do not store/render images for bots

[bwsevelyn]

When a generally benign bot makes requests of the Aspen server, medium and small images are saved locally. Generally this is great and makes loading faster for patrons, but after time this eats up space on Aspen servers and not necessarily for covers that are searched often by actual patrons.

If crawling by bots only returns data and not images, this could help save space on servers while at the same time ensuring that real patron searches are treated as originally intended.

[mdnoble73]

Some bots are specifically looking for images, is the intention to not serve those bots or to treat those ones differently?

[kylemhall]

What do these bots do? Do they benefit the libraries?

[mdnoble73]

Bing has an image search as well as Google I believe. Not sure about benefit to libraries although libraries are always pro getting their catalog into Google. Perhaps a switch to enable or disable would be appropriate. I'd say a switch would be the MVP with a configurable normal/don't cache/don't serve option on a per bot basis with a default would be the best version.

[kylemhall]

Considering those images are from other sources I cannot imagine they create much value as opposed to the record data.

[mdnoble73]

Libraries can already block content by user agent, so perhaps an option to extend that to block covers separately from content or to change an option for what is blocked for a given user agent?

[kylemhall]

Yes

[lizrea]

100% by user agent would be amazing

[bwsevelyn]

thank you, yes, by user agent would be great!