This page is based on the document IPET-MDRD-1 I 09, which is result of TT-ApMD work on best practice of writing quality title and abstract.

1. What is this Information used for

The Title and Abstract fields are important in aiding the data discovery and selection process.

The Title and Abstract contents are indexed and used by the search engine to provide search results to WIS catalogue users.

Having conducted a search, the user is initially presented with the titles (and part of the abstract) of matching records. Perusing titles is the first step in assessing whether data/products are relevant.

Search results can also be sorted by title (or change date), and if the content layout is consistent, then browsing through the titles is more efficient. Ideally, for the title field, the layout should be as consistent as possible, as this provides easier browsing of the search results.

It is therefore extremely important to create a Title that fully and unambiguously describes a product.

It is also important to have, where possible, homogenous content in fields, to help users comparing the search results to make their choice. For this reason, it is also important to define a structure for abstracts.

Perusing the abstract is often the next step, after browsing a list of titles. Abstracts ideally allow users to understand what the different returned products are, and to select the results that are interesting to them.

2. Title Creation Best Practices

Titles are critical in helping a scientist find data. While searching for appropriate datasets, a scientist is most likely to use the title as the first criteria in determining if a dataset meets their needs.

A complete Title includes who, what, where and when.

• Who: Who created the dataset. For example “EUMETSAT IASI L2 Cloud Parameters – Metop”, or “Meteo-France 4 days Forecast Products”.
• What: What is the dataset about? For example, is it a deterministic forecast product or surface synoptic observations (SYNOP)
• Where: Region covered by the product. Is it global, or covering a region, a country. It is coming from a specific station
• When: Time validity of the product. For example “3 days forecast products for Africa”. This is not always possible, because some products have no validity or are continuous.

When the Title is written, it should be crafted to help users to quickly identify whether the dataset is relevant. Therefore, ambiguity should be avoided, and one should consider all the possible interpretations of the word choices, as well as ambiguity caused by absence of more uniquely identifying information.
Below are some examples of consistent and descriptive titles:

1. Meteo-France 4 days Forecast Products for Africa – NWP Model
2. EUMETSAT Active Fire Monitoring (GRIB) - MSG - 0 degree
3. 'UK Met Office 15 minutes Lightning Detection Data for Africa from ground based ATDnet network''

3. Abstract creation – best practices

There are no specific mandatory constraints regarding the abstract content, but trying to normalise its structure will help users when browsing records in search results.

The guidelines are as follows:

1. The first sentence should summarise, in one line, what the product is.
2. The following paragraphs should provide more details regarding the product (content, validity, data collection, data/product process(es), and the organisation creating the product).
3. The last paragraph should be dedicated to the available product formats.

Below are three examples of consistent and descriptive abstracts:

ATDNet Lightning Data

[+]

EUMETSAT Active Fire Monitoring

[+]

Meteo-France 4 days Forecast Products for Africa

[+]