Content: Inventories of Content
This project, like many others, will be successful when a bunch of “stuff” that is now in an old place, will be accessible in a new place, and in a new way. In this case the “stuff” is medical content documents. The new way is accessible through improved cross-silo search capabilities. So, knowing facts about the “stuff” – how many, what format, where – is important to the success of the project.
Thus the need for an Inventory of On-line Content. I wasn’t surprised when there was no such inventory available, so I went about building one.
I started with the current Print Catalog, and built a spreadsheet framework containing a row for every area described in the published catalog. For the columns, I created fields for facts of interest. These included how many documents, what format they were in currently, how they were delivered to the web, where the underlying content was stored, the from and to date ranges, who was responsible, and how often the content was updated. I set up the spreadsheet, and logged onto the web, and went looking for content.
I wasn’t too concerned about pdf of html-format documents. I was concerned about areas where everything was in a data base, and delivered dynamically to the web. In these cases, the content would not be readily indexible by an indexing program. And the content would be difficult to deliver to a link from a search hit-list if it was accessed and formatted by application code. I wasn’t sure what I would find…
Inventories of Content - Part II
I was hot on the trail of a comprehensive inventory of all of the Client’s on-line content. I knew that the level of conversion and integration effort depended on how much content was in what format.
In our initial planning, it appeared that several important content areas were maintained in a SQL data base, and were formatted and delivered dynamically via ASP to the Users web session. How would a generic search tool index and access this content?
I had formulated a strategy, though, for the content maintained in a data base. That strategy was to publish it to a searchable HTML format on a batch basis. That would require some customization of the existing ASP scripts, but much less than building live links between the search engine for both search and presentation.
I spent many tedious hours bringing up web search interfaces in each existing content silo area, and looking for content. I could usually estimate how many underlying content items were available through searching. By looking at the page source, I could tell how the content was being delivered to the web. I followed up my web sessions with interviews with the content managers in each area to confirm my findings.
Fortunately, because some customers still wanted and received hard-copy publications of the content areas, most content in data bases also existed as .pdf files, which are easily cataloged and linked to.
I ended up with the first comprehensive inventory of on-line content – useful to the Client in and of itself. For me it is a checklist of how much stuff there is, where it is, and in what format it is. This is the set of stuff that must be searchable and accessible at the end of the project
Larry Cone
Post a Comment
View Printer Friendly Version
Email Article to Friend
Reader Comments