Meta Interchange

Rob McGee has been moderating the “View from the Top” presidents [of library technology companies] seminar for 26 years. As an exercise in grilling executives, its value to librarians varies; while CEOs, presidents, senior VPs and the like show up, the discussion is usually constrained. Needless to say, it’s not common for concerns to be voiced openly by the panelists, and this year was no different. The trend of consolidation in the library automation industry continued to nobody’s surprise; that a good 40 minutes of the panel was spent discussing the nuts and bolts of who bought whom for how much did not result in any scintillating disclosures.

But McGee sometimes mixes it up. I was present to watch the panel, but ended up letting myself get plucked from the audience to make a couple comments.

One of the topics discussed during the latter half of the panel was patron privacy, and I ended up in the happy position of getting the last word in, to the effect that for 2016, patron privacy is a technology trend. With the ongoing good work of the Library Freedom Project and Eric Hellmann, the release of the NISO Privacy Principles, the launch of Let’s Encrypt, and various efforts by groups within ALA doing educational and policy work related to patron privacy, lots of progress is being made in turning our values into working code.

However, the reason I ended up on the panel was that McGee wanted to stir the pot about where innovation in library technology comes from. The gist of my response: it comes from the libraries themselves and from free and open source projects initiated by libraries.

This statement requires some justification.

First, here are some things that I don’t believe:

• The big vendors don’t innovate. Wrong: if innovation is an idea plus the ability to implement it plus the ability to convince others that the idea is good in the first place, well, the big firms do have plenty of resources to apply to solving problems. So do, of course, the likes of OCLC and, in particular, OCLC Research. On the other hand, big firms do have constraints that limit the sorts of risks they can take. It’s one thing for a library project to fail or for a startup to go bust; it’s another thing for a firm employing hundreds of people and (often) answering to venture capital to take certain kinds of technology risks: nobody is running Taos or Horizon 8, and nobody wants to be the one to propose the next big failure.
• Libraries are the only source of innovative new ideas. Nope; lots of good ideas come from outside of libraries (although that’s no reason to think that they only originate from outside). Also, automation vendors can attain a perspective that few librarians enjoy: I submit that there are very few professional librarians outside of vendor employees who have broad experience with school libraries and public libraries and academic libraries and special libraries and national libraries. A vendor librarian who works as an implementation project manager can gain that breadth of experience in the space of three years.
• Only developers who work exclusively in free or open source projects come up with good ideas. Or only developers who work exclusively for proprietary vendors come up with good ideas. No: technical judgment and good design sense doesn’t distribute itself that way.
• Every idea for an improvement to library software is an innovation. Librarians are not less prone to bikeshedding than anybody else (nor are they necessarily more prone to it). However, there is undoubtedly a lot of time and money spent on local tweaks, or small tweaks, or small and local tweaks (for both proprietary and F/LOSS projects) that would be better redirected to new things that better serve libraries and their users.

That out of the way, here’s what I do believe:

• Libraries have initiated a large number of software and technology projects that achieved success, and continue to do so. Geac, anybody? NOTIS? VTLS? ALEPH. Many ILSs had their roots in library projects that later were commercialized. For that matter, from one point of view both Koha and Evergreen are also examples of ILSs initiated by libraries that got commercialized; it’s just that the free software model provides a better way of doing it as opposed to spinning off a proprietary firm.
• Free and open source software models provide a way for libraries to experiment and more readily get others to contribute to the experiments than was the case previously.
• And finally, libraries have different incentives that affect not just how they innovate, but to what end. It still matters that the starting point of most library projects is better serving the needs of the library, their users, or both, not seeking a large profit in three years time.

But about that last point and the period of three years to profit—I didn’t pull that number out of my hat; it came from a fellow panelist who was describing the timeframe that venture capital firms care about. (So maybe that nuts-and-bolts discussion about mergers and acquisitions was useful after all).

Libraries can afford to take a longer view. More time, in turn, can contribute to innovations that last.

My ALA Annual this year is going to focus on five hashtags: #mashcat, #privacy, #nisoprivacy, #kohails, and #evgils.

#mashcat is for Mashcat, which an effort to build links between library systems and library metadata folks. We’ve had some recent success with Twitter chats, and I’ve made up some badge ribbons. If you’d like one, tweet at me (@gmcharlt)!

#privacy and #nisoprivacy are for patron privacy. My particular interest in using our technology to better protect it. I’ll be running the LITA Patron Privacy Technologies Interest Group meeting on Saturday, (where I look forward to Alison Macrina’s update on Let’s Encrypt). I’ll also be participating in the face-to-face meeting on Monday and Tuesday for the NISO project to create a consensus framework for patron privacy in digital library and information systems.

#kohails and #evgils are for Koha and Evergreen, both of which I hack on and which MPOW supports – so one of the things I’ll also be doing is wearing my vendor hat while boothing and meeting.

Here’s my conference schedule so far, although I hope to squeeze in a Linked Data program as well:

In the title of the post, I promised mod_proxy hackery. Not typical for an ALA schedule post? Well, the ALA scheduler website allows you to choose you make your schedule public. If you do that, you can embed the schedule in a blog post using an iframe.

Here’s the HTML that the scheduler suggests:

There’s a little problem with that suggestion, though: my blog is HTTPS-only. As a consequence, an HTTP iframe won’t be rendered by the browser.

What if I change the embedded URL to “https://alaac15.ala.org/user/36364/schedule-embed”? Still doesn’t work, as the SSL certificate returned is for https://connect.ala.org, which doesn’t match alaac15.ala.org. *cough*

Rather than do something simple, such as using copy-and-paste, I ended up configuring Apache to set up a reverse proxy. That way, my webserver can request my schedule from ALA’s webserver (as well as associated CSS), then present it to the web browser over HTTPS. Here’s the configuration I ended up with, with a bit of help from Stack Overflow:

    # ALA scheduler needs SSL with a cert that matches badly
    ProxyPass /alaac15/ http://alaac15.ala.org/
    ProxyPassReverse /alaac15/ http://alaac15.ala.org/
    ProxyHTMLURLMap http://alaac15.ala.org /alaac15/

    
       ProxyPassReverse /
       SetOutputFilter  proxy-html
       ProxyHTMLURLMap http://alaac15.ala.org /alaac15/
       ProxyHTMLURLMap / /alaac15/
       ProxyHTMLURLMap  /alaac15/ /alaac15/
       RequestHeader    unset  Accept-Encoding
    

This is a bit ugly (and I’ll be disabling the reverse proxy after the conference is over)… but it works for the moment, and also demonstrates how one might make a resolutely HTTP-only service on your intranet accessible over HTTPS publicly.

Onward! I look forward to meeting friends old and new in San Francisco!