As a community we would like to know the best-practices for archiving OSF

[mjy]

OSF is seeking to provide an example of best-practices defined by the COL. One of those best-practices is "archving".

• What does COL archiving entail?

The two targets for archiving I feel we should pursue first, in order, are

• Zenodo
• Internet Archive

Additional exploration could target GItHub's policy on archiving.

Can we

• Identify the required data/software products to Archive
• Identify the metadata required. What is shared, what differs b/w Zenodo and Internet Archive
• Define/describe the process of creating the archive
• Define/describe a SOP for updating the archive(s)

Once the archive is live

• Make use of the discoverability the archive adds to link back to
  • OSF pages
  • THe corersponding TaxonWorks API
  • Resources like this issue tracker
• Ensure the DOIs, and other identifiers associated with the archive and integrated back into our products
  • The data in TW itself
  • Webpages like the OSF

[debpaul]

Just starting to delve.

Zenodo FAQ has some information on metadata requirements and standards, for example.

COL archiving.

1. See COL Archive Repository
2. note that trying to access and understand a COL archive, the downloads I tried, failed.

From GitHub see Referencing and citing content

To make your repositories easier to reference in academic literature, you can create persistent identifiers, also known as Digital Object Identifiers (DOIs). You can use the data archiving tool Zenodo to archive a repository on GitHub.com and issue a DOI for the archive.
Zenodo archives your repository and issues a new DOI each time you create a new GitHub release. Follow the steps at "Managing releases in a repository" to create a new one.

• considering the above and the rate at which our projects might do a "release" keep in mind that

You can create releases to bundle and deliver iterations of a project to users.

[MMCigliano]

If downloads from COL Archive Repository works, maybe this would be the natural repository.
We could export to ChecklistBank in a more frequent period (4 times a year)

[debpaul]

If downloads from COL Archive Repository works, maybe this would be the natural repository. We could export to ChecklistBank in a more frequent period (4 times a year)

@MMCigliano I don't know (since it didn't work). I can't see what it contains.

Also, considering points made by @mjy about connecting "the data" and "the web pages" we need to see what's inside the COL to judge if it might work. For example, I can't see "the metadata" included in the COL file. More research needed. I'll be posting a ticket to the COL github repo to see if Markus Doering can fix the download.

[debpaul]

Update @mjy @MMCigliano for some reason (unknown to me): download of COL archives works for Geoff, fails on my computer. So you might both try (so we can "see" what's in them). And I'll suss out more later, why not working for me.

[klausriede]

redundancy in data storage and archiving is always good. However, just a reminder that OSF exceeds Zenodo and COL by far, reaching back to the 1990s when Daniel Otte started to establish a digital list of types and references. I think that this supremacy in data continuity should be maintained. It might be complemented by strategies of offline longterm digital storage.