Not really all that basic; IMHO, a good question.
My short answer to the question "are [there] reasonable objections to
use ISO8601" is "yes", there are.
The primary objection is that there is little to no readily available
software out there in the world that will read an arbitrary ISO 8601
formatted string and return a structure that you (the programmer) can
use. (BTW, I would *love* to learn that I'm wrong about this.) On the
other hand, there is reasonably good support for W3C format dates and
times in a variety of languages (including, most importantly, XPath;
and thus XSLT, XQuery, and Schematron). That support includes, e.g.,
comparison functions, time zone adjustment functions, and operators
Another objection to ISO 8601 is, at least back in 2005 when we paid
nearly $100 for a copy (of the 2004 version), you have to pay for it.
(Not to use it, but to get a copy of the complete spec, which is
protected by copyright.)
Another (somewhat minor) objection to ISO 8601 is the (absolutely
ridiculous) support for the construct "24:00" for midnight.
Last, an objection to ISO 8601 that is reflected in the original W3C
datatypes specification, but I've heard (from Martin Holmes) has been
corrected in at least one of those two places, but *not* in all
corresponding software, is that in 8601 as of 1998 the year "0000"
was not allowed. Apparently in the 2000 version it is allowed. In the
2004 spec (which I just scanned) "0000" is explicitly allowed.
However, I note that in 8601:2004 dates BCE are represented with '−'
(U+2212 MINUS SIGN), not '-' (U+002D HYPHEN-MINUS). Some will
consider that an objection, some will consider it an asset.
My rule of thumb is to use the W3C datatypes (e.g., @when, @dur,
@from, @to) when they are expressive enough to say what you want. If
your project requires more expressivity than W3C permits, use the ISO
format (i.e., @when-iso, @dur-iso, @from-iso). If even 8601 is not
expressive enough for your needs, use the LOC EDTF on the custom
attributes (e.g., @when-custom, @dur-custom). (See .)
The main exception to that rule of thumb is in precision of time. W3C
requires that times be expressed to at least the second. In many
humanities applications asserting that an expression of time of day
is precise to the second is at best silly, and at worst a lie. In
some cases I have recommended using ISO 8601; but in others I have
recommended using non-ISO 8601 methods of representing this mess.
<time when="12:30:00" precision="medium">half-past noon</time>
where internal project documentation explains that a precision of
"high" = precise to the second ("12:30:30" in ISO 8601)
"medium" = precise to the minute ("12:30" in ISO 8601)
"low" = precise to the hour ("12" in ISO 8601)
All that said, two thoughts about the examples you gave. E.g.
<time from-iso="P19M" to-iso="P49M31S" dur-iso="P30M31S">
* Why use both @from & @to *and* @dur? The duration can always be
calculated from the @from and @to, so it is redundant. In some
cases, that redundancy is a good thing, it gives you a way to
check for inconsistencies. In some applications that redundancy is
a bad thing, it provides a place for inconsistencies to creep in.
(BTW, it is perfectly valid, but much less common, to use @from &
@dur or @dur and @to instead.)
* Many of your durations are not valid, because the "T" designator
before components more precise than a day is required. I.e.,
"P19M" is a period of 19 months (not minutes), and "P49M31S" is
simply invalid. They should be "PT19M" and "PT49M31S",
respectively. (This is true whether you use ISO 8601 or W3C
Not sure I actually answered what you were asking, but I hope this
helps in any case.
> Encoding points in time (date) and periods of time
> This is – bluntly – basic stuff of a bloody beginner who's encoding points
> in time (date) and periods of time, wondering after perusal of
> whether there are reasonable objections to use ISO8601 and is it good
> practice simultaneously orientating fundamentally to
> <recording type="audio" dur-iso="P10M57S">
> Aufnahmedatum: <date when-iso="1959-11-01">01.11.1959</date>.
> Länge: <time dur-iso="P10M57S">00:10:57</time>.
> <cell role="label">Inhaltsangabe</cell>
> <cell role="label">Dauer</cell>
> <cell>Rede <persName ref="http://d-nb.info/gnd/118763784"
> role="publisher">Unseld</persName>: Begrüßung und Einführung / zur
> zeitgenössischen Literatur / Biographie und Werk von <persName
> ref="http://d-nb.info/gnd/118529374" type="author">Günter Eich</persName> /
> Biographie und Werk von <persName ref="http://d-nb.info/gnd/11855817X"
> role="reader">Uwe Johnson</persName></cell>
> <cell><time from-iso="P0M" to-iso="P18M40S"
> <cell>Lesung <persName ref="http://d-nb.info/gnd/11855817X"
> role="reader">Uwe Johnson</persName>: <title
> ref="http://d-nb.info/457105232" type="book">Mutmassungen über
> <cell><time from-iso="P19M" to-iso="P49M31S"
> Briefly: Is it clever common sense or have I simply overseen till now the
> much more clever alternatives also existing? And if so, can anyone help me
> finding them?