Today is my last day here at Johns Hopkins University Libraries.
After Thanksgiving, I’ll be working, still here in Baltimore, at Friends of the Web, a small software design, development, and consulting company.
I’m excited to be working collaboratively with a small group of other accomplished designers and developers, with a focus on quality. I’m excited by Friends of the Webs’ collaborative and egalitarian values, which show in how they do work and treat each other, how decisions are made, and even in the compensation structure.
Their clientele is intentionally diverse; a lot of e-commerce, but also educational and cultural institutions, among others.
They haven’t done work for libraries before, but are always interested in approaching new business and technological domains, and are open to accepting work from libraries. I’m hoping that it will work out to keep a hand in the library domain at my new position, although any individual project may or may not work out for contracting with us, depending on if it’s a good fit for everyone’s needs. But if you’re interested in contracting an experienced team of designers and developers (including an engineer with an MLIS and 9 years of experience in the library industry: me!) to work on your library web (or iOS) development needs, please feel free to get in touch to talk about it. You could hypothetically hire just me to work on a project, or have access to a wider team of diverse experience, including design expertise.
Libraries, I love you, but I had to leave you, maybe, at least for now
I actually really love libraries, and have enjoyed working in the industry.
It may or may not be surprising to you that I really love books — the kind printed on dead trees. I haven’t gotten into ebooks, and it’s a bit embarrassing how many boxes of books I moved when I moved houses last month.
I love giant rooms full of books. I feel good being in them.
Even if libraries are moving away from being giant rooms full of books, they’ve still got a lot to like. In a society in which information technology and data are increasingly central, public and academic libraries are “civil society” organizations which can serve user’s information needs and advocate for users, with libraries interests aligned with their users, because libraries are not (mainly) trying to make money off their patrons or their data. This is pretty neat, and important.
In 2004, already a computer programmer, I enrolled in an MLIS program because I wanted to be a “librarian”, not thinking I would still be a software engineer. But I realized that with software so central to libraries, if I were working in a non-IT role I could be working with software I knew could be better but couldn’t do much about — or I could be working making that software better for patrons and staff and the mission of the library.
And I’ve found the problems I work on as a software engineer in an academic library rewarding. Information organization and information retrieval are very interesting areas to be working on. In an academic library specifically, I’ve found the mission of creating services that help our patrons with their research, teaching, and learning to be personally rewarding as well. And I’ve enjoyed being able to do this work in the open, with most of my software open source, working and collaborating with a community of other library technologists across institutions. I like working as a part of a community with shared goals, not just at my desk crunching out code.
So why am I leaving?
I guess I could say that at my previous position I no longer saw a path to make the kind of contributions to developing and improving libraries technological infrastructures and capacities that I wanted to make. We could leave it at that. Or you could say I was burned out. I wasn’t blogging as much. I wasn’t collaborating as much or producing as much code. I had stopped religiously going to Code4Lib conferences. I dropped out of the Code4Lib Journal without a proper resignation or goodbye (sorry editors, and you’re doing a great job).
9 years ago when, with a fresh MLIS, I entered the library industry, it seemed like a really exciting time in libraries, full of potential. I quickly found the Code4Lib community, which gave me a cohort of peers and an orientation to the problems we faced. We knew that libraries were behind in catching up to the internet age, we knew (or thought we knew) that we had limited time to do something about it before it was “too late”, and we (the code4libbers in this case) thought we could do something about it, making critical interventions from below. I’m not sure how well we (the library industry in general or we upstart code4libbers) have fared in the past decade, or how far we’ve gotten. Many of the Code4Lib cohort I started up with have dropped out of the community too one way or another, the IRC channel seems a dispiriting place to me lately (but maybe that’s just me). Libraries aren’t necessarily focusing on the areas I think most productive, and now I knew how hard it was to have an impact on that. (But no, I’m not leaving because of linked data, but you can take that essay as my parting gift, or parting shot). I know I’ve made some mistakes in personal interactions, and hadn’t succeeded at building collaboration instead of conflict in some projects I had been involved in, with lasting consequences. I wasn’t engaging in the kinds of discussions and collaborations I wanted to be at my present job, and had run out of ideas of how to change that.
So I needed a change of perspective and circumstance. And wanted to stay in Baltimore (where I just bought a house!). And now here I am at Friends of the Web! I’m excited to be taking a fresh start in a different sort of organization working with a great collaborative team.
I am also excited by the potential to keep working in the library industry from a completely different frame of reference, as a consulting/contractor. Maybe that’ll end up happening, maybe it won’t, but if you have library web development or consulting work you’d like discuss, please do ring me up.
What will become of Umlaut?
There is no cause for alarm! Kevin Reiss and his team at Princeton have been working on an Umlaut rollout there (I’m not sure if they are yet in production). They plan to move forward with their implementation, and Kevin has agreed to be a (co-?)maintainer/owner of the Umlaut project.
Also, Umlaut has been pretty stable code lately, it hasn’t gotten a whole lot of commits but just keeps on trucking and working well. While there were a variety of architectural improvements I would have liked to make, I fully expect Umlaut to remain solid software for a while with or without major changes.
This actually reminds me of how I came to be the Umlaut lead developer in the first place. Umlaut was originally developed by Ross Singer who was working at Georgia Tech at the time. Seeing a priority for improving our “link resolver” experience, and the already existing and supported Umlaut software, after talking to Ross about it, I decided to work on adopting Umlaut here. But before we actually went live in production — Ross had left Georgia Tech, they had decided to stop using Umlaut, and I found myself lead developer! (The more things change… but as far as I know, Hopkins plans to continue using Umlaut). It threw me for a bit of a loop to suddenly be deploying open source software as a community of one institution, but I haven’t regretted it, I think Umlaut has been very successful for our ability to serve patrons with what they need here, and at other libraries.
I am quite proud of Umlaut, and feel kind of parental towards it. I think intervening in the “last mile” of access, delivery, and other specific-item services is exactly the right place to be, to have the biggest impact on our users. For both long-term strategic concerns — we don’t know where our users will be doing ‘discovery’, but there’s a greater chance we’ll still be in the “last mile” business no matter what. And for immediate patron benefits — our user interviews consistently show that our “Find It” link resolver service is both one of the most used services by our patrons, and one of the services with the highest satisfaction. And Umlaut’s design as “just in time” aggregator of foreign services is just right for addressing needs as they come up — the architecture worked very well for integrating BorrowDirect consortial disintermediated borrowing into our link resolver and discovery, despite the very slow response times of the remote API.
I think this intervention in “last mile” delivery and access, with a welcome mat to any discovery wherever it happens, is exactly where we need to be to maximize our value to our patrons and “save the time of the reader”/patron, in the context of the affordances we have in our actually existing infrastructures — and I think it has been quite successful.
So why hasn’t Umlaut seen more adoption? I have been gratified and grateful by the adoption it has gotten at a handful of other libraries (including NYU, Princeton, and the Royal Library of Denmark), but I think it’s potential goes further. Is it a failure of marketing? Is it different priorities, are academic libraries simply not interested in intervening to improve research and learning for our patrons, preferring to invest in less concrete directions? Are in-house technological capacity requirements simply too intimidating (I’ve never tried to sugar coat or under-estimate the need for some local IT capacity to run Umlaut, although I’ve tried to make the TCO as low as I can, I think fairly successfully). Is Umlaut simply too technically challenging for the capacity of actual libraries, even if they think the investment is worth it?
I don’t know, but if it’s from the latter points, I wonder if any access to contractor/vendor support would help, and if any libraries would be interested in paying a vendor/contractor for Umlaut implementation, maintenance, or even cloud hosting as a service. Well, as you know, I’m available now. I would be delighted to keep working on Umlaut for interested libraries. The business details would have to be worked out, but I could see contracting to set up Umlaut for a library, or providing a fully managed cloud service offering of Umlaut. Both are hypothetically things I could do at my new position, if the business details can be worked out satisfactorily for all involved. If you’re interested, definitely get in touch.
Other open source contributions?
I have a few other library-focused open source projects I’ve authored that I’m quite proud of. I will probably not be spending much time on the in the near future. This includes traject, bento_search, and borrow_direct.
I wrote Traject with Bill Dueber, and it will remain in his very capable hands.
The others I’m pretty much sole developer on. But I’m still around on the internet to answer questions, provide advice, or most importantly, accept pull requests for changes needed. bento_search and borrow_direct are both, in my not so humble opinion, really well-architected and well-written code, which I think should have legs, and which others should find fairly easy to pick up. If you are using one of these projects, send a good pull request or two, and are interested, odds are I’d give you commit/release rights.
What will happen to this blog?
I’m not sure! The focus of this blog has been library technology and technology as implemented in libraries. I hadn’t been blogging as much as I used to anyway lately. But I don’t anticipate spending as much(any?) time on libraries in the immediate future, although I suspect I’ll keep following what’s going on for at least a bit.
Will I have much to say on libraries and technology anyway? Will the focus change? We will see!
So long and thanks for all the… fiche?
Hopefully not actually a “so long”, I hope to still be around one way or another. I am thinking of going to the Code4Lib conference in (conveniently for me) Philadelphia in the spring.
Much respect to everyone who’s still in the trenches, often in difficult organizational/political environments, trying to make libraries the best they can be.