Re: [whatwg/url] Omitting www might be confusing for users (#568)

Thanks for your reply, estark37!

> I don't really feel convinced that anything needs to change in the spec, since it's already a "may", allowing browsers to make their own product decisions.

It is indeed a product decision on whether to hide the "www" or so in the address bar. However, I also think that the word "may" can never be an excuse for a spec to not remove a (possibly?) unwelcomed practice.

On whether omitting "www" (or in the canonical terms, displaying only the "registrable domain") is ill-famed, I would suggest one perspective, described as below:

Chromium started to hide "www"s from the address bar from 69 (Sept 2018, later reverted) and 76 (July 2019). Two flags called `omnibox-ui-hide-steady-state-url-trivial-subdomains` and `omnibox-ui-hide-steady-state-url-scheme` were useful on reverting this behavior, effective from Chromium 76 to 79. After Chromium 79 (Dec 2019), the seemingly only way to un-elide "www" is installing the extension [Suspecious Site Reporter](https://chrome.google.com/webstore/detail/suspicious-site-reporter/jknemblkbdhdcpllfgbfekkdciegfboi) (and also, "agree to the Google Terms of Service and Privacy Policy"), since Chromium [explicitly hard-coded the extension ID](https://github.com/chromium/chromium/blob/99314be8152e688bafbbf9a615536bdbb289ea87/chrome/browser/safe_browsing/user_interaction_observer.cc#L29) to disable elision. On the page of that extension, there are tons of comments complaining that Chrome users need to install an extension to un-elide "www". In my perspective, this might be a sign implying that the "www" elision is not popular.

I have been misled by the elision in my personal experience, and some web searches I did on this topic also presented more negative reviews than positive ones, but this perspective might be more persuasive since Chrome/Chromium and Chrome Web Store have been a majority on the web browser market.

> Maybe it would be reasonable to add a caveat at the end of the section to note that there are tradeoffs to these simplifications -- maybe something like this?

I agree that it would be better to add a caveat if this is kept in the spec. It might also be good if the spec can encourage vendors to leave the choice to users (for example, as one configurable setting or flag).

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/url/issues/568#issuecomment-755111937