[whatwg] Private browsing vs. Storage and Databases

[Jeremy Orlow]

I haven't decided for sure yet, but I was leaning towards either option #2
or option #3 for Chrome.  Option 5 seems like it'll be very confusing to
apps.  It's possible it'll even have undesired consequences like websites
popping up alerts or telling the user "you need to increase your quota" and
not allowing them to use the app until they do.  Web developers already need
to handle localStorage not being part of the DOM, localStorage having a 0
sized quota, and hitting the quota while running.  Why also make them assume
that localStorage could be populated and read-only?

Another question is whether a user in a "private" browsing session would
want an app to have access to existing storage at all.  I can see arguments
both ways.

The most important thing, however, is to be consistent.  For example, if
application cache was not read only but database was, then any application
'upgrading' its version in cache would fail while trying to upgrade the
database schema.  Even when "schemas" are implicit (like in localStorage)
this can be an issue.  Michael might have more to say on this.

I also agree that it'd be ideal if all the browsers responded similarly.  If
every browser implements it differently, it's much less likely that
developers will properly test for this.

I've inlined a couple additional comments...

Jeremy


On Tue, Apr 7, 2009 at 5:24 PM, Brady Eidson <beidson at apple.com> wrote:

> A commonly added feature in browsers these days is "private browsing mode"
> where the intention is that the user's browsing session leaves no footprint
> on their machine.  Cookies, cache files, history, and other data that the
> browser would normally store to disk are not updated during these private
> browsing sessions.
>
> This concept is at odds with allowing pages to store data on the user's
> machine as allowed by LocalStorage and Databases.  Surly persistent changes
> during a private browsing session shouldn't be written to the user's disk as
> that would violate the intention of a private browsing session.
>
> Let's take the specific case of LocalStorage, which is what I am currently
> working on with WebKit.  In attempting to fix this bug I came up with a few
> possible solutions:
>
> 1 - Disable LocalStorage completely when private browsing is on.  Remove it
> from the DOM completely.
> 2 - Disable LocalStorage mostly when private browsing is on.  It exists at
> window.localStorage, but is empty and has a 0-quota.
> 3 - Slide a "fake" LocalStorage object in when private browsing is enabled.
>  It starts empty, changes to it are successful, but it is never written to
> disk.  When private browsing is disabled, all changes to the private
> browsing proxy are thrown out.
> 4 - Cover the real LocalStorage object with a private browsing layer.  It
> starts with all previously stored contents.  Any changes to it are pretended
> to occur, but are never written to disk.  When private browsing is disabled,
> all items revert to the state they were in when private browsing was enabled
> and writing changes to disk is re-enabled.
> 5 - Treat LocalStorage as read-only when private browsing is on.  It
> exists, and all previously stored contents can be retrieved.  Any attempt to
> setItem(), removeItem(), or clear() fail.
>
> Option 1 is simple but painful for applications to get such different
> behavior.
> Option 2 is only a little more complicated, but also seems unsatisfactory.
> Option 3 is simple to implement and option 4 would difficult to implement
> efficiently.


Re #4: Most sites will not be given much space and you could implement
copy-on-write, so I guess I only see this being an issue in a small minority
of cases.


> Both would lead to bizarre behavior where data that the application thought
> was saved really wasn't.


This matches up with how most private browsing sessions handle cookies,
right?  The data persists until the session is up (because some of the web
can't work correctly without them) but then they're deleted at the end.

For now we're going with option 5.  setItem() during private browsing will
> fail with the QUOTA_EXCEEDED_ERR the spec mentions.  removeItem() and
> clear() will silently fail, since the spec assumes they always succeed and
> doesn't provide a failure mechanism.
>
> It seems the same issues apply to all the storage mechanisms, be it
> LocalStorage, SessionStorage (with optional session resuming), and
> Databases.
> I have a few questions I think it would be wise for the spec to address for
> all of these:
> 1 - What *should* the specified behavior be?
> 2 - If read-only ends up being the specified behavior, should we have a
> mechanism for removeItem() and clear() to demonstrate that they failed?
>
> Thanks,
> ~Brady
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.whatwg.org/pipermail/whatwg-whatwg.org/attachments/20090407/3fdc7974/attachment.htm>