[uf-discuss] Species microformat process

[Charles Roper]

On 30/01/07, Benjamin West <bewest at gmail.com> wrote:
> On 1/30/07, Charles Roper <charles.roper at gmail.com> wrote:
> > I'm very interested in the Species microformat, but the process seems
> > to have stalled and I just wanted to poll opinion here as to why that
> > might be. Is it due to a lack of demand?

> To be honest, the
> use case for the species microformat is a little bit weak.  It could
> be that if there is a lack of demand, it is due to the weak use case
> and the gap between the research and the proposal.

OK, that is a decent criticism I feel. The use case, which is pretty
much encapsulated here: http://microformats.org/wiki/species#Proposal,
does sound a little vague to me too. There is clearly a failure on the
part of us proponents of species to communicate the uses of it
convincingly. Perhaps fleshing out use cases and clarifying the
importance of semantically marking up taxonomic names would be a good
next action?

>(The use case
> essentially describes a hyperlinking behaviour that is already present
> and used on many sites.) This is just my own opinion though.

I'm not sure that it does. There is an element of tagging; i.e.,
referencing an authoritative resource. But, just as importantly, there
is the parsing and collation use. Being able to extract, collate and
analyse taxonomic names from web pages is a tremendously compelling
concept, especially when married with married with who, when and where
data (the species being the "what"). These are the basic elements that
constitute a "biological record."

> > It seems that the successful
> > microformats have been developed, in the main, by web designers and
> > developers for web designers and developers. Could it be that web
> > designers and developers of the microformats community do not perceive
> > the value of a species microformat in the same way that they can see
> > the value of, say, hCard, hReview, XFN, etc. The more successful
> > microformats seem to be riding on the back of the "social web"
> > zeitgeist, with many (most?) being used in this kind of context. I
> > don't see species as being of particular interest to the bloggers and
> > the other social-networking, mashup-making, digerati of current times.
> > Is appealing to this demographic the key in getting a microformat
> > developed? I'd appreciate the view of people in this community.
> >
>
> If I think I know what you mean here, I disagree a bit.  Who, what,
> when, how, where are all answered by these microformats.  They are
> extremely common, and are probably the first data types to be
> represented by any new technology.  For example: how old is the
> calendar compared to taxonomy? hcard describes "who", xfn describes
> "how are they related", hcalendar describes "when".... Again, this is
> my own opinion, and I don't have any evidence.

I think what I'm trying to get at here is this: at present my feeling
(and this is just a feeling) is that the microfomats gaining most
traction are ones that are related quite closely to social networking
and, dare I say it, Web 2.0 concepts. I'm not saying that's a bad
thing, but I do feel that because there's not a lot of interest in
species and related matters within these circles (the interest lies in
technology, modern media and networking), that trying to develop a
species microformat at this point in time is going to be an uphill
struggle, simply because there aren't many people out there who would
take advantage of such a thing right now.

> > I also wanted to ask about the fundamental microformat principle of
> > "paving the cowpaths" in relation to hCard. It seems to me that hCard
> > was derived from vCard rather than being based on existing markup
> > practice. How does this square up with the cowpaths philosophy?
> >
> I'll leave this to others.  However, I do think the kind of path vcard
> took is qualitatively different than the one species is taking.

I'm not really questioning the quality of the two approaches, but
rather the fundamental uF principle of "paving the cowpaths." As far
as I am aware (but please correct me if I am wrong), hCard is an
extraction of vCard rather than an extraction of existing markup
practice. This being the case, can other microformats, in the absence
of existing practice, not be developed on the back of existing,
canonical standards?

> > This brings me to a question about Species. The Species proposal
> > doesn't really reflect current mark-up practice but instead represents
> > what might be a good way of doing things in the future if authors were
> > to start using it.
>
> I agree completely, and I think this is a problem with the current
> effort, but it's one that is solveable.

If this *is* a problem, why wasn't it a problem with hCard? I'm not
trying to get at anyone here, I'm just trying to wrap my head around
this apparent anomaly.

> > The vocabulary in the proposal isn't plucked out of
> > thin-air, though; it is taken from the taxonomic hierarchy as used by
> > biologists. It seems to be modelled on hCard in this respect, hence my
> > cowpaths question. My own feeling is that the current proposal is too
> > complex. The current usage patterns as far as I can see (in the
> > majority of cases) either have species names as plain text or
> > marked-up with simple <strong> tags, or <em> or <i>. However, I'm not
> > adverse to having a rich vocabulary of class names to call on should I
> > need them (which 9 times out of 10 I won't), as long as a species name
> > can still be marked-up very simply. This is similar to the way in
> > which hCard has a rich vocabulary, but can still be very simple.
> >
>
> Charles, this is a great analysis.  This matches my observings, at
> <http://microformats.org/wiki/species-examples-regrouped>.  I think
> the format would benefit greatly from taking a fresh look at the
> examples collected.  It look to me like there should be two pieces.
> One should be the way authors mention species in text, the other
> should be how authoritative sources represent the information about a
> species.  In any case, the first behaviour can be accomplished,
> perhaps entirely, by using tagging techniques.  Perhaps the in-depth
> use of unambiguous names can be used by a second format intended for
> publishers of the authoritative information.

Given the problems associated with the naming of species, there
certainly needs to be a way of linking a species name with an
authority in order to clarify ambiguity, link synonymy and otherwise
extract other taxonomic intelligence. But authors aren't going to want
to make a link out of every mention of a species name, although they
might well want to mark up that name as being a species. It looks as
if the LSID could well become the URI of choice for linking names to
authorities, but these, when placed in an href don't resolve in
current browsers. So the question is, how could LSIDs work with
tagging?

Many thanks for the feedback, it's much appreciated.

Cheers,

Charles
-- 
Charles Roper
www.sxbrc.org.uk