Introduction to metadata
What is metadata?
Metadata is data that describes the attributes of other data. For example, a photograph might have the following metadata:
- Time and date the photo was taken
- The size of the image
- The resolution of the image
- The camera that took the photo
- The aperture and shutter speed settings
- The lighting and flash settings
- Who took the photo
- Who is in the photo
- What is the subject of the photo
- What colours are in the photo
- Where was the photo taken
Metadata can be automatically generated by some technology, or it can be defined by the creator or user of that data. In a photograph the resolution, size, date and time, camera settings, and camera that took the photo, are all automatically captured and added at the point the photograph is taken. Other metadata, for example, who took the picture and what the subject of the photograph is may have been added by a user, for example in the form of tags or keywords added to the photograph.
Using metadata is a way to categorise or classify your data, making it easier to find and manipulate. For example, if you have added metadata to your photographs to the subject of the photo, it is easy to find all the photos including that subject and exclude all photos that do not contain that subject.
For more information about metadata, see [article on Wikipedia].
Metadata in the laboratory
Descriptive and preservation metadata are particularly important for data-driven science. Understanding and working with large data volumes is difficult without appropriate metadata or, to be precise, descriptive metadata. It is particularly important to capture - with metadata - the process by which scientific data is obtained: results should always be open to testing and be capable of replication. The relationships between objects – the research elements – are an important feature of the research process, as are the histories of each object. The digital provenance of research data consists of these relationships and histories.
The scholarly communication process relies not only on publishing analyses and conclusions, but also on the availability of the research outputs. To ensure the value and reusability, researchers must curate those outputs by capturing appropriate metadata. Designing curation into experiments is an effective solution to the provision of high-quality metadata that leads to better, more reusable data and to better science. For more information, see [J. (2008). Curation of Laboratory Experimental Data as Part of the Overall Data Lifecycle. The International Journal of Digital Curation, 3(1), 44-62. doi: 10.2218/ijdc.v3i1.41]
Why should I use metadata?
Using metadata when recording your experiment enables others to find that information, view the information that is linked together, which helps them to reuse the information and to replicate your experiments. Using metadata effectively also supports your own work, by enabling you to find your data easily in the future, to organise your data, and to help link related data together. Using metadata when you create your content enables you to locate that information without needing to trawl through everything in order to find it. You can filter out the content you don't need right now in order to narrow down the results to search through.
Using common metadata across a team helps the team to link their data together, and ensures consistency between experiments and projects.
Metadata in LabTrove
When you create a record of your experiments using LabTrove, some metadata is automatically created for you, but you can also add your own metadata to help you classify your data for future use. For Entries you create, the following metadata is automatically captured:
- Date and time of creation
- Date and time of updates
- A unique ID for each version
There is one piece of metadata that you must define as a user. For each Entry you create, you must assign the Entry to a Section. You can create and define your own Sections. The Section is useful for indicating what type of Entry you are writing, for example:
- Risk assessment
- Instrument instructions
- Reaction scheme
The different Sections that you create are displayed as links beneath the Section heading in the right hand navigation menu of each Notebook. Next to each Section title is the number of Entries that have been assigned to that Section. If you click on the name of the Section, a list of all the Entries that are classified as being that kind of Entry are displayed.
The Sections that you create will depend on the kinds of experiments and writing that you intend to do. It is helpful if a team are undertaking similar research that they have consistent metadata, including Sections.
There is a specific kind of Entry called a template that can be used to standardise the content of Entries. A template is created from a Entry by assigning it to the Section called Templates. A Entry that has Templates as the Section name is treated in a different way to a normal Entry and can be used to create a new Entry. For more information about templates, see Recording_repeated_processes.
Metadata keys and values
In addition to adding a Section to your Entry, you can add more metadata in the form of key-value pairs. The key represents the attribute of the data, and the value is the value assigned to that attribute. For example you could create a key of Instrument and use different values depending on the instrument used in the experiment. Other examples of keys you might use are:
- Analysis type
- Experiment type
- Project name
- Experiment code
- Procedural step
You can add as many key-value pairs as you need to your Entry to describe its content. It is helpful if a team are undertaking similar research to have consistent metadata, including using the same keys.
The different keys that you create are displayed as main headings in the right hand navigation menu of each Notebook. The values are displayed as a list of links under each of these main headings. Next to each value link is the number of Entries that have been assigned that value for the key. If you click on the value, a list of all the Entries that are classified as sharing that value are displayed.
What to do next
- Defining the metadata for your experiment
- Adding metadata to your entries
- You can create template Entries to help you record consistent information for different experiments. For more information, see Recording repeated processes.