This week’s “On Location” visit with a collector ties in nicely with today’s topic: defining the requirements of your project. Before we start talking more about that, however, let me tell you a little bit about my new friend Irving (we talked for 3 hours – does that make us friends?).
Irving Leif has an interesting life story, but since it’s not mine to tell, I’ll let you catch a glimpse of it in this article, and I will instead focus on our visit together. I drove three hours west of Milwaukee to meet him in the small town of Viroqua, Wisconsin (a stone’s throw from the Mississippi River), where we spent the afternoon chatting in his home and then where his books are stored.
This was my first visit with a high-end collector as a part of this project, and I was not disappointed. Seeing all those one-of-a-kind pieces of literary history in my own hands (well, almost – he didn’t actually let me touch any of the books, lol) was a thrill, and I was taken aback by how long and how hard he has worked to build his collection. It is, quite literally for him, a lifetime achievement.
We talked for a while at his house, then he took me to where his books are, we looked at all those (pictures scattered throughout this post, obviously), then we went back to his house where I told him about this project and what I’m trying to do, at which point we not only talked, but argued (politely) as well, since – as I’ve discovered time and time again with this group – old-school collectors are definitely not eager to jump on board with new-school ideas. ;)
All in all it was a super fun day for me – great conversation, a fine collection, and six hours of lovely near-autumn driving through rural Wisconsin. (I’ll throw in a picture at the bottom of this post of the 6′ x 4′ wall map in my office, with a close-up of some of my driving record in the northern Midwest.)
Collector Directory: Defining the Requirements
One of the first things you have to do when you’re building a website that’s more functionality-focused and less marketing-focused is to define the requirements. I’m doing this with Steve (my developer) right now, and once we get the technical hurdles accurately described (so at least we know what we’re up against), then we can start looking at ways to clear them.
The two biggest requirements for the Collector Directory right off the bat are privacy and categorization. These directly follow from the two major goals of the tool: to discover and connect. Discover through a robust search function and proper categorization, and connect via personal email while still keeping the collector private.
I always knew that privacy would be a big topic on this project, and planned for it accordingly. At least I thought I had, but it turns out I didn’t take it as far as I needed to. Both from the survey results and from talking to individual collectors (including Irving), it seems that collectors *DO* want to be indexed so they can be found and contacted with interesting offers or topical questions, but they don’t want to be personally identified. That’s fine, I figured as much. But they *ALSO* don’t want their personal email address to be revealed, even when choosing to respond to an offer or a question! That one threw me for a loop.
Turns out collectors really are a finicky bunch, I guess (no offense). The good news is that I know a way around this problem – the bad news is that it messes with the process a little bit and makes the flow a little more disjointed than I’d like. But at the end of the day the tool needs to work exactly how they want it to, and that’s more important than how it looks or feels, so that’s what I’ll do.
This one’s a little tougher, and may very well be the most difficult part of this entire project. I’ll be building an index of thousands of collectors, organized primarily by the topic(s) they collect. Having them tag their collections with the appropriate keywords, and then having a tag feature and search function to parse through that information is easy enough, but the harder part is placing all this information on a visual, browse-able tree so that people can discover similar groupings of topics in the proper organizational structure (this also has implications for allowing dealers to advertise or sell into the appropriate groups later on).
Not only do I not know how to do that, but even if I do,, I need to figure out how to implement it for thousands of collectors and tens of thousands of topics without killing myself (or worse, killing Steve). So my first thought was to see if we could find out what governing or standards-setting body determines the appropriate hierachy for fields of interest, and see if we could pull an API from them.
The good news: After talking to a number of librarians, it turns out that pretty much yes, the Library of Congress Subject Headings (LCSH) provides the most universally accepted categorization structure.
The bad news: API support for these standards are weak at best, while additionally the sheer volume of data is massive. This is the index structure for only the letter “D” – it runs over 300 pages: LCSH Section D
I’m all for working with the proper standards right from the beginning, but as you can imagine, we’re in no hurry to try to integrate all of this uncertainty when we’ll be starting pretty small, not to mention the fact that our needs are a little bit unique and may evolve quickly. So right now we’re working through a couple other ways to tackle the problem, including a frontrunner that I think you all may find very interesting and appealing. But alas, that’s a topic for another time. For now, more pictures!