Tomasz Neugebauer wrote:
> Alex Johannesen asks:
>
> "Where are our open API's?
> Have we done anything towards a simpler meta data system since DC?
> What are we doing to integrate / merge it with MARC? Where's our
> services that allows re purposing of our data?"
>
> These are good questions.
> I think that libraries should be moving more towards open systems:
> exposing APIs
>
>
I'm sure Tomasz realizes this, but I want to make something clear: This
is a _different thing_ than 'open source'. Even if I have a
proprietary/commercial solution, I need that solution to offer APIs
which give access to the data and services that product houses. [And I
need those API calls to be reasonably speedy to access!]. Perhaps open
source applications are more likely for market reasons to have these
open APIs, but there's no technical reason a proprietary solution can't.
"and publishing (open access) the semantic rules and structures implicit in their metadata. I think that existing cataloguing rules (eg.: AACR2), could have a great impact if they were available online open access. Why aren't they?"
And this is yet another thing again, and refers to things that are (or
are treated as) the intellectual property of LC, or of ALA, or of the
AACR2/RDA "Committee of Principles", etc. It's not a technical problem.
Which doesn't mean it's not a problem. We need to get the owners or
custodians of these rules to realize the importance of open-ness, and
what it means.
We also need these custodians to realize what it really means to be in a
format that is machine useful. I heard recently a story of someone at
one organization suggesting that an ascii list of terms and definitions
in a basically fixed-width-column format (embedded in a <pre> tag on an
HTML page full of other text, to boot!) was "machine readable". Well,
sort of, but not very usefully. Any organization responsible for large
amounts of community metadata or controlled vocabularies or standards
should have a 'metadata engineer' (I just made that up) on staff, says
me. Someone who understands the technical requirements of making this
stuff available in a machine-useful way.
Jonathan
> Tomasz Neugebauer
>
>
>
> Systems Development Librarian
> tomasz.neugebauer_at_concordia.ca
> Concordia University Libraries
>
> -----Original Message-----
> From: Next generation catalogs for libraries
> [mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Janet Hill
> Sent: Friday, September 21, 2007 12:49 PM
> To: NGC4LIB_at_listserv.nd.edu
> Subject: Re: [NGC4LIB] As a Library 'decision maker'
>
> [mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Tomasz Neugebauer
>
> My humble opinion is that Libraries have been too reluctant to publish
> their authority records in a format that can be used for computer
> science research. Libraries have rich classification schemes and
> structured subject vocabularies, semantically rich catalogues; the
> problem is that they are too reluctant/slow to make these structures
> public, to share them with computer scientists, for example
> ---------
>
> If library types and systems types are to talk successfully to each
> other
> without taking umbrage, they need to be careful not to use language that
> can
> be interpreted as accusatory, because it's almost inevitable that if an
> insult CAN be read into text, it WILL BE. And a group that feels insult
> is
> less likely to cooperate.
>
> I believe the problem is not that libraries have been either "reluctant"
> or
> "too slow." Instead, library folk may have (mistakenly) assumed that
> people
> in other fields who are involved in endeavors related to information
> organization and access ought to have realized (or discovered) that
> there is
> an entire discipline that is about information organization and access,
> that
> their experience might be useful to consider, that relevant wheels have
> already been invented, etc.
>
> It is not the "fault" of LIS folk that they have not fully anticipated
> and
> understood the needs of future systems. e.g. it seems a bit harsh to
> "accuse" library types of being " ....reluctant to publish
> their authority records in a format that can be used for computer
> science research" when the format was devised decades ago for the
> systems
> that could at that point be imagined. Or at least, if LIS types are to
> be
> faulted for "being reluctant" we might at the same time fault IT types
> for
> (to the extent that they have) not discovering and/or "being reluctant"
> to
> seek out, and utilize data and formats already in existence or to
> cooperate
> with another field in working toward a standard better suited to the
> future
> that is NOW evolving.
>
> Note, that this message is not meant as a refutation of this particular
> posting. It's more a response that comes about from the accumulated
> impact
> of messages in which each group seems to be accusing the other of being
> "the
> problem," and which seems to result more in intransigence than in
> compromise/cooperation/understanding.
>
> Janet Swan Hill, Professor
> Associate Director for Technical Services
> University of Colorado Libraries, CB184
> Boulder, CO 80309
> janet.hill_at_colorado.edu
> *****
> Tradition is the handing-on of Fire, and not the worship of Ashes.
> - Gustave Mahler,
>
> -----Original Message-----
> From: Next generation catalogs for libraries
> Sent: Thursday, September 20, 2007 3:38 PM
> To: NGC4LIB_at_listserv.nd.edu
> Subject: Re: [NGC4LIB] As a Library 'decision maker'
>
> .
>
> Computer science research in information visualization would probably
> benefit from an attempt at applying their efforts to LCSH, for example,
> but LCSH is not openly available in the same way as, say, DMOZ <
> http://rdf.dmoz.org/ > Similarly, AACR < http://www.aacr2.org/ > is not
> available in the same way as the standards published by the W3C <
> http://www.w3.org/ >; one has to purchase the former whereas the latter
> is openly and freely available online. There is an analogy here between
> the closed world of Microsoft-type software platforms and that of the
> open source systems.
>
> Tomasz Neugebauer
>
> Systems Development Librarian
> tomasz.neugebauer_at_concordia.ca
> Concordia University Libraries
>
>
> -----Original Message-----
> From: Next generation catalogs for libraries
> [mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Jonathan Rochkind
> Sent: Thursday, September 20, 2007 5:09 PM
> To: NGC4LIB_at_listserv.nd.edu
> Subject: Re: [NGC4LIB] As a Library 'decision maker'
>
> I think quite a bit of the way that the catalog _records_ are structured
> (and not structured) is also a legacy of the card catalog error, and
> equally needs to be changed just as much as the catalog interface does.
>
> But to be clear, yes, there is a lot of power in catalog records that is
> not found in Google. That should not be lost in any changes. There is
> also a lot of power that is _intended_ to be in catalog records, but the
> nature of the structure of those catalog records makes it too hard to
> take advantage of that power. That needs to be fixed.
>
> But we've had this discussion so many times before, we just keep going
> round.
>
> Jonathan
>
> Weinheimer Jim wrote:
>
>> Janet Hill wrote:
>>
>>
>>> If you define "catalog" as the discovery tool maintained by a library
>>> (or
>>> group of libraries) through which information seekers may obtain
>>>
> access
>
>>> through various mechanisms to information resources -- whether owned
>>>
> (or
>
>>> leased) by the particular institution or not -- then its use will
>>>
> certainly
>
>>> be greater than if you define "catalog" as the tool that provides
>>> access in
>>> a single listing utilizing a single methodology, and only to
>>>
> materials owned
>
>>> (or leased/subscribed to) by the local institution.
>>>
>>>
>> Also, there needs to be an understanding, especially on
>>
> librarian/professional lists, that there is a difference between the
> catalog *records* and the catalog *interface.* The interface of the
> catalog is obsolete and based on the card catalog. There have been a few
> attempts--only recently--to change this, but basically, the ways the
> catalog works hasn't changed much in the last 100 or so years. The only
> real change has been in the introduction of the keyword search, but that
> was achieved with the introduction of the OPAC. The latest "novelty" of
> the OPAC now is that it works more like a card catalog(!), in that the
> user can see the lists of headings. Big wow!
>
>> But it must be confessed that the traditional library catalog allows
>>
> types of searching that Google, Yahoo and the like *cannot do* in any
> way: the traditional catalog allows for the searching of concepts
> instead of just text. Once I demonstrate this to my students, they begin
> to want conceptual access over other resources on the web. Everybody
> needs to understand this, and only then can we make a decision whether
> the traditional catalog needs to continue.
>
>> Certainly, the catalog interface is obsolete and needs to be trashed,
>>
> but the records that allow for conceptual searching need to be retained
> and used to much greater effect than they are today. The "local
> collection" is a fiction that no longer exists with so many high-quality
> materials available on the web. One of the problems and tragedies is
> that so few people, including information professionals, seem to
> understand this today.
>
>> James Weinheimer j.weinheimer_at_aur.edu
>> Director of Library and Information Services
>> The American University of Rome
>> Rome, Italy
>>
>>
>>
>
> --
> Jonathan Rochkind
> Digital Services Software Engineer
> The Sheridan Libraries
> Johns Hopkins University
> 410.516.8886
> rochkind (at) jhu.edu
>
>
--
Jonathan Rochkind
Digital Services Software Engineer
The Sheridan Libraries
Johns Hopkins University
410.516.8886
rochkind (at) jhu.edu
Received on Fri Sep 21 2007 - 13:46:43 EDT