31fps by Sam Purtill A blog about business, technology, and life

Hard Lesson Learned Today

I have been coding for like 10 hours off and on today... I ran into a problem very early on in my day; I had to change how I was grouping all of my users by. I knew this was going to be a bit of a problem to re-engineer, but it was going to be worth it in the long run. So I began to start hacking the django.contrib.auth files, and realized about an hour into it that this was a really stupid way of doing things since I want to be able to update my auth without screwing things up when Django has makes it's new releases.

So I did something really stupid. I got frustrated and just started going through deleting files that I had written that had hacked the Django auth system. Should have saved it in another directory for backup at least. Anyways, I lost like 4 hours fixing everything I had broken. In the end, it was totally worth it. I have a completely new way of organizing my users, and it is so much more efficient and is not as heavy on the database. I am already glad I had all this crap happen to me today, I learned a ton about Django from making a big mistake.

They say in a startup, "everything that can go wrong, will go wrong." Yep.

And two phenomenal essays by Paul Graham that I read today - Microsoft and Why to Not Not Start a Startup.