I’ve noticed a number of times that the info on the stats page may be troubled.
I think a blog post from yesterday can’t have been seen 20+ times when the daily stats of yesterday indicate about 5 visitors to the blog. It doesn’t seem logic to me that each visitor would have reloaded the page about 4 times, statistically speaking?
I’ve also noticed admin actions seem to be referenced as page views. If I post, it is 0 views. If I edit, 2 views. I guess one for showing the editing page, and one for showing the edited page?
I also noticed as logged in admin I am counted as visitor as soon as I post or visit a page of my blog, regardless whether I visit the page on own domain or write.as/blogname
If I change IP, which can happen on occasion (dynamic IP’s in hotels, airports, cell phone, etc), it looks like I am again counted as separate visitor whilst I am logged in.
Maybe the visitors and page views is something worth looking into?
Can you enhance the “plain pageview counter on the server-side” to exclude the logged-in admin, and perhaps the search engine crawlers? It seems logical to me that it would do that. The logged in admin I think should be easy to track, he’s logged in. And I should think search engine crawlers such as google, bing, webarchive, … are known too?
Anyone uses so called privacy respecting services who have a free tier and could help me how to include it in my w.a, like:
This is actually how it works already, as long as you’re viewing blog at your write.as/* address. If you access your blog on *.writeas.com or your custom domain, it will count your visit, since you’re not actually logged in on those domains (and we don’t have a mechanism for doing that right now).
Same goes for bots – we check all visitors against a list of crawlers, and ignore known ones.
Does the view incrementing happen every single time you post / post+edit? Do you have federation enabled? If so, people may be visiting your post between the time you first publish and edit your post.
I do have federation enabled. I just never got around actually configuring mastodon or any fediverse. Sounded too complicated past the one click to enable.
Would serendeputy.com, or archive.org, or archivebot be seen as bots/crawlers? They are three possible sources I know of.
@matt Hope you don’t mind another analytics reference. the archive.today service (archive.fo; archive.ph) also references in the write.as analytics. Hope it helps!