Metadata with Linked Data – also known as ‘structured data’ – is a beautifully interwoven record of descriptions, authorship, provenance, context and history that can be shared with everyone. Written in RDF and JSON-LD, it becomes, like music, universal. What is Metadata Anyway?

What can you do with ImageSnippets?

ImageSnippets is a complete metadata editing interface that enables someone who knows little to nothing about RDF, OWL, ontologies, or even URIs to create descriptions for images using Linked Data (also known as structured data) which is written in RDF.

This type of metadata can then be used as part of the overall management of the assets. Your metadata is published openly as 5-star linked data to a triple store, which can be queried through our SPARQL endpoint and accessed through

RDF metadata is persistent, transportable, and read by semantically-aware machines.

As ImageSnippets uses linked data on all of its images, an entire database of images can be easily sorted. When proper and consistent data is provided to all images, a user can utilize ImageSnippets semantic search technology to help find a single image out of many others. This is supported by our “Triple Tags.” These tags connect a subject, property, and term. The term often comes from online databases such as DBPedia, Yago, and Art and Architecture Thesaurus. If a term is not available, a user can generate their own database with individual entities within the ImageSnippets software.

-Users can assert their copyrights in a way that is not easily stripped from the image before they share and post, thereby reducing the likelihood their images will be classified as orphan works.

By utilizing ImageSnippets’ triple tags, as well as other structured data you can add to your images, searching for images becomes much easier. By using the linked data, ImageSnippets provides more refined results based on the structured data attached to that image with a subject,property and object value.


Your metadata is published as 5-star linked data openly and daily to our triple store that can be queried with our SPARQL endpoint and accessed through Not only can metadata be added for the purpose of transport, re-use, and findability by semantic crawlers, there are numerous benefits to using linked data as part of the framework of an asset management system in general. Some of the advantages include:

-Layering more metadata with their images as more knowledge about the images reveals itself in various contexts. The context can be defined by the use of the properties and values of the triples, not by adding more database tables.

-Datasets can be ontologies previously engineered and loaded into ImageSnippets or they can be datasets that evolve as part of the curation process. The creation and evolution of the dataset terms can be orchestrated exclusively by an administrator or with collaborative input from a team of users.

Images managed in ImageSnippets can be shared on almost any social media site or webpage where links are shareable. If you choose to share an image from the application, shortcut options are available for sharing to sites like Google+, Facebook, and Twitter. You can also paste a short URL link to share the image. Clicking a link shared from ImageSnippets displays the image and basic information you have added, including copyright terms. Triple tags and other metadata are preserved with a triple store, in the html file; additional metadata (including desired IPTC and EXIF data) remain embedded in the image itself along with a link back to the file containing the triple data.
Images published on Web media are often stripped of their descriptive metadata. ImageSnippets records XMP metadata in RDF and provides a separate record of ownership and copyright information that cannot be stripped from the image. ImageSnippets additionally places a URI into the XMP of all published images which links back to the file containing the RDFa data. This creates a ‘link cycle’ connecting published images to a permanent record of their ownership and copyright metadata, making it harder for the image to become an orphan work.

Images shared online are often falsely attributed to incorrect events or with incorrect history. User’s rights are violated because copyright intentions are not always clearly known or visible when images are shared. ImageSnippets records those rights in a way that is both clearly visible and also easily read by machines and search engines.

ImageSnippets can generate output as JSON-LD or other RDF syntax forms. The standard default snippet is an HTML file containing RDFa. This can be published directly to public media such as Facebook, and the body of such a snippet can be pasted into Web pages where the RDFa is transparent to browser rendering.

What do we mean by ‘metadata’?

Metadata is any information you want to attach to an image. This includes the content of traditional camera-generated EXIF data, but it also goes far beyond it, to include descriptions of what the image is an image of; the history, ownership and provenance of the image; artistic or emotional descriptions or classifications of the image, or whatever some particular image corpus might want to say about it. For example, a corpus of images taken in a classic car repair shop might be built around metadata connecting images to vehicles, craftsmen, part numbers and project identifiers. All of this, and more, can be represented using the ImageSnippets RDF-based format for metadata.

Are you a GLAM person (Galleries, Libraries, Archives and Museums)?

Please Go Here!

Getting the most out of ImageSnippets:

If you are not a large organization or don’t have a lot of images on your own server, you will get more benefits from ImageSnippets if you upload them to our server first because you can then manage the metadata stored inside your image. You can DELETE data you do not want (EXIF location and camera settings for example) and you can ADD metadata that will then be embedded in the image – we support all IPTC and XMP fields. We also add a self-referential link inside the image back to an HTML file containing data about the image.

It should be understood that we do not want to profit from owning your image, or your data. We are not selling those commodities either. We just want you to be able to share and manage as much or as little as you would like to share.

If your file size is larger than 1024 x 768 – we reduce it to smaller dimensions proportionally. Why?

We are primarily interested in getting better at data preservation as it is presented on the web and we want to help you share your image data more accurately. We are operating on the premise that if someone really wants to buy or license your image that you have shared from ImageSnippets, they will contact you directly and you can supply them with a higher resolution file directly as you desire.

If you want to simply link your images from Flickr or another URL, we will import all the metadata you have in your original file, your Flickr keywords, username and link to the original image URL. Since we won’t have permissions to edit files stored on other servers, we won’t write any embedded metadata back into your original file, but we will still store this data into our published RDF triple store.

For customers who have a LOT of images on their own server and they don’t want to upload them to the system, you can:

Keep your images on your own server and we will work with you to interface the ImageSnippets system with your images. The import can be done with a csv list of your image URI’s.

If you do not need to change the data in the actual file itself – and you just want to share through us – the process is simple and there are numerous benefits to this method for large organizations. You keep your original images exactly where they are on your server – and we let you create lots of rich linked data triples that we publish for you, that will point search engines directly back to your original image. If you want the IPTC/XMP data changed or saved in your original file – then we can work with a system administrator to be able to write the embedded data back to your file.

Semantic Search Technology

Because ImageSnippets metadata is stored in a common format and with descriptive names used consistently, semantic search technology can be used to rapidly and reliably find images conforming to simple or complicated descriptions. The key technology used here is the RDF graph, comprising a connected ‘tree’ of descriptive branches all growing out from the image is its ‘root’. Each branch is a triple S P O which says something (that it’s Property is an Object) about a single Subject. Each descriptive term is an Internationalized Resource Identifier (IRI, the Web standard for names of any kind) which is extracted automatically from online descriptive encyclopedias such as DBPedia, Yago and the Art and Architecture Thesaurus. In brief, and without the technical detail: the image description can say virtually anything about anything using names that are understood world-wide, can be arbitrarily large and complicated, and can be searched and processed very efficiently by software. This includes ownership of the image, copyright information, provenance and standard camera-generated metadata and existing image tags, as well as describing what the image depicts or shows: all of this is directly expressed in RDF.

Sign Up Now!