Jon was a better pick for head archivist (the profession not the godhood) than Sasha according to me, a Jon apologist/Sasha stan with a masters degree in information science

1) a disaster pit of an archive like Gertrude's requires a work horse like s1 Jon, not someone easily lost in the weeds like Sasha. For a body of information that huge, with subject matter that broad (even just 15 fear subjects is broad when you consider the overlap and highly subjective, contrasting nature of the statements), you want to create as little work for yourself as possible until you've built a foundation. Ergo, in this early stage you want a project manager who has the ability to burn through busy work, resists distractions, and remains open to his team's input for future endeavors while keeping them on track.

I also have a hc that a disorganized archive bolstered the Eye's hold rather than undermined it because that meant that only Eye-aligned people had the ability to navigate it. So in a normal case, I'd assume they could just call up one of their sister archives and borrow their database or taxonomy, because why reinvent the wheel, but here I suspect that messy records were an industry standard designed to lure in people with a craving for tidy answers. It's possible that no-one at any location had successfully organized their statements, or at least not before getting indoctrinated with some mark or another and abandoning the task.

2) Referring to statements by their dates isn't a bad idea when first tackling a project this unclear. When you create a spreadsheet map of your body of information (literally just a giant inventory list), I was taught to also provide an ID column to designate each collection item with a uniformed shorthand for easy, clear communication and notes. Think dewey decimal. It can be something simple as sequential numbers or dates. Why use dates and not statement giver names? Confidentiality is likely a behind the scenes factor in statement acquisition. Some of the statements are anonymous. Some names double. And the series demonstrates that statement givers rarely visit twice in one day, much less twice a week or month, and even if there are, they can tack a .1 on the end or something. This shorthand ID key will likely change no matter what once they familiarize themselves with the content enough to build a more thoughtful catalogue. So since that makes this a placeholder anyway, dates as a good-enough-lets-keep-moving ID would be my pick, too, ESPECIALLY because of point #3