[dpub-arch] 24 March follow-up: CLOCKSS, signposting

Hello DPub IG,

Thanks again for allowing me to join the 24 March DPub Archival Task Force Call. I've since officially joined the DPub IG and will participate in the Archival Task Force as time allows.

I'm following up our last conversation with a couple of resources that may be of interest:

*         How CLOCKSS works (https://www.clockss.org/clockss/How_CLOCKSS_Works) - the scholarly web content to be preserved in the CLOCKSS network is either harvested (front-end, presentation files, crawler, HTTP) or transferred (back-end, source files, FTP or RSYNC). Transfer is easier for us, since the content is packaged more discretely; ongoing and platform-specific adjustments on our end are instead needed to identify and map the collection of harvested web resources to intelligible and discrete archival units. In the best case, PWP would seem to provide a mechanism for web harvest to also serve discrete, publisher-designated packages for archiving.

*         Signposting the Scholarly Web (http://blog.dshr.org/2015/12/signposting-scholarly-web.html) - a comparatively lightweight enhancement to current scholarly publishing on the Web that by itself would solve a major technical challenge we have: identifying which links off of a publication landing page point to web resources that need to be preserved as part of the archival unit. I don't know how critical the specific recommended implementation is versus the conceptual idea (which is similar to the W3C Web App Manifest), but I'd be happy to reach out to Herbert, Michael, and/or David if their insights would be additionally helpful.

Please let me know if you have any questions.

~Nicholas
--
Nicholas Taylor | Web Archiving<http://library.stanford.edu/projects/web-archiving> Service Manager | Stanford University Libraries DLSS<http://library.stanford.edu/department/digital-library-systems-and-services-dlss>