Hi Tomasz and Will -- Coversheets intuitively appear to make a lot of sense because, as Will says, they enable a level of branding and can ensure license details are clear. However, they are *explicitly bad* for indexing by Google Scholar [1], [2]. They disrupt automatic metadata extraction processes used by GS, a feature that GS has become very good at doing, e.g. to supplement associated metadata.

For a time, we were using automatic coversheets at my previous institution (Strathclyde) but I disabled them circa 2016 as it became clearer that they could be problematic (not just for GS, incidentally). We are using coversheets at my new institution (Glasgow); however, this is something we will seek to review soon.

Alternative options:

• Instead of pre-pending the coversheet, you could consider post-pending, thereby including relevant branding, copyright info, bibliographic citation, etc. without screwing up GS indexing. The original coversheet plugin (assuming it works with 3.4+) enabled such a configuration.
• Seek to develop a ‘watermark’, arXiv-esque plugin! 😉 I attempted this while at Strathclyde but, a) I’m a coding charlatan and so it wasn’t very good, and b) even when it was successful, it was spectacularly difficult to anticipate the formatting of accepted author manuscripts such that any watermark could be sized and located appropriately on the page. Watermarks would end up obscuring text, etc. It was a disaster.

The team at Strathclyde began added footers to manuscripts containing any relevant info. They were already performing a lot of file handling (e.g. improving formatting, accessibility, etc.) that inserting details in a footer wasn’t especially onerous.

One final comment. I undertook some repository discoverability research a few years ago [3]. Coversheets were identified as a potential issue, but I concluded that it was impossible to measure the negative influence they had on repository indexing because repository visibility tends to be about the ‘accumulation of marginal gains’. But given that GS explicitly advise against them, I suppose we must conclude that they are indeed a bad idea.

Hope this helps!

George

[1] Acharya, Anurag. (2015) Indexing repositories: pitfalls and best practices, https://purl.dlib.indiana.edu/iudl/media/6537033b6s

[2] Westin, Monica. (2021) Google Scholar Indexing for Repositories: Best Practices and Fixes for Common Indexing Problems, https://youtu.be/C-miRaROsaE?si=3sGDP-bD9MwDE7BX

[3] Macgregor, George. (2020). "Enhancing Content Discovery of Open Repositories: An Analytics-Based Evaluation of Repository Optimizations" Publications 8, no. 1: 8. https://doi.org/10.3390/publications8010008

--

Dr George Macgregor | Assistant Director – Digital Library

Information Services | University of Glasgow

Web: https://purl.org/g3om4c  | Fediverse: @g3om4c@code4lib.social
orcid.org/0000-0002-8482-3973

Mobile: +44 (0)7977 858281
--

The University of Glasgow is a registered Scottish charity: Registration Number SC004401