[DICT] May 15 conference call

[Jeff Alexander]

Hello,

This is a reminder that the dictionaries working group will be holding its regular conference call today at 2100 UTC. [1]

The agenda is available [2] and lists the order of use case priorities for discussion today.  The goal is to finish our discussion of the highest priority items as soon as possible so we can move on to defining atomic properties. 

Jeff

[1] http://www.timeanddate.com/worldclock/fixedtime.html?iso=20120515T21

[2] https://docs.google.com/document/d/1orr-Rbv7c8G8GOFkmtWprHqUC66yDe_8y9b3P12_AM4/edit

[Jeff Alexander]

Meeting minutes for today's call are now available:

https://docs.google.com/document/d/1orr-Rbv7c8G8GOFkmtWprHqUC66yDe_8y9b3P12_AM4/edit 

Jeff

[Yasuo Kida]

My attendance at the F2F meeting has been confirmed. Will be staying at the hotel where the meeting is held. What time does it start?

Thanks,

- kida

[Yasuo Kida]

Sorry for my poor memory. What was action items before the next con call on 5/29?

- kida

On 2012/05/15, at 16:39, Jeff Alexander <je...@intangiblepress.com> wrote:

[Jeff Alexander]

I'd like it if people could review the "Needed Publication Properties" section of the charter document before the next call:

http://idpf.org/charters/2012/dictionaries/#Needed_Publication_Properties 

At the F2F, our goal will be to settle on a list of 'atomic elements' that need to be addressed by the spec. This section of the charter document is not a draft list as such, but it does provide useful background and a starting point for that discussion. 

Apart from that, we need to finish up discussion of just a few use cases. (Stay tuned!)

Jeff

[Robert Bolick]

General comment on Needed Publication Properties

For me, the purpose of this section seems to overlap with the implied purpose of the section called Structure and Semantics.      Could we distinguish or clarify the purposes of these two sections?

Particular comments on Needed Publication Properties

1)  Package metadata:  There are many other categories of dictionary besides monolingual and multilingual.  For example, medical, biographical, geographical, economic, etc.   Although one could argue that more often than not the metadata that clarifies the category is embedded in the title (The Oxford Dictionary of National Biography, Chambers Biographical Dictionary, Webster's Dictionary of American Authors), there are instances where the title does not.   Therefore, should there not be metadata in the pkg that declares the "type of dictionary"?    If so, is this something that should be picked up from ONIX?   And if so, should we consider going down the ONIX and ticking those items of metadata that we think are core for dictionaries?

BTW, the sample record in the ONIX_Book_Product_2.1_rev.03.pdf happens to be a record for a dictionary!  

2)  Package metadata:  metadata for intended audience.   These metadata elements are also covered in ONIX (PR 14), so my questions above apply here as well.   

3) Entry structure:  In the first bulleted item, should "is" be "can be"?   "A dictionary or glossary entry is both content "

4) I'd like to draw attention to the Use Cases document (rows 21-29).   Row 21 includes a first stab at the atomic elements of a dictionary entry.   Is it good enough?  https://docs.google.com/spreadsheet/ccc?key=0AgaNbXmzpYW1dHdmRFZCelBWb3VWMVVzMDlpX1dLckE&pli=1#gid=0

5) Entry Structure:  Second bulleted point.   Apologies if this has been addressed, but I don't understand what is meant by "delivered mechanism," which implies, of course, that I don't understand the sentence fully.  Also possibly contributing to my confusion here and in the next bulleted point is their formulation in terms of what the author needs to be able to do.   Shouldn't the formulation be in terms of what the user must be, should be able to do, see, etc., or what the entry's structure must or should consist of?  Otherwise, will we wind up stipulating CMS rules for dictionaries?

6) Entry structure:  Third bulleted point.  Isn't "entry's domain" a package metadata element?  It seems to be covered by ONIX PR13.

7) Entry structure:  Fourth bulleted point.  Is this describing the same things as the subatomic elements referred to in Row 26?   I think the term "metadata" is slightly misused in Row 26.  True that "n", "adj", and "vi" are elements that describe the headword entry's grammatical function, but they aren't package metadata.   Does this bulleted point belong under Headwords and Inflections?

8) Entry structure:  Fifth bulleted point.  I am struggling with this one.   The example given does not describe "an entry within another entry."  Does it?   I'm thinking that an entry within an entry would be a headword that has multiple definitions (e.g., legacy, which would include the nominal definition of inherited assets and the adjectival definition describing software or systems to be replaced).   Is this bulleted point related to the fifth bulleted point under Headwords and Inflections?

9) Headwords and Inflections: Again, Row 26 in Use Cases touches on this.  The example "in*val*id" in the bulleted point gives only a partial guide to pronunciation.   Do we need to expand this bullet point or add another to cover the inclusion of a phonetic representation as a subatomic element?   If yes, do we need to make it explicit that the package metadata should declare the phonetic system being adopted by the dictionary?

--
Robert Bolick

http://uk.linkedin.com/pub/robert-bolick/4/8bb/ba2

[Jeff Alexander]

Bob, thanks for reading this over and commenting. 

Before I address your questions / comments specifically, I should just say that I felt it was useful for people to re-familiarize themselves with this section as background for the F2F and the next phase of our work. As for the exact wording of this section though, I don't see us revising or reusing its language per se. 

>For me, the purpose of this section seems to overlap with the implied purpose of the section 

>called Structure and Semantics.      Could we distinguish or clarify the purposes of these 

>two sections?

Needed Publication Properties was an initial attempt to define the properties that would be required in an EPUB dictionary publication in order to support the use cases in the original document. So it is outdated inasmuch as it doesn't reflect our recent discussions of priorities and concepts. 

Structure and Semantics is simply a glossary of lexicographic terminology.  We may incorporate some of these concepts into our semantics, but it's not a spec or requirement list.

>1)  Package metadata:  There are many other categories of dictionary besides monolingual 

>and multilingual.  For example, medical, biographical, geographical, economic, etc.   

That's true. But really, these are more like two different axes. While specialized subject dictionaries are generally monolingual, they can be bilingual as well (eg, a German-English scientific dictionary). 

I'm not up-to-speed enough with ONIX to speak to its usefulness for this or for 2) yet.   

On 3): yes, that's probably a better wording for it. 

>4) I'd like to draw attention to the Use Cases document (rows 21-29).   Row 21 includes a 

>first stab at the atomic elements of a dictionary entry.   Is it good enough?  

Hmmm. I don't know that we should try to define what a dictionary entry must contain from a user point of view. The publisher determines the content, which may or may not contain a classification. (For instance, subject dictionary entries often do not have them--think of a dictionary of idioms, mythology, music, etc.)

Thanks for taking a stab at it. I'm looking at this more functionally--what is atomic element X in the context of an EPUB ecosystem?  So my first stab at "dictionary entry" would be more like:

Dictionary entry: an organizational unit of a dictionary that contains one or more terms, and that can be rendered as syndication content by a reading system during word lookup

I'll try to reply to the remaining points soon.