Adds browser preferences to configure first-run flow on Windows 8.

chrome/browser/ui/browser_ui_prefs.cc

Index: chrome/browser/ui/browser_ui_prefs.cc
diff --git a/chrome/browser/ui/browser_ui_prefs.cc b/chrome/browser/ui/browser_ui_prefs.cc
index 502bb18615b00828c675be78493155811024838f..79028e9a64c88faf93af0a0bc8584e9c987e3cd9 100644
--- a/chrome/browser/ui/browser_ui_prefs.cc
+++ b/chrome/browser/ui/browser_ui_prefs.cc
@@ -89,9 +89,13 @@ void RegisterBrowserUserPrefs(PrefService* prefs) {
   prefs->RegisterBooleanPref(prefs::kCheckDefaultBrowser,
                              true,
                              PrefService::UNSYNCABLE_PREF);
-  prefs->RegisterBooleanPref(prefs::kDefaultBrowserFlowDialog,
-                             true,
+#if defined(OS_WIN)
+  // As with Mac-spacific code above, it should be in a platform-specific
+  // section somewhere, but there is no good place for it.
+  prefs->RegisterBooleanPref(prefs::kSuppressSwitchToMetroModeOnSetDefault,

[gab, 2012/07/24 17:43:16]

Double-negation in settings usually leads to confusion.

I would prefer: kSwitchToMetroModeOnsetDefault and have it default to true.

[motek., 2012/07/25 10:08:58]

While I agree with the principle, I have to disagree with this suggestion in
this specific context. This is about some default behavior which itself is
triggered in some unusual circumstances (as in: most of the time nobody cares). 
We want to alter this behaviour for some reason, this must be a deliberate
action. Thus, the default of 'change the default behaviour' would have to be
true. 
I feel this would be as confusing (or as understandable) as what I propose. In
fact, personally I feel better about 'suppress' than about 'switch to metro' set
to true because it applies in very specific circumstances. 

On 2012/07/24 17:43:16, gab wrote:

[gab, 2012/07/25 16:39:55]

Ok, maybe I'm not clear on the "very specific circumstances", I was under the
impression this was going to be a user option in settings (and even if it isn't
now, someone could definitely turn it into one later..).

I don't see how it being a very specific setting (i.e. barely ever turned on (on
as in disabled... see how confusing it is as "on" has the opposite meaning for
the pref and what it triggers...)) is a good argument against having a positive
property always set to true (in fact if you look in browser_ui_prefs.cc where
defaults are set, there are much more options that default to true than to
false).

[gab, 2012/07/26 14:52:08]

Discussed offline, this is a browser pref, not a setting, this all makes sense
now :), please do use "suppress_..."!

+                             false,
                              PrefService::UNSYNCABLE_PREF);
+#endif
   prefs->RegisterBooleanPref(prefs::kShowOmniboxSearchHint,
                              true,
                              PrefService::UNSYNCABLE_PREF);