Dear Mar Shiprim readers,
This month we thank Rune Rattenborg & Émilie Pagé-Perron for creating an introductory primer for using the CDLI to browse, search, edit and cite artifacts uploaded to the platform by a team of volunteers going back to three generations of scholars. We hope this primer will be useful to students and scholars new to the platform as well as to seasoned users who may be interested in contributing to CDLI. Rune and Émilie also share some of the platform’s aspirations for the future.
Look out for more primers in the future – are there topics or tools within Assyriology and related fields you’d like us to cover? Get in touch at marshiprim@iaassyriology.com.
With warm wishes,
Pavla Rosenstein
Mar Shiprim Editor
The Cuneiform Digital Library Initiative: A Primer
Rune Rattenborg & Émilie Pagé-Perron
1 Introduction
Welcome to the Cuneiform Digital Library Initiative (CDLI) primer! This guide will give you an introduction to the history of the CDLI and a basic overview of the CDLI research platform and data collections. It will enable you to conduct simple and advanced searches of the hundreds of thousands of artifacts in the database, filter and download your search results, register a user account with CDLI in order to edit and contribute data through our crowdsourcing system, understand how you can get the most out of our copious bibliography, and how to cite and reference the CDLI in your work. Finally, we provide some directions on current and future projects of the CDLI.
Figure 1.1: The co-directors of the Cuneiform Digital Library Initiative (www.cdli.earth). From left to right: Émilie Pagé-Perron, Jacob L. Dahl, Bertrand Lafont, and Jürgen Renn
The data collections of the CDLI hold records on ca. 390,000 artifacts inscribed with cuneiform, including hundreds of thousands of photos, line drawings, and transliterations. Linked to the artifacts are records on ca. 700 archaeological proveniences, ca. 1,200 public and private collections, and a bibliography of ca. 16,000 secondary publications (all figures as of April 2025). The artifact index also includes a steadily growing collection of external resource links, currently counting more than 400,000 links between artifacts in the CDLI and their corresponding record in other databases, making the CDLI as much a main gateway to the wider digital ecosystem of cuneiform studies as an essential database in its own right.
Next to these data collections, the CDLI also houses a series of scholarly journals and papers managed by the editors and the wider CDLI community. All are published in a digital format (.html and, for CDLJ and CDLB, also in .pdf), are entirely free of charge, and fully open access. If you are interested in publishing with us, do get in touch with CDLI editors at cdli@ames.ox.ac.uk. You can take a closer look at the journals and guidelines for submission here:
- The Cuneiform Digital Library Journal, a peer-reviewed journal for full-length research papers
- The Cuneiform Digital Library Bulletin, a peer-reviewed journal for shorter, specialist analyses
- The Cuneiform Digital Library Notes, a notice board for the publication of short notes
- The Cuneiform Digital Library Preprints, a venue for publishing drafts of research papers or manuscripts
2 Community
The CDLI is managed by four co-directors (see Figure 1.1) overseeing strategic priorities and daily operations. As of 2025, the directorship of the CDLI consists of Dr Émilie Pagé-Perron (University of York), Professor Jacob L. Dahl (University of Oxford), Professor Bertrand Lafont (Centre National de la Recherche Scientifique), and Professor Jürgen Renn (Max-Planck-Instiut für Geoanthropologie). Together, they represent three generations of scholars who have devoted considerable parts of their professional career to the building of the CDLI.
The directors are supported by a larger group of associates and collaborators. Associates are domain experts in Assyriology, archaeology, and computational linguistics advising on aspects of the CDLI data model and platform. Some of them also act as editors of the data collections, reviewing and approving contributions from the user community. Collaborators contribute with expert knowledge on particular bodies of cuneiform inscriptions, for example collections housed in specific museums, or relating to other research databases. They also count seasoned experts in a range of other areas, including digital imaging and modeling, research engineering, data management, and programming. Together with our global community of contributors, these all help to augment and further develop the data collections held by the CDLI.
The CDLI is and always will be entirely non-profit. We rely on public funding, private donations, and scholars, programmers, and translators volunteering their time and expert skills to keep the platform running. If you are interested in contributing to the CDLI community, you are always welcome to contact the editorial team at cdli@ames.ox.ac.uk. Our user community contributes to the platform in a wide number of ways, including not only the preparation of imagery, transliterations, and translations of cuneiform artifacts, but also cleaning and checking of metadata, curating our extensive bibliography, and building extensive related data collections on museum collections and archaeological proveniences, among many other things. Regardless of your level of experience, your skills can certainly be put to use at the CDLI!
If you have the means or the desire to do so, you can also support the CDLI financially through Open Collective, a crowdfounding platform that allows individuals and institutions to donate money to projects through an online platform. It also provides full transparency with regards to contributions and expenses. Anyone can access information on how donations to the CDLI are being spent and on which projects, as you can see from the Cuneiform Digital Library Initiative (CDLI) Open Collective portal.
3 Getting started
To start you off, let’s go to the main home page (at cdli.earth, see Figure 3.1) and take a look at the menus. This is the default entry point of most visitors to the CDLI, so it will be good for you to know where you can go from here.

Figure 3.1: The top part of the main page of the Cuneiform Digital Library Initiative (www.cdli.earth), including navigation bars (① and ②), the simple search tool (③), and various shortcuts (④).
To ease access to the search engine, the simple search tool is placed centrally on the main page. This feature also includes a shortcut to the advanced search tool. Below are given shortcuts to the Bibliography, the Browse menu, and the Documentation pages. We will get back to all of these shortly.
In the top of the main home page, you have two sets of navigation bars. The lower one, to the right of the logo, provides you with some essential shortcuts. These include the Browse menu, giving an overview of the different indices, Contribute, which provides instructions to registered users on how to contribute to the data collections, About, for those who want to learn more about the background and history of the CDLI, and Search, which contains a drop-down list leading to the various search features. This navigation bar is available on any page of the CDLI.
In the extreme top right, you find the user navigation bar, which includes features mainly for registered users and administrators. This navigation bar is also available on any page of the CDLI. It includes shortcuts to Publications, namely the various journals published by the CDLI, and Resources, which provides a full overview of the various reference lists, tools, and user manuals related to the CDLI data collections. Finally – if not logged in – you have the Login button where registered users can log into the CDLI, and Register, where you can sign up to be a user with the CDLI if you do not already have an account – or you need to create a new one.
4 Identifiers and indices
A lot of specialist terms and references will appear in the following, so before we dive into the details of searching the artifacts, we will briefly go over the meaning of the most important concepts. For further information on any aspect of the CDLI database, we would always advise you to search the Documentation pages. If you still cannot find what you are looking for, remember that you can always contact CDLI staff at cdli@ames.ox.ac.uk.
First, let us take a look at artifact identifiers, composite numbers, and seal numbers. These are the primary keys for identifying and linking individual artifacts inscribed with cuneiform with unique textual compositions, and with unique seal impressions. Once you understand how these identifiers relate to each other, finding things will be a lot easier. Second, we will look at the many other indices, or data tables, that are linked to the main artifact index. Here you will get an idea of all the different types of data that are connected to individual artifact records in the CDLI, what they mean, and how they can be used to search for the information you need.
4.1 Identifiers
The core of the CDLI data collections is theartifactsindex, which takes the cuneiform artifact as its primary entity. A cuneiform artifact is the physical object upon which a cuneiform inscription is placed. The artifact is recognised by itsartifact identifier. In addition, a unique textual composition can be identified by itscomposite number. A seal, or the impression of a seal on an artifact, can also be identified by itsseal number(see Figure4.1for an example).

Figure 4.1: An inscribed brick of Sennacherib, king of Assyria, giving the artifact identifier P428341 (①) and the composite number Q003563 (②). The tag ‘witness’ (③) indicates that this artifact identifier designates an inscribed artifact related to a composite number (from (https://cdli.earth/P428341).
4.1.1 Artifact identifier
The artifact identifier, or the so-called P-number, is a unique primary key for every individual cuneiform artifact contained in the artifact index of the CDLI. An artifact is any one physical object, or multiple fragments of the same object, inscribed with cuneiform, as well as seals, which may or may not carry an inscription. Artifact identifiers are generated by the CDLI, and consist of a six-digit number. Once assigned to an artifact, artifact identifiers are never reused or reassigned. They provide a permanent and stable digital identifier for all artifacts included in the CDLI and, by extension, for the vast majority of cuneiform inscriptions known around the world. Artifact identifiers can be searched through the artifact index. The stable URI of a CDLI artifact will always include the artifact identifier in the end, so for example https://cdli.earth/P123456.
Note that, contrary to most museum catalogues, an artifact record in the CDLI can include any number of fragments, indicated for example by multiple associated museum or excavation numbers, as long as these have been joined to form a complete or partially complete original artifact. This means that the number of records returned from a search on a particular body of cuneiform artifacts in the CDLI and, say, the catalogue of the British Museum, may not always be the same, since museum catalogues will often count fragments as separate entities.
4.1.2 Composite number
The composite number, also referred to as the Q-number, is a unique primary key for text compositions. A text composition is the unique inscription itself, not the artifact upon which it is inscribed. To take an example, multiple artifacts may bear the same display inscription of Ur-Nammu, so the composite number is a unique identifier of that specific inscription. The composite number consists of a ‘Q’ prefix followed by a six-digit number. composite numbers for cuneiform inscriptions are maintained by the Q Catalogue of the Open Richly Annotated Cuneiform Corpus (ORACC). A composite number can be applied to a cuneiform artifact when the text inscribed on that artifact constitutes a witness, that is, a rendition, of the specific text composition. composite numbers can be searched through the artifact index. The stable URI of a composite number found in the CDLI will give the composite number at the end: https://cdli.earth/Q000001.
A composite number has its own dedicated artifact record, and therefore its own dedicated artifact identifier, in the artifacts index of the CDLI. As you may have guessed, composite numbers are managed by an ORACC project because many subprojects at ORACC are ordered according to unique compositions, not unique artifacts. A composite number can be linked to several different artifacts bearing the same textual composition. At CDLI, we call the linked artifact a ‘witness’ to the composite, and the composition record itself the ‘composite’. This will appear immediate below the header on the individual artifact record (see Figure 4.1), and can also be selected in the filters. A composite number also allows for the creation of scores, namely the comparative arrangement of renditions of the same composition from different artifacts. You can search scores in the Composite Texts section of the Resources page.
4.1.3 Seal number
The seal number is a unique primary key for a specific seal, including the iconographic element as well as any cuneiform inscription it may have. Note that the CDLI includes seals both with and without cuneiform inscriptions. The seal number consists of a ‘S’ prefix followed by a six-digit number. A seal number can be linked to a cuneiform artifact when said artifact is either the cylinder or stamp seal identified with the seal number, or when said artifact carries an impression from said seal. Seal numbers can be searched through the artifact index. The stable URI of a seal number found in the CDLI will give the seal number at the end: https://cdli.earth/S000001.
A seal number also has its own dedicated artifact record, and therefore its own dedicated artifact identifier, in the artifacts index of the CDLI. Like composite numbers, a seal number can be linked to several different artifacts, provided that each artifact carries part or all of an impression of the same physical seal. The seal number itself will then designate the physical seal used to make the impression. A seal in the CDLI artifact index can then be labelled ‘physical’ if the seal itself has actually been found, or ‘composite’ if it has only been inferred from an impression. This will appear immediate below the header on the individual artifact record, and can also be selected in the filters. Just like with composite numbers, seal numbers allow us to integrate information about seals across a wide range of artifacts. You can find out more about our seals data in the Seal Portal section of the Resources page.
4.2 Indices
Okay, so the main index of the CDLI contains artifact records, identified by artifact identifiers. Some of these artifact records are linked together by composite numbers and seal numbers, which indicate relationships relating either to the particular textual composition inscribed on an artifact or the seal impressed upon an artifact. These are the three essential dimensions of the artifacts index at CDLI. Now, let us try to expand a bit on all the other things that the artifacts index ties together.
Most of the basic information, or metadata, associated with an artifact found in the CDLI is drawn from separate indices, or data tables. The home page of the CDLI will lead you to the artifacts index, but it is important to know the existence of at least some of the related indices.
You can find an alphabetically ordered list of all publicly available indices of the CDLI on the Browse page. This overview can be reached from any other location on the CDLI by using ‘Browse’ in the navigation bar on the top right. You can reach any index of the CDLI directly through the URL of that index – see the list below for some of them.
These indices are controlled vocabularies, meaning that they contain a list of carefully curated unique values that are the same throughout the database. Both simple search and advanced search of the artifact index can use these indices as filters. They can only be edited or expanded by registered users, and most of them only by CDLI editors. In the next section of this primer, we will demonstrate how you filter by categories in artifact search to further refine your results. But first, let us look at the indices. The following indices are linked to the artifacts index:
- Materials – the material of the inscribed artifact
- Material aspects – the aspect of the artifact material
- Material colors – the colour of the artifact material
- Artifact types – the type of artifact inscribed
- Genres – the genre of the inscription
- Languages – the language(s) of the inscription
- Periods – the historical period of the artifact
- Dates – the date of the artifact as given in or derived from the inscription
- Archives – the archive that the artifact belongs to
- Proveniences – the provenience of the artifact
- Regions – the region where the provenience of the artifact is located
- Collections – the current collection that holds the artifact
- Bibliography – scholarly publications relating to the artifact
- Authors – authors relating to publications relating to the artifact
- External resources – external resource links for the artifact
Take some time, either now or later, to explore the various indices and get an idea of what they contain. And remember that these indices are not just oriented towards the artifacts, but also form the link between artifact metadata in the CDLI and resources from elsewhere in digital space (see Figure 5.1 for an example). A value included in the material index can, for example, be matched to corresponding values in the Art & Architecture Thesaurus of the Getty Research Institute, meaning that the index integrates with more general vocabularies for the study of cultural heritage. A record in the provenience index can include links to the corresponding record of an archaeological site at Pleiades, Geonames, or Wikidata. An artifact linking to a provenience in CDLI is then not only linked to Assyriological data collections, but also many external databases used by other research domains.
5 Search
So far we have concerned ourselves with the structure and elements of the database. How to use the CDLI in a practical scenario? Knowing your way around the various search features is essential to exploring and using the CDLI database. There are three principal ways to search theartifactsindex;simple search,advanced search, and by giving theURL ID queryto selectartifact identifiers. We will go through each of these in turn.
Note also that, if you are looking for artifacts belonging to any category found in the controlled indices, for example aprovenience, acollection, or abibliographic reference, you can also use these to get the list of all artifacts associated with this category value. If, for example, we go to the provenience record forBabylon, that record will also include a direct link to an artifact index search giving all artifacts found at Babylon (see Figure5.1). If we go to the collection record of theBritish Museum, the same type of link will give you all artifacts in the CDLI index held at the British Museum. And if you seek out a specific record from the bibliography, say, the first volume of the seriesTextes Cunéiformes, Musée Du Louvre,Lettres et Contrats de l’époque de La Première Dynastie Babylonienneby François Thureau-Dangin, then this record will also give you a full list of all artifacts appearing in this publication.

Figure 5.1: The provenience record for Babylon, modern Bābil, in southern Iraq (https://cdli.earth/proveniences/200). Note the link to related artifacts on the upper left. Also note the links to external resources, such as Pleiades, Real-Lexikon der Assyriologie, and (not in this image) Wikipedia, OpenStreetMap, and Geonames.
5.1 Simple search
Now for the simple search tool. The simple search tool is available on the main page of the CDLI (https://cdli.earth/), being one of the first things you see when entering the site (Figure 3.1 and Figure 5.2). From any other location on the CDLI, the search page can be reached using ‘Search’ in the navigation bar on the top right. Going to the search page (https://cdli.earth/search) will land you on a full search of the artifacts index, from which you can proceed to apply filter by categories.

Figure 5.2: The simple search tool of the CDLI as it appears on the main page (https://cdli.earth/) and the artifact search page (https://cdli.earth/search). The tool includes links to search settings (①) and the advanced search tool (②). Note also the link to the search guide documentation on the top left (③) and the drop-down list to specify the domain to be searched (④).
By default, a query in the simple search field will search all fields in the artifacts index. You can narrow the search by defining a specific domain from the drop-down list on the right-hand side of the simple search field. The following domains can be specified:
- Publications
- Collections
- Proveniences
- Periods
- Transliterations
- Translation
- Identification numbers
A full description of how to use the search field and what criteria to insert is provided in the Documentation and can also be reached by clicking the ‘Search Guide’ link just above the simple search. If you want to search specific fields or domains, note that search terms can be interpreted differently depending on the field settings. Here, we will just go over a couple of essential rules:
- Entering multiple search terms in simple search is done by clicking ‘Add search field’ just below the search field. This will open up an additional search field where you can, for example, apply criteria relating to another domain than the first, and specify whether the relation between the search fields should be ‘and’ or ‘or’ in the drop-down to the left of the additional search field.
- Use double quotation marks (““) to search for an expression (words that follow each other, without taking into account punctuation). A search for
"Nineveh (Kuyunjik)"
, for example, will return results withNineveh
directly followed by(Kuyunjik)
, but notNineveh (mod. Kuyunjik)
. - If search terms are enclosed in forward slashes (/), they will be interpreted as regular expression patterns (if you don’t know what that is, there is a nice online resource for learning about and trying out regular expressions right here:https://regex101.com/). For example,
/^Nineveh/
will return only results whereNineveh
appears at the beginning of a line. - If search terms include question marks (?) or asterisks (*) at the beginning or the end of a word behave like a wildcard, respectively for one and multiple characters. a search for
*ineveh
, for example, will return results withNineveh
. - When searching for identifiers or in free search, preceding zeros in artifact identifiers, composite numbers, and seal numbers are normalised to support further matching. For example, searching for
BM 09999
orBM 9999
would match bothBM 09999
andBM 9999
. - When searching for transliterations (after specifying the domain) the search will behave differently. For example, searching for
ku
will not findku<sub>3</sub>
but will findku#
. Further details on transliteration search are given below.
5.2 Advanced search
The simple search is a versatile tool for making quick searches with one or two parameters. If you want to make more elaborate searches, you should try the advanced search. The advanced search can be reached from any location on the CDLI by clicking ‘Search’ in the navigation bar on the top right and selecting ‘Advanced search’ from the drop-down. The advanced search can also be reached by clicking ‘Advanced search’ to the right and below of the simple search field.
Going to the advanced search page (https://cdli.earth/search/advanced) will land you on the form giving you the option to apply one or more terms to one or more specific fields in the artifact index. Here, you can generate much more complex searches covering the full range of information relating to artifact records in the CDLI. Again, we will highlight just some of the main features available with this tool, and add some more specific guidelines on searching transliterations:
- In advanced search, multiple search terms can be entered in all fields using the typed-in operators
%AND%
and%OR%
. If using%AND%
between two terms, the search will only return results where both are present. If using%OR%
, the search will return results where one or the other is present. So, in advanced search, typingUruk %OR% Larsa
in the provenience field will return all artifacts from Uruk or Larsa. Note that this is a way of applying multiple search criteria in the same domain that is slightly different from the drop-down options used in the simple search (see above). - In advanced search, you can enable case sensitive search, so distinguishing between capital and small letters
You can search transliterations using the simple search as well as the advanced search tools, but the latter offers a number of additional features that will help you generate more precise and reliable search results. Searching transliterations also need to account for the intricacies of the ATF (ASCII transliteration format) file format for transliterations common in Assyriological text databases. If you are not familiar with ATF, ORACC has a very instructive guide on working with ATF that will be an essential read if you want to work more with transliterations. You can read more about searching transliterations in the Search Guide on the Documentation pages. Below are just some of the basic rules for searching transliterations as these apply to the CDLI. See our documentation on transliteration search for more details.
- Searches of transliterations are bound to sign values, meaning that
ku
will returnku
(𒆪), but notku3
(𒆬) - Phonemes with special representation in ATF, like
sz
,s,
,t,
andh
can also be searched with their equivalentsš
,ṣ
,ṭ
, andẖ
, andš
andj
also with alternative representationssh
andc
- Flags are ignored when searching transliterations, meaning that a search for
ku
will also return damaged or uncertain readings ofku
(𒆪), likeku#?
- If enclosing transliteration search terms in double quotation marks, search terms will be interpreted as-is, and so a search for
"ku"
will returnku
as well asku3
- Wildcards can be used in transliteration searches as well, so
?
can be used to represent any single character, and*
any number of characters - In advanced search, you can specify single line search to return only values contained in the same line of the transliteration, or use sign permutation search to search by sign names to look for alternative readings
5.3 URL ID query
If you already know exactly which records you want to find, you can search for the pertinent identifiers using an expanded URL. This works for standard CDLI identifiers, namely artifact identifier, composite number, and seal number, and will return a complete list of all artifact records where the given identifier appears.
All you need to do is to build the URL for the search, using the same format as we use for the stable URL of an artifact identifier, composite number, or seal number. This consists of the base domain address, ‘https://cdli.earth/’, followed by the individual identifiers that you want to include separated by commas, with no spaces. Try out the URL ID query given below in your own browser:
The search results from this query gives you only the records of these specific identifiers. This can be useful if you want to provide a single list of results from the artifacts index for a specific selection of artifacts, compositions, or seals – for example if you want to send a search link to someone or put a link to a selection of artifacts in a printed text.
5.4 Filter by categories
Results fromsimpleandadvanced searchqueries, as well as results from aURL ID query, can be filtered according to values from controlledindicesusing the filters on the left-hand side of the search results. This is another way of narrowing your search as you go, but also extremely helpful when you want to explore search results or get an idea of how results are distributed across different categories – especially if you have a great many records to look through!

Figure 5.3: The upper part of the filter menu of the CDLI search results (detail from https://cdli.earth/search), located on the left-hand side of the search results. The period filter (①) has been expanded and the period ‘Neo-Assyrian (ca. 911-612 BC)’ checked (②). If pressing ‘Apply’ now (③), the search results will be filtered to include only the 36,314 records from this period.
The filter categories on the left side of the search results can be expanded for you to see which values are included in the search results and how many records belong to each (see Figure5.3). Simply check the values that you want to retain in your search results, then scroll to the top or bottom of the filter menu and click ‘Apply’. The search results are then qualified according to the filter values that you are interested in.
Note that the search results will show the filters menu by default. If you want to hide the filters to give more room to search results, simply click the arrow to the right of the ‘Filters’ header.
6 Download and reuse
Once you have executed your search and filtered the results according to your needs, you may want to export or download the results. As the CDLI is an open access resource, the results of all searches from the artifacts index can be downloaded in full and in a variety of formats. You are not required to be a registered user with the CDLI to download search results. When satisfied with your search, simply click ‘Export’ on the top left of the list of search results. This will present you with download options for three basic types of data, namely metadata / catalogue, text / annotations, and bibliography (see Figure 6.1).
Importantly, always remember that the download will only include the records visible on the first page of your search results. The default is 25 records, the maximum is 10,000. If you have more than 25 records in your search results and you want to download them all, scroll to the bottom of the search results page and set the results per page to a higher number.
Note that if you want to work with exports of data exceeding 10,000 records, you should acquaint yourself with the CDLI API. This tool requires more extensive technical skills, but is an amazing resource if you want to work with larger collections of data.

Figure 6.1: The standard download of search results pop-up window visible on the artifacts search results when clicking ‘Export’. Listed are download options and file formats for metadata (①), text (②), and bibliography (③).
6.1 Metadata / catalogue
Metadata downloads include catalogue data related to the artifact, so all the information found in the artifact record except the text transliteration and translation itself. This is typically stored in a tabular format, like .csv (comma-separated values) or .tsv (tab-separated values). These file formats can be imported to Excel, OpenRefine, LibreOffice Calc or similar spreadsheet editors for further editing. Metadata can also be downloaded in .json, .rdf, and .xml-formats, but we will not go further into types of files here.
6.2 Text / annotations
The text itself, so transliterations, translations, annotations, word and sign lists can be downloaded in a variety of plain and linked text formats depending on what you need. The most common will be the primary text data in ATF, but you can also get annotation data in CoNLL, .json, .rdf, and .xml-formats. Words and sign lists come in .txt and .json, and include frequency of words and signs generated from your search. You can read more about the many tools developed for working with CDLI text and annotation data in the ‘Tools’ section of our Resources page and further below (see Resources).
6.3 Bibliography
You can also download the bibliography linked to your search results as a separate file. This is extremely useful if you want to capture all of the bibliographic references associated with a certain selection of artifacts from the CDLI, either for an assignment or an article, or as a starting point for working with a certain topic or group of inscriptions. These downloads are available in .csv or .bibtex formats. The .bibtex format is especially useful as you can easily import this file into your own reference manager, like Zotero, Endnote, Paperpile or similar.
7 Edit and contribute
Uniquely, the CDLI platform includes comprehensive front-end crowdsourcing features for most of its data collections, a service that has been in place since 2022. This means that anyone is free tosign up for a user accountand, once their crowdsourcing privileges have been confirmed by the editorial team,contribute and edit datato the CDLI via the front-end user interface. All edits and contributions arelogged and credited to the user, meaning that the various indices also provide a log of changes to individual records over time.
The CDLI has more than 400 registered users and receives around 1,000 submissions each month. A single submission may involve edits to anything from one to several hundred or thousands of records. Registered users can contribute and edit data in many different ways. You can find an overview of the various contribute forms and manuals of the CDLI on theContributepage. This overview can be reached from any other location on the CDLI by using ‘Contribute’ in the navigation bar on the top right.
7.1 User registration
In order to edit or contribute data to the CDLI, you will need a registered user account. You can register a user account with the CDLI athttps://cdli.earth/register(see Figure7.1). Note that you will need a 2FA (or two-factor authentication) application on your mobile device prior to registering. The CDLI recommends Google Authenticator, but any 2FA application will do. There are numerous proprietary and open-source options to choose from atGoogle Play App StoreandApple App Store.

Figure 7.1: The register form for new users at (cdli.earth/register).
Importantly, you should note that, when creating an account, you accept that we store your email address and your name. You understand that we will link your account to a public author profile (see theauthorsindex). This profile will display and credit you for any submitted and accepted edits to text, text annotations, images, image annotations, and metadata. You can read more about this underChange logs and user credit.
When you have completed the registration, make sure to contact CDLI staff atcdli@ames.ox.ac.ukto activate your crowdsourcing privileges. Then log out and log back in. Now, you are all set!
7.2 Submitting changes
As noted earlier, registered users can, depending on their level of authorization, edit just about any part of the CDLI data collections subject toapproval by editors. Once your user account and crowdsourcing credentials have been activated, and you havelogged in, you should see your user name appearing in the right-hand side of the user navigation bar (where it used to say ‘Register’) (Figure7.2). Also, when navigating to any artifact or publication record, you will see an ‘Edit’ button appearing in the upper right corner of that record.

Figure 7.2: View of a single artifact record, here P414454, an administrative text from Puzriš-Dagan. Note that the name of the logged in user appears in the user navigation bar on the top right (①), and that there is an ‘Edit’ button above the primary metadata field to the right (②) (from https://cdli.earth/P414454).
Note that, as of April 2025, the standard editing forms and submission procedure apply to edits of the artifacts and publications indices only. If you want to add proveniences or collections, or something else, do get in touch with the editors at cdli@ames.ox.ac.uk.
Now, there are two basic types of user submissions; edits to an existing record, and adding new records. We will start by pointing you in the general direction for adding new records, but otherwise concern ourselves with a basic example of how to edit metadata in an existing record.
7.2.1Adding new records
To add a new metadata record to the artifacts index, you can go to the single artifact edit form (remember that you need to be logged in to your user account to access this form!). Directions to the various forms for adding metadata, text, or annotation are listed under the Contribute page.
Note that data drawn from controlled indices, for example material, artifact type, genre, language, provenience and collection, will be limited to the values given in the drop-down list for that field. If you do not find the value that you need, get in touch with the editors at cdli@ames.ox.ac.uk to update the relevant index – or to find out what existing value to use.
7.2.2Editing existing records
Editing existing records can be done from the artifacts index search results overview and from the individual artifact record. If you click ‘Edit’ in the top right corner of the artifact record, you will see a drop-down list, giving you ‘Edit Metadata’, ‘Edit Images’, and ‘Edit Annotations’. We will focus only on the first here, so click ‘Edit Metadata’.
This will bring up the edit artifact form, with a bar on top displaying four numbered steps, namely ‘1. Edit’, ‘2. Confirm’, ‘3. Details’, and ‘4. Done’ (Figure 7.3). As you will see, we are currently on the first step .
You can add or change information in any free-text field, and change any existing value in fields linked to controlled indices. Here, we will limit ourselves to a simple update of the text contained in the ‘Designation’ field, giving instead of the museum accession numberWP 09973
a designation based on the primary publication,CDLN 2023/001, 01
. Once we have made the desired changes to the record, scroll to the top or bottom of the form and click ‘Continue’.

Figure 7.3: The uppermost part of the artifact record edit form for P414454. Scrolling down will take you to all the fields of the form, and clicking ‘Continue’ (①) will take you to the next step, ‘2. Confirm’. Note the bar on the top showing the progressive steps for data submission (②). The image to the left shows the original value in the ‘Designation’ field, the image to the right the new value that we want to add in the same field (③).
In the next step, ‘2. Confirm’, which is now indicated by the progression bar on top (Figure 7.4), you are asked to verify the submitted changes. The four buttons below the header gives you the options ‘Upload’, which will proceed to uploading the submitted change, ‘Upload later’, which allows you to store this edit in the change set cache and create one or many more, and then submit them all with the same details, ‘Edit’, which takes you back to the editing form, and ‘Discard’, which deletes the suggested change altogether. As we are satisfied with the change, we can now click ‘Upload’.
Note that, if you choose to temporarily store an edit in the change set cache using the ‘Upload later’ feature, and create additional edits before uploading, all your submissions must be completed before the end of your work session, otherwise the edits will be deleted.

Figure 7.4: The changeset report for submitted changes to the artifact record for P414454. Note that the progression bar for data submission is now at ‘2. Confirm’ (①). Below are buttons for upload, upload later, further editing and discard of proposed edits (②). The yellow warning bar below notifies you that the change has not yet been submitted. In the overview of changes to artifacts (③), the text marked in red is the previous value in the given field, the text marked in green is the value to be submitted. Here we can verify that the changes to be submitted are correct. If the changes cannot be implemented, for example because of a missing value in a controlled index, the changeset report will inform you. Note also the counter next to the logged-in user name in the user navigation bar on the top right (④). This is the changeset menu for the user, indicating that one changeset, namely this one, is currently pending upload.

Figure 7.5: The details form of the submitted changes to the artifact record for P414454. Note that the progression bar for data submission is now at ‘3. Details’ (①). The update type has been automatically set to artifact data (②). An explanatory text has been added manually in the ‘Event Comments’ field (③), and an author record associated with the submitted changes has been selected under ‘Authors’ (④). Scrolling down you will also find a list of all artifacts impacted by the submitted changes.
The last step, ‘4. Done’, confirms submission of the changes and shows the detailed information given for the submission and an overview of the changes included in this submission, listing all artifact records impacted. Your submission is now in the update review queue, which is checked regularly by the editorial team (see Editorial accept below).
Here, you also have the option to edit submission details entered in the previous step ‘3. Details’ by clicking ‘Edit’, and to withdraw the submission from editorial review and delete it altogether, by clicking ‘Delete’. If you are happy with your submission, you can leave it as is and close the browser window. Once an editor has confirmed your changes, they will be implemented in the database and the accepted submission listed in the ‘Update Events’ log.

Figure 7.6: The update event to the submitted changes of the artifact record for P414454. Below the header are given the details information submitted under ‘3. Details’ (①), and below a list of changes included in the submission (②), ordered per artifact impacted. Note the ‘Edit’ button (①), where you can revise details for the submission, and the ‘Delete’ button (③), which will withdraw the submitted contribution from editorial review and delete it.
Note that, when you have gotten used to the single edit form and to editing existing records, you can also begin to explore additional features for editing and updating data, specifically the bulk upload form. This feature is implemented for multi-record additions and updates to the artifacts and bibliography indices, and accepts .csv and, for the bibliography only, .bibtex-format files. If you are interested in using this feature, please look through the relevant documentation pages for editing artifact data using a file and get in touch with the editors when you have your files ready for upload.
7.3 Editorial accept
All submissions of new or updated data to the CDLI data collections are subject to editorial review, meaning that your changes will not be implemented and made public until they have been checked by an editor. If your submission is not approved or you do not receive a response regarding potential changes to your submission, you can always contact CDLI editors at cdli@ames.ox.ac.uk. When doing so, please remember that all editors and other staff at CDLI are providing their services on a voluntary basis.
In some cases, editors may ask you to amend your submission before or after the submission has been approved. In order for the editorial team to contact you, please make sure that the email address associated with your user and author account is always up to date. If users cannot be reached for questions regarding submissions, the submission will most likely be rejected.
7.4 Change logs and user credit
When confirming your crowdsourcing credentials, the CDLI editorial team will also link your user account to the corresponding record in the CDLI author index. This index contains a list of all authors linked to records in the bibliography and to the user profiles of authors who are registered users of the CDLI. All edits made by a user are credited to the author record of that user – as well as to any other author record that the user sees fit to acknowledge when submitting data through the crowdsourcing system.
As you rack up edits to the CDLI data collections, your CDLI author record forms a single point of reference for documenting and keeping track of your contribution to open access research. As your author record can also link to records in the publications index, the author record provides an overview both of submissions to the CDLI data collection and those of your publications that are listed in the CDLI bibliography. You can go to the profile of Rune Rattenborg to see how an author record in the CDLI may appear, including both personal details, biography, publications, and submissions. Take your time to curate your author record, as it showcases the work you are doing to increase digital access to cuneiform sources worldwide!
8 Bibliography and external resources
While you may be spending the most time on CDLI looking at artifacts, we would also like to take this opportunity to show you the bibliography index and how you can use this for study and research. The CDLI maintains one of the largest digital bibliographies for Assyriology, with over 16,000 titles. This bibliography is directly linked to a collection of 390,000 artifact records, and we are currently in the process of expanding reference links between the bibliography and other indices, for example proveniences and collections, to strengthen and promote referencing of all the scholarly publications contributing to our data collection.
You can access the bibliography directly through the shortcut on the CDLI main page by clicking ‘Bibliography’. This will take you to the bibliography index. From any other location on the CDLI, the bibliography can be reached by clicking ‘Resources’ in the top menu bar and selecting ‘Bibliography’ from the drop-down menu. Like the artifacts index, the default bibliography index view will give you the search results of all bibliographic records currently in the index, with the option to enter search terms for a number of specific fields given on the left-hand side of the search results.
As with search results in the artifacts index, you can also download all the results from a search of the bibliography. When satisfied with your search, simply click ‘Export Publications’ and select the preferred format of the download (Figure 8.1).

Figure 8.1: The download of publication search results pop-up window visible on the publication search results when clicking ‘Export Publications’. Listed are download options and file formats for flat data (①), linked data (②), expanded data (③), and as bibliographic data for reference management software (④).
As noted in the introduction, individual records of the artifacts index of the CDLI also contain external resource links to corresponding records in other online databases. When viewing a single artifact record, external resource links of the artifact are given in the ‘External Resources’ pane. These include links to artifact records in a wide range of research and museum databases. The former includes, for example, the many subprojects housed by the Open Richly Annotated Cuneiform Corpus (ORACC), the Database of Neo-Sumerian Texts (BDTNS), the Electronic Babylonian Library, and Archibab, among many others. The latter includes the British Museum, the Louvre, and the Yale Babylonian Collection, to name but a very few. While an artifact record at the CDLI may contain lots of information on a cuneiform inscription, it also provides direct links to many other resources that can provide you with a wealth of further knowledge. Make use of it!
9 Citation and referencing
Remember that digital resources, just like books and journal articles, are the product of sustained scholarly effort and just as relevant to properly cite and reference as things printed on paper. The CDLI – and many other digital resources in the field – is a product of decades of incessant data collection and meticulous curation. Make a habit of referencing digital resources as diligently as you would analogue resources, even if it requires you to learn a bit about how to correctly cite born-digital material.
When citing the CDLI, you have several options available. If you want to reference the entire resource, simply use the reference given below:
CDLI Contributors (2025)Cuneiform Digital Library Initiative. Available at:https://cdli.earth/
For all records and documents found in the data collections of the CDLI, a citation field is always available in the bottom of the browser window. Here, you can copy the citation to your clipboard in five different citation formats, namely Chicago, APA, Harvard, BibTex and RIS. Also note the text bubble, hovering in the lower right-hand side of every single record view, which gives you a copy option of the same citation formats.

Figure 9.1: The citation field at the bottom of the single artifact record window of P414454, including buttons for specific citation formats (①) and the corresponding citation (②). The floating icon to the right (③) follows the user scroll, and provides a pop-up of the same data as the citation field anywhere on the page (from https://cdli.earth/P414454).
10 Resources
If you want to learn more about the many aides and resources available at the CDLI for working with digital cuneiform, go take a look at our Resources section. This includes quick links to the CDLI Wiki, a legacy knowledge database of Assyriology with a lot of useful information, as well as the Documentation pages and the Bibliography.
Below these, you will also find a very useful set of links to various reference tools, including the immensely helpful list of Abbreviations for Assyriology, which will tell you plainly just what this or that shorthand means, as well as references for sign readings and general sign lists, among other things.
Finally, there is a sub-section concerned with tools. Most of these are more advanced digital solutions for specific tasks related to natural language processing and computational linguistics. Documentation is available both in the CDLI Github repository for each tool and in the Documentation pages. Contributors to CDLI have published a few articles regarding the application of these tools. If you are interested in using CDLI tools but you are not sure where to start after looking at the documentation and articles, please feel free to reach out to us.
11 Future projects
The CDLI maintains the largest index of cuneiform inscriptions worldwide, acting as a canonical reference for the correct identification and indexing of artifacts in printed publications, online resources, and written and visual media appearances. Our goals for the future aim to strengthen and further support that role, building and expanding upon a unique digital monument to the cuneiform world and the shared heritage of the Middle East.
Hence, global coverage remains a key priority. We regularly receive input of data on cuneiform artifacts from individual Assyriologists and an interested public, including for example Richard Firth of the University of Bristol, who has submitted more than 30,000 edits to our data collections, as well as from museums and collections around the world. Research projects also help to augment and expand the coverage of the CDLI. In collaboration with Geomapping Landscapes of Writing (GLoW) of Uppsala University, we are currently updating metadata coverage to include finds of cuneiform inscriptions at hundreds of archaeological sites not previously found in any online catalogue (see Ryberg et al. 2024 for a brief overview).
A closely related primary aim is the continuous addition of high-resolution imagery of cuneiform artifacts to our data collections. We continue long-lived collaborations with large and small museums and foster new ones. In the last few years, we have focused on acquisition of images of cylinder seals and we continue to deploy sustained efforts in digitising all types of inscribed artifacts. There has also been a shift in recent years with projects undertaking their own digitisation campaigns and generously sharing their data with the CDLI afterwards. CDLI always welcomes the opportunity to help preserve archival high-resolution digital photographs and disseminate web ready imagery for free.
Finally, we are continuing to expand access to a variety of user communities around the world. The launch of the updated framework, which underpins the user interface that you have been introduced to in this primer, was developed from 2017 to 2021 with an official production release in 2022. The aim was to offer extended functionality to researchers, but also to drastically improve access to various audiences, including machines, for data processing. This was made possible also through the research project Machine Translation and Automated Analysis of Cuneiform Languages, which provided some of the required functionalities. The next major step in our platform development is the translation of the user interface into Arabic, along with all of our controlled vocabularies.
While these are our primary strategic aims, remember that the CDLI is built and maintained by its user community, including you. As an initiative promoting open and reproducible research to a global scholarly and general audience, the platform is open for you to join up and start building your own research project within an established and thoroughly tested digital research framework. If you want to juggle ideas, or talk about potential projects you would like to run, do feel free to contact us!
12 Get in touch!
In closing, thank you very much for reading through this primer! We hope it will be of use to you. And again, do not hesitate to to contact the editorial team at cdli@ames.ox.ac.uk for any questions, queries or suggestions. We are always happy to help!
This work has been supported by Riksbankens Jubileumsfond, grant M24-0028: Echoes of History: Analysis and Decipherment of Historical Writings (DESCRYPT).
Please note that this primer is also available in .html and .pdf on Zenodo and will be expanded and updated in the future. Please check the latter resource for the most recent version.
13 References
-
Archibab. https://archibab.fr/
-
Babylonian Collection | Yale Peabody Museum https://peabody.yale.edu/explore/collections/yale-babylonian-collection.
-
British Museum. https://www.britishmuseum.org/.
-
Cuneiform Digital Library Initiative (CDLI). https://cdli.earth
-
Database of Neo-Sumerian Texts (BDTNS). http://bdtns.cesga.es/.
-
Electronic Babylonian Library. https://www.ebl.lmu.de/
-
GeoNames. https://www.geonames.org/.
-
Musée Du Louvre Official Website. https://www.louvre.fr/.
-
Open Richly Annotated Cuneiform Corpus (ORACC). http://oracc.museum.upenn.edu/
-
OpenStreetMap. https://www.openstreetmap.org/.
-
Pleiades: A Gazetteer of Past Places. https://pleiades.stoa.org/
-
Wikipedia, the Free Encyclopedia. https://www.wikipedia.org/
-
Dib, Firas. 2019. ‘Regex101: Build, Test, and Debug Regex’. Regex101. Accessed 25 December 2024. https://regex101.com/.
-
Geomapping Landscapes of Writing (GLoW) – Uppsala universitet. https://www.uu.se/institution/arkeologi-antik-historia-och-kulturvard/forskning/forskningsprojekt/forskningsprojekt–assyriologi/geomapping-landscapes-of-writing-glow.
-
Smidt, Gustav Ryberg, Carolin Johansson, Nils Melin-Kronsell, Seraina Nett, and Rune Rattenborg. 2024. ‘CIGS-AE: A Quantitative Index of Cuneiform Inscriptions’. Cuneiform Digital Library Notes 2024 (2). https://cdli.mpiwg-berlin.mpg.de/articles/cdln/2024-2.
-
Tinney, Steve and Eleanor Robson. 2019. ‘Working with ATF to Edit Texts’. Open Richly Annotated Cuneiform Corpus. Accessed 25 April 2025. https://oracc.museum.upenn.edu/doc/help/editinginatf/index.html.