[If a library building is open for all but a few hours, but contains
servers, it is bad practice for cleaning staff to shutoff building power
when finishing their shifts. Not that this has ever happened, especially
not at UNC]

There are some generalizations in the data shown on your blog post that
might be able to give even tighter results.

It looks like many locations / services have opening hours that are related
to other locations and services.

In some cases, the relationship is necessary: if a physical  service is
provided from a location that is entirely contained within another
location, then the hours for the first service must be contained within the
hours for the host location.

Multiple libraries of the same general class may have similar hours.

Variations from normal hours may be driven by common factors;  if the
university closes at 3pm, all libraries may close at 3pm. Violations of
expectations are usually highly salient e.g. changes from normal hours
(unless one expects hours to change and they don't ).

If a nonpublic computer is associated with catalog searches and views that
predominantly relate to holdings in a few locations, the hours of those
locations is probably salient.

Showing opening hours for today/tomorrow when displaying records may be
useful; showing opening hours for top locations for items shown on a search
screen may also be useful.  If location is a facet then this is an easy
feature to add.

On Nov 27, 2013 9:25 AM, "Sean Hannan" <[log in to unmask]> wrote:

> I¹d argue that library hours are nothing but edge cases.
> Staying open past midnight is actually a common one. But how do you deal
> with multiple library locations? Multiple service points at multiple
> library locations? Service points that are Œby appointment only¹ during
> certain days/weeks/months of the year? Physical service points that are
> under renovation (and therefore closed) but their service is being carried
> out from another location?
> When you have these edge cases sorted out, how do you display it to users
> in a way that makes any kind of sense? How do you get beyond shoehorning
> this massive amount of data into outmoded visual paradigms into something
> that is easily scanned and processed by users? How do you make this data
> visualization work on tablets and phones?
> The data side of calendaring is one thing (and for as standard and
> developed as the are, iCal and Google Calendar¹s data formats don¹t get it
> 100% correct as far as I¹m concerned). Designing the interaction is wholly
> another.
> It took me a good two or three weeks to design the interaction for our new
> hours page ( over the summer. There
> were lots of iterations, lots of feedback, lots of user testing. ³User
> testing? Just for an hours page?² Yes. It¹s one of our most highly sought
> pieces of information on our website (and yours too, probably). Getting it
> right pays off dividends.
> I don¹t know if you¹d find it useful (our use cases are not necessarily
> your use cases), but I ended up writing up the whole process as a blog
> post
> (
> ).
> -Sean
> ‹
> Sean Hannan
> Senior Web Developer
> Sheridan Libraries
> Johns Hopkins University
> On 11/26/13, 6:41 PM, "Barnes, Hugh" <[log in to unmask]> wrote:
> >Great edge case, thanks for sharing that one!
> >
> >I think currently that could only be _encoded_ as a separate opening in
> >the CSV file for loading into the database, which won't work because of
> >my assumption. There simply isn't a way to express it. The relevant
> >fields for the load file are startdate, enddate, opentime, and closetime,
> >the last two being formatted as only "hh:mm", so it's assumed they relate
> >to each single day in the range.
> >
> >However, I edited a "closes" field value directly in the test database,
> >and to my surprise it rendered sensibly. I would have thought it would be
> >rejected by a validity test I have which checks that the day portion of
> >the start and closing datestamps are the same [1].
> >
> >I can't justify spending time on this in the near future, since it's a
> >use case we are unlikely to need here. However, I'll log an issue, or you
> >may. Thanks again.
> >
> >Cheers
> >Hugh
> >
> >[1]
> >
> >-----Original Message-----
> >From: Code for Libraries [mailto:[log in to unmask]] On Behalf Of
> >Bohyun Kim
> >Sent: Wednesday, 27 November 2013 11:28 a.m.
> >To: [log in to unmask]
> >Subject: Re: [CODE4LIB] calibr: a simple opening hours calendar
> >
> >Hugh,
> >
> >Thanks for sharing. A quick question. If a library opens past midnight,
> >does that count more than one opening a day or no?
> >
> >~Bohyun
> >
> >
> >On Nov 26, 2013, at 5:04 PM, "Barnes, Hugh" <[log in to unmask]>
> >wrote:
> >
> >> Hi folks
> >>
> >> I took a calendar script posted to this list by Andrew Darby some time
> >>ago and made some changes. I don't think there is any of Andrew's code
> >>left, so I've rebranded it with an acknowledgement. (If I had my time
> >>again, I might have coded it from scratch rather than built it over
> >>Andrew's script, but that's somewhat academic.)
> >>
> >> The whole scoop is in the readme on Github:
> >>
> >>
> >> TLDR: With PHP, MySQL, some fiddling and data entry, you can publish a
> >>library opening hours calendar on your website in more than one language
> >>if you wish. It's a little quicker to enter common period patterns than
> >>it used to be in Google Calendar. The output is more accessible,
> >>customisable, multilingual, semantic, and hopefully more extensible
> >>(iCal etc) than previously.
> >>
> >> Here's a branded reference implementation:
> >> - it won't necessarily reflect the
> >>latest version.
> >>
> >> Use it, improve it, feed back, or log issues right there on Github if
> >>that works for you.
> >>
> >> Many thanks to Andrew for providing the foundation!
> >>
> >> Cheers
> >>
> >> Hugh Barnes
> >> Digital Access Coordinator
> >> Library, Teaching and Learning
> >> Lincoln University
> >> Christchurch
> >> New Zealand
> >> p +64 3 423 0357
> >>
> >>
> >>
> >> ________________________________
> >> P Please consider the environment before you print this email.
> >> "The contents of this e-mail (including any attachments) may be
> >>confidential and/or subject to copyright. Any unauthorised use,
> >>distribution, or copying of the contents is expressly prohibited. If you
> >>have received this e-mail in error, please advise the sender by return
> >>e-mail or telephone and then delete this e-mail together with all
> >>attachments from your system."