[Design] [Last call] Bug 5403: Re-importing/Re-subscribing to a Trashed item

[Alec Flett]

Mimi Yin wrote:
> 2. Is there a way for us to over-write the "version" of the item 
> that's in the Trash with the "version" of the Item that's coming in on 
> re-Import/re-Subscribe? In other words, the Item in the Trash has a 
> bunch of stuff on it that says what SmartCollections it was a member 
> of before it was Trashed. the Item in the re-Import/re-Subscribe doesn't?
>
This is a tricky case because it means we have to distinguish between 
"overwriting" an item that's in the trash, vs. "overwriting" an item 
that's in another collection.

Personally I think all of this is totally feasible, and I think it would 
be nice to keep items from magically reappearing after they have been 
trashed, but the more special cases we have, the more bugs it will 
expose or create in other places where the trash comes into play. We 
just have to decide if this is a case where improving the experience for 
this case is worth potentially strange behavior (bugs) in other 
trash-related cases.

Here's my vote: we start with the simpler option that morgen proposed - 
that when something comes out of the trash, that it gets restored to all 
collections (since that has very few edge cases in the code, and much of 
that is already implemented) and see if the magically-reappearing-items 
issue actually bothers people? Personally this is the kind of thing that 
I could imagine shipping even 1.0 with and waiting until we have a wider 
audience to complain about it...:)

Alec

> Could we essentially Delete the Trashed version and start fresh with 
> the Import/Subscribe version? I think that's the correct user mental 
> model.
>
> Mimi :o)
>
> On Apr 5, 2006, at 7:48 PM, Morgen Sagen wrote:
>
>> Well, it's complicated, but follow me on this:  :-)
>>
>> Removing an item from a SmartCollection places that item in the 
>> Trash.  Items in the Trash don't appear in other SmartCollections.  
>> Moving an item out of the Trash causes the item to reappear in the 
>> SmartCollections it used to appear in.
>>
>> If we alter the behavior in the case of an import, and that import 
>> moves an item out of the Trash, we would need to explicitly *remove* 
>> that item from all other SmartCollections.  This means the item is 
>> now in each of those collections' Exclusions list.  If the user then 
>> moves that item to the Trash and later moves that item out of the 
>> Trash, that item will *not* appear in the original collections, since 
>> the item is in their Exclusions lists.  As a result, the end result 
>> of moving an item out of the Trash is inconsistent, and depends on 
>> whether an item happened to have been 
>> 're-imported-while-in-the-Trash' at some point in the past.
>>
>> I suggest we leave the behavior as it is, which is consistent: any 
>> time an item leaves the Trash, it re-joins its previous collections.
>>
>> ~morgen
>>
>>
>> On Apr 5, 2006, at 5:44 AM, Mimi Yin wrote:
>>
>>> Morgen, do you have any issues with going with this proposal? We 
>>> would want the behavior to be consistent for both re-Subscribe and 
>>> re-Import.
>>>
>>> Please see original write-up for more background: 
>>> http://lists.osafoundation.org/pipermail/design/2006-March/004358.html
>>>
>>> On Mar 15, 2006, at 5:10 PM, Jeffrey Harris wrote:
>>>
>>>> Hi Mimi,
>>>>
>>>>> So I think the right thing to do is to have the Import and Subscribe
>>>>> behave in accordance with Jeffrey's A proposal, which corresponds to
>>>>> Proposal #2 above: Remove the item from the Trash, but don't 
>>>>> restore it
>>>>> to all the collections it used to be a member of.
>>>>>
>>>>> And stick with the B proposal, which corresponds to Proposal #3 
>>>>> above:
>>>>> Remove the item from the Trash and restore it to all the 
>>>>> collections it
>>>>> used to be a member of, for when users explicitly remove an item from
>>>>> the Trash or undo Trashing an item.
>>>>
>>>> Seems reasonable to me.  It does mean a bit more work for the sharing
>>>> layer, though.
>>>>
>>>> Sincerely,
>>>> Jeffrey
>>>> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
>>>>
>>>> Open Source Applications Foundation "Design" mailing list
>>>> http://lists.osafoundation.org/mailman/listinfo/design
>>>
>>> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
>>>
>>> Open Source Applications Foundation "Design" mailing list
>>> http://lists.osafoundation.org/mailman/listinfo/design
>>
>> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
>>
>> Open Source Applications Foundation "Design" mailing list
>> http://lists.osafoundation.org/mailman/listinfo/design
>
> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
>
> Open Source Applications Foundation "Design" mailing list
> http://lists.osafoundation.org/mailman/listinfo/design