search

ArtResearch / artresearch.net

ArtResearch ResearchSpace application hosted at https://artresearch.net
5 stars 1 forks source link

Original record links: Repository records #145

Open spykoulouris opened 3 years ago

spykoulouris commented 3 years ago

In this photograph record the original record link is the one of the Metropolitan Museum. Why did that happen?

https://artresearch.net/resource/?uri=https%3A%2F%2Fpharos.artresearch.net%2Fresource%2Ffrick%2Fphoto%2F.b13362847%2F3107100130764_001

These links should only be the ones created by the Pharos partners.

spykoulouris commented 3 years ago

Same here. There are two links, one of the Frick's catalog the other from an external repository:

https://artresearch.net/resource/?uri=https%3A%2F%2Fpharos.artresearch.net%2Fresource%2Fpharos%2Fartwork%2F61c15aa04d2d87342550b5d2967ee4fef44c6ee4

@MinadakisNikos @lklic should we keep the links to external providers?

MinadakisNikos commented 3 years ago

Hi spyros it happened because they are storing their link and MET links in the same fields. If we want to distinguish we can filter based on the prefix of the url but it will be visible in the next iteration.

spykoulouris commented 3 years ago

@lklic what do you think?

lklic commented 3 years ago

I think that we need to separate out these links during the data transformation and create a new field in the artwork template that points to the original museum record. I think that the source record links at the top of the page should only be reserved for the source catalog, as this can be easily managed by us.

Museum links will be very useful to users of the platform but we will also have issues with link rot. On a previous project I worked on, 90% of the links to museum records were no longer active after 4 years.

One way to mitigate link rot is to create our own instance of perma.cc. This project led by the Harvard innovation lab is hosted by them but limited to 10 links per month per individual. The project is open source however and we can create our own instance that is free.

This would also open up very interesting possibilities for crowdsourcing this information-- where users could add links to the museum record and have this archived page be always available to end users. In any case I think this would require extra work that is not really within the scope of what we are doing here.

Let's keep the issue open, discuss it with the GUI committee, and then make a plan on when/if to implement this. I personally think it offers some very exciting possibilities for research by integrating websites and data that are not semantically enriched or published as LOD.

lklic commented 2 years ago

Moving this issue to the back burner. The issue regarding the original record from the institution is documented in #264.

Here we need to decide if we want to archive web pages from museum collections or other websites, that we would like to be able to present to users.