ALEG

I thought I'd produce a "diary" describing what I did each week, what I didn't do, and what I hope to achieve next week and subsequently.

Firstly, thanks to you both [MLA, ALM] for a great introductory week; easing all the administrative details, providing me with lots of relevant background material, thought-provoking conversation and feedback and general friendliness!

What I've done

• I think I've progressed quite some way into acquiring the necessary "cultural" background to the project. I now have a broad appreciation of what data you need to maintain to migrate the current AUSTLIT and BAL databases and I've a reasonable idea of how the Lu Rees, Responses to Asia & Multicultural writers parts could fit together.

• I have a broad appreciation of how the AUSTLIT indexing practices work, although I think talking through the details with Tessa and Kathy next week will make me more confident.

• I can understand the work done on representing some of the AUSTLIT material in USMARC format.

• I have a vague appreciation of Kinetica, some of the issues related to identifying holdings (location of user, patchiness of holdings on Kinetica, possible costs accessing Kinetica, reformating Kinetica's response, issue of multiple editions), and what RAAM is.

• I've thought about how to make the data entry/maintenance of the new system easy to use/productive, but only in general/vague terms (reduce typing through use of pick lists, popup menus, templates, defaults...).

• I've very tentatively thought about a data model which would support the data storage and query requirements as I currently understand them, but I realise this is a very early attempt, and more for my benefit to test my understanding than anything else at this stage (sketched on my whiteboard).

• Had a quick chat with Ross Coleman regarding SETIS and how people who have lobbed somehow into SETIS could pick up information from ALEG (maybe local holdings of print copies, associated works, reviews, etc) and how ALEG could link to the full text version on SETIS (hardcoded URL stored in ALEG? by searching Kinetica and finding the SETIS holding? by SETIS?), and whether a full text search entered in ALEG (eg, find works mentioning "convicts and Parramatta" within the same paragraph) could be issued more-or-less automatically on SETIS...

• I don't have a first hand knowledge of how the indexers currently work (next week!).

• I don't know anything about how the current AUSTLIT system is designed/implemented (maybe not important yet).

• I haven't finished reading my way through those manilla folders, which I think is actually high priority - the more I can pick up in way of background, the better. I'm especially keen to assimilate Debbie Campbell's subject gateway material.

• Talk to Tess/Kathy about how it works now, what they think of the process, their suggestions for change.

• Talk to expert marc person ("Christine"?)

• Find out more about Kinetica and NLA plans for making holdings data available at no cost (?)

• Think about issues for distributed maintenance of the system - are web based tools the only or best option?

• Think about the nascent data model, understand more about the user queries that need to be satisfied now and in the future. Try to make things as general as possible to accomodate future requirements, but still make the thing implementable within 6 months.

• Think about relationship of MARC to the data model - is MARC just an exchange format (but not a data storage, data manipulation format)?

• Think about Z39.50 - both providing a search interface as a Z39.50 target, and finding out whether we actually ever need to use Z39.50 as a component of the gateway to retrieve externally held resources (Kinetica holdings are an obvious external resource (but Z39.50 is just one way to get to these), are there others?)

• Think about Voyager - do we use it for:
  • data entry/storage/searching
  • data entry/storage
  • storage/searching
  • storage
  • not at all?

• Start thinking about platforms (esp if not using Voyager) - Solaris -v- NT, I guess Oracle...

• Produce a draft of the Stage 1 report (Technical overview and functional requirements), as by the end of next week I'll be close to the end of the Stage 1 budget (8 days)

• Lots to understand, but on track to complete Stage 1 within 8 working days.