[uf-new] Currency brainstorming

[Andy Mabbett]

In message <46FD5A5F.8060409 at brixlogic.com>, Guillaume Lebleu
<gl at brixlogic.com> writes

>>> In Andy's example above, the date is not a property of the "hmoney"
>>>class, but of a TBD "hprice" class.
>>
>> I don't see any advantage in making such a distinction, nor any
>>problem in not doing so,
>>
>> Perhaps you could enlighten me, with real-world published examples?
>>
>The advantage is less confusion.
>
>If I'm a publisher and I need to mark up POSH the following "The Euro
>stood at 1.41 US dollars in September 2007",
>
>We don't want publishers to mark up:
>
>"The Euro stood at
><span class="money">1.41
>   <span class="currency" title="USD">US dollars</span>
>    in
>   <span class="date">September 2007</span>
></span>".

I'm not aware that anyone has suggested that that is what we want.

>Of course that is, unless you think the following is not better a
>better reflection of the intent of the author:
>
>"<span class="exchangerate">

[...]

Nor do I see any proposal for marking up exchange rates. This is a
proposal for marking up amounts of money in single currencies (which
could I agree, be combined into an exchange rate microformat at some
later date.

>I looked at most of the historic prices examples you presented:
>http://microformats.org/wiki/currency-examples#Historic_prices
>
>For what I can see, in most of these examples, the datetime near a
>money amount usually related to the relationship between a currency and
>another, or a currency and a commodity, not to the currency itself.

For some value of commodity:

        In 1950 the average weekly wage was X

        The Beatles first single sold for one shilling

        The last Monet painting to be auctioned fetched $95 million in
        2005

>When a datetime is nearby, it usually refers to the datetime of the
>posting, or of the general context in which the events described must
>be taken.

The latter is what's being discussed here.

>In other words, in most cases and example shown, the datetime of the
>money amount is implicit, inferred from the context, not explicit.

I don't follow.

> Theses date time should not be marked up as part of the money amount,
>but as part of a "price" class, "article" class, or "exchangerate"
>class.

There is no proposal or such classes; if you are making such a proposal,
then I'm afraid your reasoning is still not clear to me; and neither is
your claimed reduction in confusion (in fact the reverse).

>sA datetime explicitly linked to a money amount is much rarer. The only
>case I can think of where a date is directly related to a currency is
>in "Damage in Bay County, Florida alone totaled US$50 million (1975
>dollars)" http://en.wikipedia.org/wiki/Hurricane_Eloise.
>
>In this example, you are correct, we should have:
>
>"Florida alone totaled <span class="hmoney"><span class="currency"
>title="USD">US$</span><span class="value" title="50000000">50
>million</pan> <span class="date">(1975 dollars)</span></span>"

"1975 dollars" is not a meaningful date (I'm not even sure it's
meaningful English).

>So, while I can find an example that support your feature suggestion, I
>believe the above example where the date of the currency is not
>implicit is rare enough to be left aside for now for the purpose of
>moving this proposal forward and avoiding confusion within publishers.

I don't believe it is rare; I could provide many further examples, but
they would just involve repetitions of the models already discovered.

-- 
Andy Mabbett