408788 - Ambiguous wording in "Have I visited this website before today?"

Status: VERIFIED FIXED

(Firefox :: Page Info Window, defect)

Description

[Axel Hecht [:Pike]]

As Michael pointed out in bug 377076 comment 12, the wording 

"Have I visited this website before today?"

in http://mxr.mozilla.org/mozilla/source/browser/locales/en-US/chrome/browser/pageInfo.dtd#116 is ambiguous, depending on whether you group the "before" to "visited" or to "today".

Comment 1

[Johnathan Nightingale [:johnath]]

To be clear then:

 - The primary intent of the line is to communicate to the user whether this is a site they have visited "in the past," BUT
 - It's relatively straightforward for sites to build magic that gets their name added to history hundreds of times, either to deliberately undermine history-based UI, or just because they're buggy, SO
 - We want to use a test which removes that possibility by looking for history entries that existed before today, since those are substantially more work to inject, enough so that the "attack" (such as it is) is unlikely to occur.

Hence, the intended grouping (and maybe the low bar for a string recommendation here) is:

   "Have I visited this website (before today)?"

We don't want to drop the temporal reference ("Have I visited this website before?" or "Have I visited this website in the past?") because those will make our behaviour seem buggy on newly visited sites, where visits won't immediately show up.

Other suggestions are welcome.

Comment 2

[Michael Ventnor]

"Have I visited this web site prior to today?"

Comment 4

[Axel Hecht [:Pike]]

Can we get this fixed in 3.7? Sounds like a nice bug to get started.

Comment 5

[Kadir Topal [:atopal]]

Right, I'll take it :)

Comment 6

[Kadir Topal [:atopal]]

Attachment: Michael's wording

I took Michaels wording, since it's clear and doesn't need parentheses.

BTW: my first patch to the product after 7 years in the project :)

Comment 7

[Johnathan Nightingale [:johnath]]

Comment on attachment 419784 [details] [diff] [review]
Michael's wording

>diff -r e9f33e85a979 browser/locales/en-US/chrome/browser/pageInfo.dtd
> <!ENTITY  securityView.privacy.header                   "Privacy &amp; History">
>-<!ENTITY  securityView.privacy.history                  "Have I visited this web site before today?">
>+<!ENTITY  securityView.privacy.history2                 "Have I visited this web site prior to today?">
> <!ENTITY  securityView.privacy.cookies                  "Is this web site storing information (cookies) on my computer?">

Rev'ng the entity here is sort of a fine line, we're not changing the semantic meaning so much as clarifying it, but I guess it will likely cause the other english-based locales to update themselves. If you *do* rev the entity though, then you need to also update the source code anywhere that entity is used, so that it can resolve the entity once again. 

Does that make sense?

MXR says you've only got one place to update anyhow: http://mxr.mozilla.org/mozilla-central/search?string=securityView.privacy.history

Comment 8

[Axel Hecht [:Pike]]

Yea, I wouldn't rev the entity here.

Comment 9

[Kadir Topal [:atopal]]

Attachment: Patch without rev change

Wow, I screwed up a one liner! 

Hope it works better this time ;)

Comment 10

[Kadir Topal [:atopal]]

Attachment: Patch without rev change AND correct indentation

Well, third  time  is  a  charm, or something :)

Comment 11

[Dão Gottwald [:dao]]

http://hg.mozilla.org/mozilla-central/rev/6eac956960a0

Comment 12

[Tanner M. Young [:tmyoung]]

Verified Fixed.

Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.3a1pre) Gecko/20100120 Minefield/3.7a1pre (.NET CLR 3.5.30729) ID:20100120043108