[uf-discuss] NetNewsWire ditches support for microformats
chris.messina at gmail.com
Sat Aug 16 09:41:40 PDT 2008
It's funny, since this was posted to the list June 5 of last year
(NetNewsWire 3 supports microformats):
But now, turns out that future versions of NNW will no longer support
Judging from discussions around this decision (on a private mailing
list), there seems to be little real value from this feature in
NetNewsWire, a very popular feed reader for the Mac.
On the one hand, it's hard to argue with the detractors who point out
that there's actually little microformatted content in feeds (NNW only
detects hcalendar and hcard -- the most widely deployed microformats)
and that even when such content shows up, adding hcards to your
address book, or clicking the Add to Calendar button provided by NNW
is hardly beneficial when feeds that have an iCal equivalent either
tack it on as an enclosure or link off to a remotely hosted .ics
file... basically making the embedded hcalendar redundant (though
arguably still best practice).
I have a lot of respect for Brent Simmons -- and his choice to
*remove* microformats after he'd been convinced (and had decided) to
add them really concerns me. To the best of my knowledge, there are
few if any other aggregators that support microformats consumption --
and fewer still that expose an end-user interface  for converting
microformatted content on the client side.
This makes me reconsider some things.
* Are microformats valuable in client-side applications?
* Should their presence be made known to end-users? Where there is
data that can be converted, should an interface be provided to do so?
* Does the slight overhead/performance hit that parsing HTML to look
for microformatted content in feeds outweigh their benefits
(performance was cited as one gain from removing the parsing code)
* Is parsing microformats, given their fluid nature and lack of
concrete specificity (compared with more machine-friendly formats like
JSON or XML), too burdensome for developers who have the choice of
alternative, machine-friendly formats? Do the availability of parsing
libraries allay this burden? Where we lack libraries in certain
languages, would the existence of such libraries help to promote wider
* Have we actually identified valuable use cases for microformats in
* Is adding hcards to address books and hcalendars to calendars really
the best possible value of adding microformats parsing to an
application? If not, what other application has demonstrated
additional, or different, value?
For my own part, I think that Apple's data detectors concept  is
probably the most compelling example where embedded microformatted
content could be useful -- at least in terms of reducing ambiguity or
increasing accuracy (if an ADD picks up the term "yesterday" without
context, an embedded microformat that follows that date-pattern could
provide absolute time values). Additionally, we see Microsoft flirting
with support for microformats with the addition of "hSlice" detection
in IE8... so there is, conceptually, hope -- although hSlice sadly
didn't come out of this community, but one company's bastardization of
I also see benefits for improvements for client-side search... and
this is something that I'd like to propose to Brent... and that I had
intended for Flock : Apple's Spotlight is able to pull out data
from documents by file type, rather than object type. Microformats
like hcard and hcalendar provide hooks that would allow client-side
search to differentiate objects, or to allow you to divide and
conquer, as it were, for example, if you were searching for a
person... you could search by author, by name, by address, etc... and
locate related biographical information were the data formatted as an
Anyway, the news got me noodling, and I'm curious what other folks think.
Open Source Advocate-at-Large
factoryjoe.com # diso-project.org
citizenagency.com # vidoop.com
This email is: [ ] bloggable [X] ask first [ ] private
More information about the microformats-discuss