Confirmed users
513
edits
Auto-tools/Projects/PublicES: Difference between revisions
→Postmortem (February 5th 2014)
Klahnakoski (talk | contribs) |
Klahnakoski (talk | contribs) |
||
| Line 155: | Line 155: | ||
= Postmortem (February 5th 2014) = | = Postmortem (February 5th 2014) = | ||
My biggest mistakes were in time estimation. Generally, I underestimated | My biggest mistakes were in time estimation. Generally, I underestimated communication/coordination overhead, which can be significant: | ||
* '''Security reviews take time''' - The security team was fast and responsive, but it still took time to schedule meetings, and fix the design. | * '''Security reviews take time''' - The security team was fast and responsive, but it still took time to schedule meetings, and fix the design. | ||
| Line 167: | Line 167: | ||
** scheduling IT to push the next version | ** scheduling IT to push the next version | ||
: all takes time; a couple of days at least. Compare this to development: where you can squash a bug in an hour. | : all takes time; a couple of days at least. Compare this to development: where you can squash a bug in an hour. | ||
Another issue, mentioned above, is the extra effort in adding tests to the code-base: More tests => more test code => more bugs in test code => more bugs found in production code => more tests. A very good thing for sure, but I should have recalled from my past that tests account for 2/3 of coding effort. I should have tried to estimate the effort to rewrite the existing code, then doubled it to get an estimate for writing the testing code. I had lazily glossed over that non-trivial amount of effort. | |||
Finally, Python has idiosyncrasies, and the bugs caused by utf8/latin1 were the most time consuming. | |||