Despite racking up 11K on StackOverflow over the years, I’m not very active on meta. Despite this fact, this wildly unpopular policy announcement caught my attention:
Yes, you read that right, all code taken from StackOverflow is going to require attribution as of March 1st, 2016.
This isn’t necessarily a bad thing. I’ve occasionally included a comment with a link to a SO posting to clarify unusual code or my thought process but what concerns me the most is how vague the post gets when it comes to what is actually code:
But what is code?
We will give you guidance on identifying code in an upcoming FAQ, plus guidance on how best to comply with the attribution requirement. But ultimately, identifying code will be a judgement call on your part. We have full faith in your ability to do this.
In other words, “we’re doing this now but we’ll tell you what it actually means later.”
This might be the single most important definition in this policy. Looking at my history from today, I visited StackOverflow to troubleshoot a minor problem. The solution was this post. In summary, I needed to add a “,null” to my method call so the correct overload would be selected. Other than different strings in the parameters, my code is identical to the accepted answer. Should I attribute that?
I’ve replaced my entire library of reference books over the last 15 years with online resources. If I need to lookup syntax etc, it is always quicker to do so online (and stackoverflow ranks top among my searches). I’m not alone.
I wonder how long it will be before some lawyer decides to spider out all the “code” in stackoverflow and start comparing that against source trees.