Ticket #600 (new task)

Opened 1 year ago

Last modified 9 months ago

handle multiple abstracts correctly on ingest

Reported by: stevec Assigned to: rich
Priority: medium Milestone:
Component: ambra Version: 0.8
Keywords: Cc:
Blocking: Blocked By:

Description (Last modified by rich)

may require some migration for existing articles.

necessary to display certain abstracts on ToC, download correct abstract with citiations, etc. for articles with multiple abstracts.

Dependency Graph

Change History

08/22/07 16:59:05 changed by stevec

  • priority changed from high to medium.
  • milestone changed from 0.8 to 0.9.

see [3556]

bumping to medium and changing milestone since the immediate need is taken care of.

09/02/07 22:41:40 changed by

  • milestone deleted.

Milestone 0.9 deleted

09/20/07 11:11:20 changed by amit

  • owner changed from stevec to rich.
  • type changed from unassigned to clarification.

Will articles have more than one abstract? In that case how we do determine which one to display? If not, the upstream process needs to be fixed to catch such errors.

09/21/07 14:23:45 changed by amit

  • milestone set to 0.81.

10/05/07 15:48:58 changed by amit

  • priority changed from medium to high.

Upgrading priority as we need an answer if this is going to affect code logic.

10/05/07 16:32:28 changed by rich

  • owner changed from rich to amit.

Clinical trials articles may have two abstracts: 1. <abstract> 2. <abstract abstract-type="editor">

I think that this has already been resolved between Steve and Susanne. citation.xsl is displaying both abstract fields in the HTML. The citation download is grabbing the main abstract which is the correct abstract for download.

10/05/07 16:39:09 changed by amit

Unfortunately not in case of the migration script. I believe that picked up the first abstract (instead of the one without the abstract-type). Similarly we will have to check what the ingest does when it stores the abstract in Mulgara.

10/05/07 16:44:34 changed by amit

  • owner changed from amit to rich.

So in general for ingestion/migration the 'real' abstract will not have abstract-type? Will that be true across all journals? This is so that we extract the correct abstract into mulgara.

10/05/07 17:49:12 changed by rich

True across all the PLoS journals. But it looks like PMC wants multiple abstracts designated with an abstract-type: http://www.pubmedcentral.nih.gov/pmcdoc/tagging-guidelines/article/style.html

Let me research a bit more....

10/05/07 17:52:00 changed by amit

All right. All it probably means is another data migration...:) Please let us know ASAP.

10/09/07 17:34:57 changed by rich

We need to ingest multiple abstract types per article. The editorial teams want to display a link to an article's Author Summary from the ToC. The author summary is tagged as <abstract abstract-type="summary">. There is also the <abstract abstract-type="toc"> which is the article blurb displayed on the ToC.

I still clarification regarding a single abstract with no type. We may need to change the tagging to always have a an abstract type even if there is only one abstract e.g. <abstract abstract-type="main">

10/16/07 09:54:00 changed by susanne

I have a question in to PMC about this issue. I will update the ticket when I have a response from them.

10/17/07 09:58:22 changed by amit

  • owner changed from rich to susanne.
  • milestone deleted.

Reassigning to Susanne and moving it out for 0.8.1.

10/25/07 10:13:31 changed by susanne

PMC said it is not necessary to assign an abstract-type to the main abstract, even when you have multiple abstracts within the article. However, if Topaz needs us to do it, we could update this when we do our XML cleanup for the data migration and we could get PLoS NTDs to start being marked up this way. Otherwise, it would be easier for us if we could just leave things as they are, which is that the main abstract is always simply <abstract> and the secondary abstract or abstracts are identified by an abstract-type.

10/25/07 11:15:05 changed by amit

  • owner changed from susanne to russ.

No, assigning a 'abstract-type' to the main abstract is not necessary. However, this means someone probably needs to take a look at the ingestion application (specifically the XSLT) and to make sure that the correct main article abstract is being extracted and also extraction of other abstracts. Corresponding to that the Article model will need to be changed and it might mean another migration script. Since the Publishing Application is being slowly transferred over, this is not a Topaz decision anymore. Assigning to Russ.

12/17/07 15:55:10 changed by rich

  • owner changed from russ to rich.
  • priority changed from high to medium.
  • type changed from clarification to task.
  • description changed.
  • summary changed from download citation chokes on articles that have more than 1 abstract to handle multiple abstracts correctly on ingest.

03/24/08 11:09:59 changed by rich

There will be implications of the abstract types for Bio/Med migration.

From Susanne:

I believe you are thinking of the one-sentence blurb (which is usually tagged <abstract abstract-type="toc"> but sometimes tagged <abstract abstract-type="teaser">) that occurs only in PLoS Biology and PLoS Medicine. These are used in the e-mail alerts and on the table of contents. We do not display the Author Summary (which is usually tagged as <abstract abstract-type="summary"> but sometimes tagged <abstract abstract-type="synopsis">, <abstract abstract-type="author">, or <abstract abstract-type="special">) on e-mail alerts. They are too long! We display them within the article only. PLoS Medicine also has the Editors' Summary (<abstract abstract-type="editor">) which is quite long, and needs to be displayed within the article only.