If you care about your writing, then your aim is for people to read your whole post, not just parts of it. And you might think you want to know how much time people spent on your page (i.e. what your bump rate is) but even that tells you nothing: what are you going to do with that information? And that's not rhetorical: do you actually work on your writing based on whether people scroll past the fold? Because if you do, I'd love to hear what your workflow around that is.
Unless you're trying to monetize your readership through ad placements, analytics like these are a little lie we like to tell ourselves matter, when really they don't. They're nice to look at, but eye candy isn't a good enough reason to run tracking and behavioral profiling on a blog.
> And much like profiling any program I’ve written, the hotspot is always something different from what I’d have expected. Having analytics doesn’t change what I write. I still enjoy the information.
I like seeing the hotspot so I know what to focus on in the future. It's really no different than in startups, let's say you build features A, B, and C, but most people are focusing on C, knowing that info ensures you better serve your users. For a blog, your user is your reader. You might say server level analytics might work, but just as in startups, it assumes that there is only one feature (or for blogs, only one idea) per page. Hence, intra-page analytics can be useful.
and being in public visiting a store vs reading a website in the privacy of your own home are clearly very different situations.
>but that is what is happening regardless
So the justification is "everyone else is doing it" - OK.
Enough going around in circles.
That people know about this type of tracking with stores and do not with websites is not my problem, and at best, they should be educated about that, and indeed we have laws against this exact thing, such as CFAA, and no judge will let you off the hook just because you said it was in cyberspace and not meatspace that you invaded someone else's servers. Therefore, both are analogously the same.