I'm open to suggestions, Ya'aqov.
I've been talking up the idea of some sort of dashboard for our services. Display uptime and response time. It will be tougher to automatically detect a database update and report it. I'll give that some thought for stuff running over my software stack.
This seems like the right forum to solicit other suggestions. Has anyone done this before? It seems like there ought to be some lists lying around somewhere of information that would be helpful to service consumers.
Ralph
-----Original Message-----
From: Code for Libraries [mailto:[log in to unmask]] On Behalf Of Ziso, Ya'aqov
Sent: Sunday, March 21, 2010 1:09 PM
To: [log in to unmask]
Subject: Re: [CODE4LIB] WorldCat Terminologies
I'm certain that as Ralph indicated, this file has been kept weekly up-to-date.
The html page header will be, eventually, fixed as well to reflect accurately the file's last update and its SRU searchability. The fact remains that for all: terminologies/identities/xISSN/xISBN >> WC-DEVNET is the customer support and quality control.
We have no other address for maintenance, and possibly OCLC Research's dedicated staff lack such address as well. Yes, these experimental services reside on OCLC servers.
Unfortunately, given this customer support model, OCLC Research will be constantly put in a defensive position and all we can do is flag problems and maintain this loop.
(unless any of you has an idea for a loophole and, please, bring it on!)
Ya'aqov
-----Original Message-----
From: Code for Libraries on behalf of Jonathan Rochkind
Sent: Sun 3/21/2010 12:19 AM
To: [log in to unmask]
Subject: Re: [CODE4LIB] WorldCat Terminologies
Yeah, the statement that it's a static copy from 2006 would have stopped me in my tracks if I had somehow happened accross the page, which I probably wouldn't have, but now I've bookmarked it so I might find it again -- but will probably forget that it's REALLY up to date even though it says 2006 on it. Nice catch Karen.
Karen, that looks to me like an HTML front-end for an SRU service, I bet it's got an SRU api. Which one of these days I'll get around to figuring out how to write code for.
________________________________________
From: Code for Libraries [[log in to unmask]] On Behalf Of Karen Coyle [[log in to unmask]]
Sent: Saturday, March 20, 2010 11:29 AM
To: [log in to unmask]
Subject: Re: [CODE4LIB] WorldCat Terminologies
Quoting "LeVan,Ralph" <[log in to unmask]>:
> I hate to muddy the waters, but I can't resist here.
>
> Research also exposes a copy of the LC NAF at
> http://alcme.oclc.org/srw/search/lcnaf
>
> It gets updated every Tuesday night.
Unfortunately, that page states right up front:
"A static copy of LC's Name Authority File from February of 2006"
That might confuse visitors. Maybe a quick revision is in order? :-)
Also, API access?
kc
>
> This is something I've been maintaining for years and is what
> Identities points at when you ask to see the NAF record associated
> with an Identities record.
>
> This particular service has none of the linked-data-type bells and
> whistles I'm putting into VIAF and Identities, but easily could, if
> there was interest. I believe I've made the indexing on it
> consistent with what I do in Identities.
>
> Looking at the configuration file for the load of this database, I
> am omitting records with 100$k, 100$t, 100$v, 100$x or any 130
> fields. I'm sure Ya'aqov (or other similarly expert Authority
> Librarian) could tell you why I am omitting them, because I can't
> off the top of my head.
>
> This service is actually running as a long established model of how
> similar services should run in Research. While it is not running on
> a machine operated by our production staff, it is automatically
> monitored by them, they have restart procedures in places when the
> service becomes unresponsive and problems are escalated by email
> when the restart fails to fix the problem. (Those emails come to me
> and where they get treated appropriately.)
>
> Let me know if there are questions about any of this.
>
> Ralph
>
>> -----Original Message-----
>> From: Code for Libraries [mailto:[log in to unmask]] On Behalf Of
>> Ya'aqov Ziso
>> Sent: Friday, March 19, 2010 3:29 PM
>> To: [log in to unmask]
>> Subject: Re: [CODE4LIB] WorldCat Terminologies
>>
>> Jonathan, thank you, in full accord. Yes, the crux of the matter is Names
>> (NAF being the more expensive library subscription
>> and the one not available for free like http://id.loc.gov
>>
>> At http://orlabs.oclc.org/Identities/ I searched Oclapton, ericš and wonder:
>>
>> * are all WorldCat+NAF 49 retrieved headings there helpful? do we need in
>> the list Oclayton, cecileš (etc.)
>> * Names that havenšt made it into an authority record are definitely
>> helpful, but can we suggest a way to sort and rank them more usefully (for
>> the user) on the page?
>>
>> Your thoughts?
>> Yašaqov
>>
>>
>>
>>
>>
>>
>> On 3/19/10 2:35 PM, "Jonathan Rochkind" <[log in to unmask]> wrote:
>>
>> > I don't think the inclusion of non-NAF headings in Identities is
>> a flaw, it's
>> > a benefit to the purpose of Identities not to be held back by the somewhat
>> > glacial pace of change in NAF. But you're right, the right tool
>> for the job,
>> > I don't know that any of the existing OCLC "free" (or included
>> with other OCLC
>> > membership/services) services are the right tool to replace any existing
>> > purchased authorities tools or sources. It depends on what you're
>> using them
>> > for, of course. I agree that the brochure statement was potentially
>> > misleading, but these (Identities, Terminologies, Research
>> Terminologies) are
>> > still very interesting and useful services.
>> > ________________________________________
>> > From: Code for Libraries [[log in to unmask]] On Behalf Of
>> Ya'aqov Ziso
>> > [[log in to unmask]]
>> > Sent: Friday, March 19, 2010 2:14 PM
>> > To: [log in to unmask]
>> > Subject: Re: [CODE4LIB] WorldCat Terminologies
>> >
>> > Karen,
>> > Seems like pulling-teeth was worth it. Thank you for these updates and for
>> > making them available for all interested.
>> > Essentially, given your 6 months latency compared to
>> http://id.loc.gov) and
>> > the inclusion of NAF and non-NAF headings in Identities,
>> > both Terminologies and Identities are not yet at a level of databases that
>> > libraries can go for, forgoing what they currently need to buy, load, and
>> > maintain locally.
>> > Your withdrawal of the brochure statement is justified and
>> apology accepted.
>> > Yašaqov
>> >
>> >
>> >
>> >
>> > On 3/19/10 1:35 PM, "Karen Coombs" <[log in to unmask]> wrote:
>> >
>> >> > Yašaqov,
>> >> >
>> >> > Identities is not based on a name authority file it is based
>> on name data
>> >> in
>> >> > WorldCat. These two are not the same thing. Names within
>> Identities come
>> >> from
>> >> > several different fields within the WorldCat MARC records
>> including 1xx,
>> >> 6xx,
>> >> > and 7xx fields. This is why Identities contains names which
>> have not made
>> >> > their way into a name authority file
>> >> > (http://www.worldcat.org/identities/np-levan,%20ralph%20r).
>> Also Identities
>> >> > doesnšt contain authority records, it contains Identity
>> records. Identity
>> >> > records contain different information from Authority records.
>> >> >
>> >> > There are name authority projects going on within OCLC Research. The
>> most
>> >> > active is VIAF http://www.viaf.org/ . This service contains
>> name authority
>> >> > information from several national libraries, not just Library
>> of Congress
>> >> and
>> >> > right now not all of NACO has been loaded, only differentiate personal
>> >> names.
>> >> > More information on the VIAF project is available at
>> >> > http://www.oclc.org/research/activities/viaf/
>> >> >
>> >> > Regarding up to dateness of Terminologies as I said in my previous
>> message
>> a
>> >> > schedule for updates was posted to the Developer Network
>> listserv and the
>> >> > Developer Network wiki
>> >> > (http://worldcat.org/devnet/wiki/Terminologies_Updates).
>> Andrew Houghton
>> >>
>> has
>> >> > also informed you that information about how recently a given
>> terminology
>> >> is
>> >> > located on the Experimental Terminologies page (
>> >> > http://tspilot.oclc.org/resources/index.html).
>> >> >
>> >> > Karen
>
--
Karen Coyle
[log in to unmask] http://kcoyle.net
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet
|
