Had many fun bugs in my time: Buffer size on line printer out of sync with what the program (COBOL) was sending and would produce variable results every so oft

Had many fun bugs in my time: Buffer size on line printer out of sync with what the program (COBOL) was sending and would produce variable results every so often when WRITE BEFORE and WRITE AFTER got used. Then the ones which work fine with debug libraries and not on production ones as turns out the debug libraries had a unintended fix to a bug that was yet to be known. Also had programs work fine when you don't use debug libraries. Then we have driver bugs, when if you have even touched graphics will of encountered, more so when we have a shift in driver model as ME introduced and was stable enough for windows XP, same with Vista which is stable in 7 and 8. Surfice too say the hardest bugs almost always turn out to be hardware or some external software being documented wrong or undocumented features. Also new revisions in hardware can casue the smallest of issues in the niches of uses, but it can be you and it is a small lonely world when you try to fix one of those. Though for me the hardest bugs are always the ones which you know were abouts they are, you are just unable to prove it too those enable to investigate it too the level of being able to prove it is there issue. Also some bugs for some are easier for others and we have all had a bit of code or aspect of life that we seem blinkered from seeing what is wrong; Somebody else could look at it and solve it in seconds. Then we have also been that somebody looking at others code or issue and seen the problem and solution. Sometimes we see the problem before they can see it as a problem. Had classic in the 80's doing a mailing list update to add on postal codes (same as USA ZIP codes only UK flavour). I advised that some address's will already have a postal code tagged onto the address lines and in some cases could end up with not only 2 post codes on the address but also not always the same as the one entered may of been slightly wrong. This was dismissed until 5pm friday on prepping to head 300miles home south for the weekend. Was late heading home that weekend. Though that would be an annoying bug more than a hard bug, however hard it was upon my weekend. But we learn how to highlight issues better, or overhighlight. I have learned much from weather warnings at least and how they lean more towards the worst case situation thesedays more than say in the 80's. Lastly though we have those intermitant bugs, so rare and niche that the impact is so neglagable and negatable that they are easier to work around instead of trying too fix. Mostly such fix's would cost more to identify fully and address than any work around. I suspect we have encountered more of those than you realise. With that, is how hacking was born after all. Without bugs would we of had the hacker history gave us - that in itself is worth a thought.