy the whole 10-hour drive home.
Software Zen:
delete this; -
Using a computer with insufficient resources. A few years ago (more than I care to remember) I was developing a hardware graphics processing module using FPGA's. I had what I considered the final iteration and wanted to run one last round of simulation for the entire design. Just after lunch (1:00pm), I set up the simulator for 30 seconds of real-time processing (1500 frames) and started the run. At 6:00pm it was still processing (phew!) so I left everything running and went home. The next morning (7:00am) it was still running but around 9:00am the software reported the simulation was complete, so I clicked the [Continue] button and the computer then took another 2 hours post processing files until it finally generated an "Insufficient Memory" error message, deleted all the results and terminated the program. I had 32Gb of memory installed (was 8Gb) and the repeat simulation took just over 1 hour... (my original expectation) All that for a 30 second simulation. BTW the design was successful.
Live long and prosper
Spending a beautifully sunny Saturday morning in the cold server room of Skopje airport, because the (redundant) servers weren't communicating with each other, only to find that it was because the installation engineers hadn't followed my very clear, very detailed instructions on installing the Synaptic software. They hadn't bothered following the section on configuring exceptions for Synaptic's internal firewall, so the servers weren't being allowed to talk to each other. As is typical, once I'd found the problem, it took only a couple of minutes to fix -- but the day was ruined.
I wanna be a eunuchs developer! Pass me a bread knife!
-
Being told that all my SQL had to be on a single line, and not formatted. It would be as readable as formatted SQL, which is true if you just read out aloud the sentence. Wrote a VS plugin that would show a formatted SQL statement after selecting it and pressing a button. Wasn't allowed to copy/paste that formatted statement into code though, it had to be on a single line. Lots of scrolling just to read the statement wouldn't matter, that was preferred to people "wasting time" on formatting.
Bastard Programmer from Hell :suss: If you can't read my code, try converting it here[^] "If you just follow the bacon Eddy, wherever it leads you, then you won't have to think about politics." -- Some Bell.
-
Being told that all my SQL had to be on a single line, and not formatted. It would be as readable as formatted SQL, which is true if you just read out aloud the sentence. Wrote a VS plugin that would show a formatted SQL statement after selecting it and pressing a button. Wasn't allowed to copy/paste that formatted statement into code though, it had to be on a single line. Lots of scrolling just to read the statement wouldn't matter, that was preferred to people "wasting time" on formatting.
Bastard Programmer from Hell :suss: If you can't read my code, try converting it here[^] "If you just follow the bacon Eddy, wherever it leads you, then you won't have to think about politics." -- Some Bell.
Eddy Vluggen wrote:
Being told that all my SQL had to be on a single line, and not formatted.
I could see that as a way to stop all dogfights about The One and Only True Way to Format Code. You should consider starting to program in APL. In APL one major goal is to write the entire application on a sigle line.
-
Using a computer with insufficient resources. A few years ago (more than I care to remember) I was developing a hardware graphics processing module using FPGA's. I had what I considered the final iteration and wanted to run one last round of simulation for the entire design. Just after lunch (1:00pm), I set up the simulator for 30 seconds of real-time processing (1500 frames) and started the run. At 6:00pm it was still processing (phew!) so I left everything running and went home. The next morning (7:00am) it was still running but around 9:00am the software reported the simulation was complete, so I clicked the [Continue] button and the computer then took another 2 hours post processing files until it finally generated an "Insufficient Memory" error message, deleted all the results and terminated the program. I had 32Gb of memory installed (was 8Gb) and the repeat simulation took just over 1 hour... (my original expectation) All that for a 30 second simulation. BTW the design was successful.
Live long and prosper
Had a boss who should have been running his company rather than writing code. I inherited a codebase that had a function that was about 700 lines long, and it had been copied 7 times, each with different parameters and had very subtle changes being made to the bodies of each duplicated function. After about 3 years of seeing this code almost daily (it was pretty much central to everything the app did), I took it upon myself to refactor the code so there was a single function. The next logical step would've been to refactor the one function some more into smaller chunks, but by that time I was still too afraid of breaking it that's where I stopped.
-
Eddy Vluggen wrote:
Being told that all my SQL had to be on a single line, and not formatted.
I could see that as a way to stop all dogfights about The One and Only True Way to Format Code. You should consider starting to program in APL. In APL one major goal is to write the entire application on a sigle line.
Member 7989122 wrote:
I could see that as a way to stop all dogfights about The One and Only True Way to Format Code.
Yup, was also one of the reasons :thumbsup: Still, there is no "one true way"; simplest solution would be to "accept all", or have some tool do the formatting. Not wanting to have discussions on were whitespace goes is a bad argument to stop all formatting.
Member 7989122 wrote:
You should consider starting to program in APL. In APL one major goal is to write the entire application on a sigle line.
Only worked there for a year, two tops. No discussions on how to format the VB.NET code since VS automatically did that.
Bastard Programmer from Hell :suss: If you can't read my code, try converting it here[^] "If you just follow the bacon Eddy, wherever it leads you, then you won't have to think about politics." -- Some Bell.
-
Using a computer with insufficient resources. A few years ago (more than I care to remember) I was developing a hardware graphics processing module using FPGA's. I had what I considered the final iteration and wanted to run one last round of simulation for the entire design. Just after lunch (1:00pm), I set up the simulator for 30 seconds of real-time processing (1500 frames) and started the run. At 6:00pm it was still processing (phew!) so I left everything running and went home. The next morning (7:00am) it was still running but around 9:00am the software reported the simulation was complete, so I clicked the [Continue] button and the computer then took another 2 hours post processing files until it finally generated an "Insufficient Memory" error message, deleted all the results and terminated the program. I had 32Gb of memory installed (was 8Gb) and the repeat simulation took just over 1 hour... (my original expectation) All that for a 30 second simulation. BTW the design was successful.
Live long and prosper
I remember working with a old Dual Core between 2008-2009. I have a lot of problems with the power supply!!! :mad:
