[Foundation-l] Personal Image Filter results announced

[Thomas Morton]

On 7 Sep 2011, at 23:04, MZMcBride <z at mzmcbride.com> wrote:

> Thomas Morton wrote:
>> This is largely an engineering problem; and it can probably be overcome with
>> some architecture work. As we are going to be implementing a major new
>> feature *anyway* it's not something to reject outright, I think :)
>>
>> Obviously given the complexity of the category tree system any such
>> engineering wouldn't be infallible - but you could match it to most use
>> cases. Ultimately it is just a collapsing tree problem, and they are ten a
>> penny to a decent engineer :)
>
> I think some of your comments are exhibiting an unfamiliarity with the
> tangled mess that is MediaWiki/Wikipedia. Have you done much work on
> MediaWiki or worked with the replicated databases at all (particularly the
> databases of the larger sites)? An outside voice is great, but yours comes
> off as rather naïve.
>
> MZMcBride

I've not proposed any actual solutions, or changes etc. Simply said
that the problem raised is an engineering problem and so needs to be
considered from that perspective.

>From my off hand knowledge of MW it won't be particularly easy - but
as one of my professors used to say "nothing is easy, but someone will
be able to fix it"

The next step is to figure out what engineering would be needed to
provide these features and whether that is possible

Had anyone seriously assessed this? (and if the answer is yes, then fine)

Tom

(I tend to hold a positive attitude to such problems until they are
solved or shown insoluble; everyone tells me my proposed solutions at
work are "impossible" but they work out more often than not!)