What Next? Part 2: The Open Source ILS


Posted On May 30 2007 by

It’s true that I am one of the skeptics. I’ll state that up front. But, in truth, my skepticism toward building an open source integrated library system was born in optimism that the vendors of proprietary software would be paying close attention to the landscape. Alas, I don’t think they were.

Some tried
Most people have forgotten that Dynix wrote a white paper in 2003 called“Horizon Open Technology: A Vision for the Future” that was the basis of the original Horizon 8.0 (I looked all over for this white paper online and had to fall back on my own archive). The system would use Lucene and uPortal, and would support interchangeable back-end databases. The architecture was going to be open and modular. Within months of Dynix’s merger with Sirsi, Lucene was replaced with FAST, and uPortal was replaced with options like Horizon Information Portal and Sirsi’s Enterprise Portal System (EPS). But that’s all academic now that Horizon 8.0, too, belongs to the ages.

Another key aspect of open systems is the growth in open source technology. Not all open source technology is good, but there is tremendous promise here. Operating systems (Linux), application servers (JBoss), and search engines (Lucene) are among the more powerful open source developments that Dynix believes holds great promise and which the company intends to leverage.

–excerpt from 2003 Dynix Whitepaper

VTLS continues to tout its open source components, and was one of the first vendors to repackage Fedora (an open source repository application) and brand it Vital, essentially selling some of the interfaces and support for the underlying open source architecture.

Talis, a UK-based ILS company, continues to make strides in its support of open source services. The Talis Platform is the basis for many open source add-ons, thought the company’s ILS itself remains a proprietary system. Nevertheless, Talis remains nearly alone among “traditional” vendors proselytizing open source solutions.

Some finish
I’ve mentioned Koha and Evergreen on this blog and on the pages of American Libraries many times. LibLime has made some success of installing Koha at smaller libraries. Evergreen, the open source ILS created for the Georgia PINES consortium of public libraries, is now supported by its creators through a new company called Equinox. Roy Tennant called this “The Dawn of a New Era.” I prefer to call it a potential tipping point.

The open source tipping point
Libraries and open source are like peanut butter and chocolate to many people. Our altruistic nature, penchant for openness, and shoestring budgets make libraries a virtual Petrie dish for open source experimentation. OSS4LIB has been around for nearly (over?) seven years. Many libraries are probably using open source software without even realizing it . . . things like Linux and apache, sendmail or Zotero.

So, where’s the tipping point? Well, it might just be the commodity status that I mentioned in starting off this series. The authors of a book called Open Source 2.0 suggest as much. That is, when the software reaches commodity status, companies will emerge to provide services associated with that software. Redhat’s Linux is a very good example of this. The question is whether the ILS has reached this point yet. It could very well be barreling in that direction.

Some paradox
Still, there is the skeptic in me. I am inclined to ask some baiting questions and point out what I think are some serious issues in the open source movement. I think there is still a “build or buy” paradox in going the open source route. While the services that support open source are bound to be cheaper than buying a proprietary ILS, libraries will have to ask themselves why they want an open source system. It could be philosophical—either in wanting to support a move toward openness, or in wanting software that supports easier local development and add-ons. It could be financial, or the idea that “good enough” is often just that. The point is to make an informed decision—one that is based on the freedom of open source and not its relative free-ness.

There is also what I call the “marketing paradox.” You will hear from open source vendors that one of the benefits of using open source is that if you are unsatisfied with the associated services, you can just get them from someone else without having to change software. I don’t want to downplay this benefit: It would be great to not have to migrate every time there is a bump in the road. And let’s face it, there have been a lot of bumps lately. Nevertheless, this is a cultural change for libraries, but one that I think we should be open to. The good news for libraries is that the marketing of these services within this framework should naturally keep costs down, especially as more open source–supporting organizations emerge.

Finally, there is the issue of patents. MySQL’s Marten Mickos called them the Achilles heel of the entire software industry (see this interesting article from last summer’s internetnews.com). Practically the opposite of open source reaction, patents create a visceral response from most librarians—though I would argue that libraries are as likely to rid themselves of patented hardware and software as they are to rid themselves of copyrighted content in their stacks. How patents will impact the library software market and library software development remains to be seen. I eagerly await a thorough treatment of this topic (from a library point of view) from someone much more educated on the subject than I am.

Software is software
I have been relegated to quoting myself quoting someone else, but I first pointed this out in a column last fall. Dan Chudnov, creator of OSS4LIB, was the one who gave me this truism: [added text] software is software. This led me to think that [end of added text] it does not matter whether the software is open or proprietary; what matters is whether the software is good or bad. Those are my words, not his, but they are meant to capture [added text] what I took away from that conversation [added text] the essence of what he said to me over five years ago (some things just stick with you).

At the end of the day, the software has to work. If it doesn’t work, you have to be able to make it work. If open source makes for a better path to making things work, then all the better. If competition from open source makes vendors or closed-source systems more alert and responsive, then all the better. Better is good.

 

[This post originally appeared as part of American Libraries’ Hectic Pace Blog and is archived here.]

Last Updated on: January 19th, 2024 at 12:22 am, by Andrew K. Pace


Written by Andrew K. Pace


10 responses to “What Next? Part 2: The Open Source ILS

  1. In the end it does have to work – but there is a technical aspect to this as well as a usability aspect. Our system – which provides support to the libraries in 22 school districts in rural, western NY – carefully measured these two aspects of “it must work” and came up with a compromise. There are two major points of heavy lifting in an ILS. The first is everything to do with inventory control (the automation system) and the second is search (the OPAC). In the past, we have looked to companies that excelled at inventory control to also excel at searching. An odd expectation to say the least. But what if we could turn to multiple solutions? Therein lies the real power of open source.

    What we ended up doing in our system was deciding that we didn’t even want to touch the inventory control part of the puzzle. Our current ILS actually does a pretty nice job of inventory control. Their searching, however, is quite painful. So what if we just tackled the searching bit? Though we could have gone with Lucene itself, we decided to go with IBM Omnifind Yahoo! Edition for a pre-packaged Luecene installation for up to 500,000 records. We then used the open source CMS Drupal to handle the search results and display then in a more friendly library format (i.e. it works from a usability view) than we had before.

    We went into this knowing full well the challenges and potentials for open source. We call our system FISH – Free (as in Kittens) Integrated Search Handler – knowing that what we saved in money would be made up in time. However since we built this on established packages like Omnifind and Drupal, we were really only writing a limited amount of new code. Our last challenge is getting real-time status from our inventory control vendor into our system. The challenge is that the vendor doesn’t realize that the money/time paradox is better represented as a money/time divergence. I am willing to spend either time OR money to solve this problem, but not to spend a great deal of money and then still have to jury-rig a solution (which takes longer in the restrictions of proprietary code).

    The resulting portal is in beta at http://fish4info.org

  2. Great post, you provided a lot of food for thought. Oh, and I can work with “tipping point”. Maybe this is the latest quote I can steal from you. 😉

  3. he bloggers at the Institute for the Future of the Book have a related post (http://tinyurl.com/3auc65) titled “The People’s Card Catalog.” It is, in Christopher’s useful distinction, addressed to the search issue, not inventory control, and points to recent improvements in Google Book Search as suggestive of where we want to go (although not necessarily with Google) — a single node for each book, connected to reviews, references, and other related information including a link to WorldCat.

    The post and my comment come from the user perspective. If there’s a better search interface out there, whether open or proprietary, do libraries need to spend a lot of effort and/or money creating new ones? As increasing numbers of our users arrive at our OPACs from the web, what kinds of search tools do we need to provide? Is this a battle we all have to wage separately? Or can we each connect our inventory control systems to the networked interfaces that are developing Out There?

  4. Hmmm, interesting quote from Dan Chudnov, because when I made the same point at Code4Lib 2007, he heckled me. Well, whatever. I was anticipating a “four legs good, two legs bad” post from you, and I find you strangely receptive to open source. Disappointing, because who will be my straw person-of-any-gender?

  5. That’s definitely not what I said. It *absolutely* matters whether it’s free software or not, and your column last year had it right: all software can be good or bad, and meet your needs or not, independent of its license. What free software gives you, as you sort of state at the end, is the freedom to choose to take the stuff that nearly fits your needs or is bad today but shows promise for tomorrow and find a way to make it fit your needs more closely.

    I heckled you, Karen, because you emphatically said “there’s no such thing as free software.” I won’t sit quietly when somebody publicly denies something so important about the professional world I work in.

    To both Andrew and Karen: whether you are making a separate point or not, you would do well to more closely consider the actual words you’re using when you talk about free software (hopefully also when you informally misquote things I’ve said!) because your words count. There’s a worldwide community of free software developers out there right now without whom our libraries and the network infrastructure we depend upon to deliver services might very well collapse, and that community is vastly bigger than ours and has had a massive impact on libraries already. It doesn’t benefit anyone to belittle their efforts through cute similes (peanut butter and chocolate? kittens? WTF?) and misstatements of what’s important about freedom. Our community is capable of more substantive discourse where subtle (and not-so-subtle) meanings are drawn out explicitly, and talking down to us instead doesn’t raise any of us up.

    Love, -Dan

  6. Dan, as someone who has often been misquoted or had things said taken out of context, I want to assure you it was not my intention. I want to blame a badly placed colon in the sentence above, but laziness and inattention to detail are more likely suspects. I have never had to make such an edit to a post before, but I have made one now that I hope you find satisfactory. You bear no responsibility for what I take away from a conversation, even when you are party to it. You make librarians think, and that is a good thing.

    As for the cute language, that’s not really what I was going for. I’m more concerned about accusation of “misstatements of what’s important about freedom.” One of my concerns has always been that those who talk the loudest get heard the most, and when that talk is directed at decision-makers and comes coupled with “free and easy,” some decision-makers will jump at such an opportunity in an uninformed way. Let me be clear–I do not mean you. The Evergreen project is also NOT an example of this kind of mentality. They went into their projects with eyes wide open, thanks in large part to a serious and well thought out plan.

    You and I both know that one of those important things about freedom in software development–community development–does not always happen. We know that freedom is sometimes costly. But anyone who has developed applications next to an ILS also yearns for freedom.

    I guess I am trying to stake out some middle-ground. I’m frustrated that so many think it does not and should not exist.

  7. And I get frustrated because I know all of you, and I think you are all more on the same page than not. But the one thing out of this that I want to make sure is clear is that the word “free” is very problematic.

    “Free” can mean “freedom” or it can mean “no upfront monetary cost” and much of the sturm und drang I see around the open source issue is related to mixing up the two. In a way, it’s unfortunate that open source software typically (but not always) come without cost, since the most important point about open source is that it’s open. If it doesn’t do what you want, you are free (that word again) to change it. And that’s why Dan heckled Karen at Code4Lib and why they are both right in their own ways.

    So…I really want you all to kiss and make up, since I honestly and fervently believe that we all want and support the same things, or at least close enough to count for something. But we all do need to redouble our efforts to get the message right, and to watch our words closely enough that it means what we really want to say, and that what we say is the right thing to be saying.

  8. Roy – the thing is, there’s a huge community of people who’ve fought these word wars already, and it’s pointless for us to recapitulate them. If a member of our community (prominent or otherwise) presumes to stand before peers and pontificate about whether “free software” means what it sounds like or not, it’s their responsibility to do their homework themselves to ensure that they know what they’re talking about.

    The most important point about free software and open source is that it offers you freedom.

    Andrew – I don’t believe it’s a matter of finding middle ground. There’s nothing to take sides on, because as Roy suggests, it’s not that we disagree on principles. I just believe that they’re being articulated very poorly by people who can do better.

    I just don’t think it’s too much to ask of someone to spend enough time studying the free software movement and some of its glossy materials to be sure they get this stuff right when they stand before others (online or otherwise) to talk about it. Would you talk about copyright without understanding the implications of the DMCA?

    For the short course, start with these:

    http://www.fsf.org/licensing/essays/free-sw.html

    http://www.fsf.org/licensing/essays/free-software-for-freedom.html

    You don’t have to agree with everything you read there, but imho you have a responsibility to at least understand the issues at play, and to represent them fairly, if you want to speak usefully on these matters.

  9. The weird part is that Dan and I are much more on the same page than Andrew and I are, and I think Andrew misses some key advantages to open source that I delineated. But the fact is that just as matter can neither be created nor destroyed, someone, somewhere, has to expend resources to create and maintain software. Note that even with resources expended, Evergreen is still a major hoo-ha cheaper than any other route. But it’s misleading to say that software is not free.

    (I waited with great disappointment hoping to be heckled at NASIG, even after saying LOCKSS was better than Portico. Darn.)

    I also don’t think it’s helpful to take the fundamentalist approach that a discussion point has already been discussed and ergo there’s no use in discussing it, yada yada. That’s like the fundamentalist Xians who say, “the Bible said it, that ends, it, I believe it.” That is exactly what you are doing, Dan. You’re assuming that because I don’t agree with you I don’t know what I’m talking about. But the theories of the open source movement are just that, theories. It’s no more persuasive for you to club me over the head with the (often quite tedious) catechism of open source puritans than for someone to thump the Bible and tell me that their interpretation is the only one that matters.

    I guess my other concern is that you think my words are hazardous. I bet I have brought librarians over to open source because I *am* an independent thinker. For the most part, library administrators are not stupid cattle who do not “get it” about open source; they haven’t been convinced, for the most part, by arguments from their point of view. I can *tell* you about people who spout off about open source; I done saw it with mine own eyes. They are the problem, Dan. Even if you think I’m wrong (which I’m not), from a utilitarian perspective, you should be asking yourself about the market impact of my position.

    Roy, after being lectured like a five-year-old, I am not quite ready to give Dan a big ol’ sloppy soul-kiss, even though he and I are in the same sleeping bag though he doesn’t realize it yet (ew, I just grossed myself out). However, I *am* ready to go write a post debating Andrew’s points and sharing some perspectives about open source libraries should consider.

  10. I hope this can be the last comment along this thread, but invite the debate to continue. If Karen is truly going to blog to counter the points that I made (I had almost forgotten that I had made any) and she and Dan want to continue the debate, maybe they can do it over at her blog instead of mine. 🙂

    I have never edited a comment on this blog (I have asked one or two people to think a second time about the language they have used, which is what I think good moderation is about…a failsafe on the angry submit button), and I did not want to edit Karen’s words above, but Dan, I encourage you to not take the bait of being compared to a Fundamentalist.

    I mis-blogged a quote and tried to make up for it. I think the debate we are having is important, but the context has created a tempest in a teapot.

    And don’t blame Roy for wanting people he knows are friends to just get a long. You’re so northern California, Roy 😉

    Part three this week will take the vendor’s to task a little bit more. Finally something all of us can agree on?

%d bloggers like this: