SPEC Kit 329: Managing Born-Digital Special Collections and Archival Materials · 43
contents where possible. We have ﬁnished developing and testing this process and are ready to image our ﬁrst batch of
3.5” floppies, followed shortly by 5 1/4” floppies and hard drive after we acquire the hardware (drives, drivers and write
Legacy software: needed legacy equipment to access and transfer ﬁles. Donated mixed material collections: donor may
not own rights to all of collection that was contributed. Images in dissertations that might have fair use rights but not
necessarily general dissemination rights: how to deal with this.
Limited staff comfortable with ingest. Although we have an ingest process that has now been formalized and
undertaken with more than 50 accessions, we still only have a couple of staff members who possess the sufﬁcient
technical skills and understanding of digital records issues to undertake even the rudimentary steps in the accessioning
process. This leads to resource constraint issues as more and more digital records on media are being taken in, even
if they are not actively collected. To grow this program more, we need more, and lower-level staff to undertake much
of the accessioning process, as they currently do with paper. Minimal description practices don’t match ingest process.
We are following and forensic model of accessions where we are creating forensic images of storage media during
accessioning and setting those images aside for further processing. However, the current model for archival accessioning
on paper is to undertake minimal arrangement and description during the accessioning process, thereby eliminating
a backlog requiring future processing. Hardware and software ingest lab development was time consuming and
difﬁcult. Although we have now built up a signiﬁcant shared lab to enable the ingest of born-digital records from many
different types of storage media, the process of building such a lab took several years, expertise, and funding. Each new
collection seems to bring new technical issues that must be dealt with.
Major issue is technological — especially how to receive content from private donors. Still being worked on.
Media obsolescence/failure. This includes outmoded storage systems like 5.25” floppy disks and zip disks. Even if we
have hardware to accommodate them, we sometimes ﬁnd that the content is corrupted or otherwise inaccessible.
We have a small collection of old drives and other resources nearby; after that we consider outsourcing but will often
store as is or even deaccession, depending on resources and anticipated value of the content. Software obsolescence:
sometimes it isn’t even obsolete, it’s just got a small market share, like AskSam. So far, we have been able to ﬁnd
programs to access and migrate/normalize this content. File formats: we have received proprietary camcorder ﬁles, for
example, which we had difﬁculty assessing the value of. Upon further investigation, these were found to be metadata
ﬁles and thumbnails. We determined in the end that we would keep them.
Met with outgoing dean and transferred email account to library servers once he left the position. Outlook PSTs are
highly proprietary. Transferred deceased faculty member’s email account to library servers. Mac to Windows migration
was very time consuming. Email account is Eudora and no easy way to convert emails to less proprietary format.
Transferred digitized president’s ofﬁce correspondence from CDs to library servers. Transfer process took hours.
Obsolete ﬁle formats. Readability of legacy media. Lack of identifying information accompanying legacy media
(unlabeled, no contextual information).
Obsolete media storage. To date we have been able to outsource this to a vendor. Lack of any repository to store
or manage personal materials donated. We’ve taken in a few batches of material and have stored them with only a
promise of byte stream recovery and have temporarily turned other material away.
Obsolete media, ﬁle systems, and ﬁle-formats; e.g. 8” floppy disks, FAT variant disk formats, and WordStar ﬁles
(existing converters did not work). Data loss from media corruption. Managing the politics surrounding SEI/PII. Some
disks have content the donor did not expect to be there, was private, and outside of our collecting scope. Some capture
mechanisms are poor or incomplete compared to the original versions; e.g., social media and enterprise systems data.