Skip to page content
Institutional Repository Infrastructure for Scotland
Repository toolkit

Toolkit for institutions and repository managers

Repository deposit workflow

Setting up the technological infrastructure is only the first step in building your institutional repository. Once you have decided on the repository remit and adopted appropriate policies you will need to think about developing the repository service. This section aims to provide you with a basic practical approach to take when you first start to provide a running repository service.

When an item has submitted to a repository it normally sits in part of the administrative user interface commonly known as the buffer (or task pool) to await attention from repository staff. The repository manager normally has to approve the item before it is archived online. We present here a checklist of recommended actions that the repository manager (or other) should take.

Ingest

Most repositories only allow material from university staff and postgraduate students from the institution. If you have this policy you will need a mechanism to check whether the depositor is a genuine member of the institution. If your repository doesn’t have an authentication mechanism then the recommended way of doing this is to check the corresponding university web pages and staff lists. However, due to the limits on getting a staff email account it is fairly safe to assume that if the call was made from the institutional domain then the caller is genuine.

Next it is necessary to perform a check on the actual item in question and the collection is has been deposited to. Once an item has been archived in a specific collection it can be difficult to move to another. A little time spent checking this now can make your life easier further down the line.

Care should be taken to open up the file to ensure it is uncorrupted. Concurrently a brief assessment should be undertaken primarily to assess the validity and status of the item, i.e. is the item as described? What format has been submitted?

Copyright

Depending on the type of item you will need to check the copyright status. For new and previously unpublished material deposited by the rights holders, generally you will not need to clear copyright. However, it may contain material, e.g. images, whose rights are owned by a 3rd party.

The situation gets more complicated when you want to archive a previously published piece of work. Usually, authors pass over their copyright to the publishers, or assign a non-exclusive licence which prohibits further distribution by other parties. In this case you need to either clear the rights, or follow the rules of a blanket agreement assigned by publishers.

For journal articles you can use the SHERPA/RoMEO database of Publisher copyright policies and self-archiving to find a summary of permissions that are normally given as part of each publisher's copyright transfer agreement.

Searching through the online SHERPA database quite often will quickly give you the copyright information that is required. Sometimes you might come across a publisher that is not listed in the SHERPA database, or you might be dealing with other types of content, e.g. book chapters or conference proceedings. If this occurs then direct contact with the publishers will be needed. The procedure is as follows:

  1. Find contact details on publisher's website- usually under copyright permissions.
  2. Open standard copyright permissions letter (see Useful resources) and change details to suit your situation.
  3. Send permission letter via email to contact address.
  4. Save any responses from the publishers as text files in an appropriate place.
  5. If the response is positive continue with the submission, otherwise terminate the workflow.

Commonly, journal articles will be subject to the following archiving conditions set out by the publishers:

  1. archiving not formally supported
  2. can archive pre-print (i.e. pre-refereeing)
  3. can archive post-print (i.e. final draft post-refereeing), which is further sub-divided:
    • Author's own version of accepted paper (word processor version)
    • Publisher's version (e.g. the typeset PDF)
  4. can archive pre-print and post-print

Here is where it gets confusing - each publisher tends to have determined their own conditions and/or restrictions which dictate archiving rights or activities. In addition, they tend to use their own terminologies to describe them. A typical condition is to acknowledge the publisher's copyright in the work. Restrictions are more prohibitive, typically requiring some additional action on behalf of the author.

For our purposes we define a repository as:

  • a non-profit/non-commercial, institutional, open access e-print server,
  • NOT an author's personal web site or departmental web page, or password-protected site.

Coversheet

Depending on house style you may wish to add a coversheet to items within the repository. An example coversheet is available in Useful resources.

File conversion

Depending upon your policy of file formats (see Policies) you may need to do some conversion at this point if the submitted files are not appropriate.

Metadata check

Before the item gets archived online the metadata needs to be checked for mistakes by repository staff. Model guidelines for creating metadata in institutional repositories are available in the Draft IRIScotland metadata agreement (PDF).

Critically where the conditions state that we must link to the publisher version, or that the published source must be acknowledged, then you must ensure that the appropriate metadata has been recorded in the publication details section. Links to the publisher's version should also be added in the appropriate metadata field. Where possible, the Digital Object Identifier should be used as the protocol uses a resolver service limiting the effect of broken links. Where the DOI is not known the free DOI lookup service from CrossRef can be used.