I like the term a lot. However, popup blocking is actually a real feature since sometimes you want to block popups and sometimes you don't. Designing a system that makes sorting out allowed vs. disallowed in a user friendly way is non-trivial and is in fact a feature of a browser. It's especially a feature if enabling means, as it did for the Internet Explorer team, going back and changing old functionality to allow for blocking.
I'd call it an anti-feature at the browser API level. Code to launch a new browser window is nonessential and it was included as a feature long before anyone thought about whether it was a good idea for the product as a whole.
Today, popup windows are an accepted feature, though still insanely overused. The only time I have ever found popups at all valuable is for instruction annotations during stateful sessions where winding up in a situation where the 'back' button is needed is undesireable. I would still usually prefer to simply open a new tab.
Ok: one other time, this web application at my last company had a subtle bug where putting two consecutive slashes in the URL of the login page caused weird things to happen. Their answer was to launch the app in a popup window with the correct URL.
I think if it had been always impossible to launch popup windows by default, then websites would be generally better designed.
It's well thought out article. One key insight that Benjamin Hill offers is that anti-features are often driven by business model requirements. If a competitor "un-implements them" (enabling the feature) it's hard for the original firm to follow without changing business model. This applies beyond open source vs. proprietary competition.
That is a great term. I've been searching for years to sum up that concept so elegantly. Probably since first time I encountered copy-protection on a video game (X-Wing on a 486, I think it was)
