Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I'm sure what you mean to imply is "it would be trivial", and I agree, but any professional developer can tell you it wouldn't be literally a day.

It would start off as being a week, maybe two or three as the developer runs into a few tricky corner cases. Then it would sit in a backlog for a few weeks until QA sees it. Then a few days later you would get it back with a bunch of issues, some of which are legitimate bugs, others that are differences in what each side thinks the requirements should be. This would in turn get passed to various stakeholders who would garden shed their vision for an all in one paint program, that would lead to a large meeting between developers, sales, product management, and many others. Eventually executive would step in with a specific vision with no obvious involvement from everyone else, and then the actual development would begin.

As one of my co-workers says, "It will be ready two weeks after the requirements stop changing."

I'm not even being sarcastic. I've seen this same pattern on at least a dozen new products I've helped develop. This is when things go reasonably well. When this don't go well, this horrible cycle repeats for years until either some eldritch horror is released, or the entire project gets shutdown.

It takes more than 30 seconds to open a file, it takes more than 5 minutes to make a trivial one one change, it takes more than a day to test and push that one line change, and it takes a week before that change is verified and merged. An entire application - regardless of how specific the requirements and how well understood the implementation - will takes weeks to months to get to the point where it can be released to the public.



paint.net would not be trivial to build.


There's a major difference between forming an argument and simply touting an opinion.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: