09 February 2010

When insanity reigns, someone makes a packet.

I know that I haven't blogged for a while, but I have been spending all my time tweeting for the past few months. I find that tweeting allows me to say what I want, in the very short time I have, without having to compose too much. However, I have been pushed into writing by a deep sense of frustration and even anger. I have been frustrated and angered by how we have all allowed ourselves to be bamboozled by a bunch of sharks peddling out-of-date wares for astronomical prices. More than that, how these bamboozlers have almost brought the whole system to its knees.
No, I am not talking about the bankers, though I certainly could be, nor am I talking about international corporations (don't even get me started!). I am talking about a small group of specialist software companies who wrote some intractable software for museums back in the 1990s and have been fleecing the museum community ever since.
I am not going to name any names here, that would be unfair and even foolish. Nor am I going to point the finger at my friends and colleagues who have bought, and continue to pay through the nose, for these "systems" (you know who you are). However, I want to say a few things about the roughly 8 to 10 major Collection Management Systems (CMS) which have come to dominate most mid-sized to very large museums for their documentation over the past 15 or so years.
First of all, a bit of background. Since the 1960s museums, or organizations supporting museums, have been trying to write the all-singing all-dancing CMS. Different groups started at slightly different times and in different places, but you can look up the work of the Museum Documentation Association (now the Collections Trust) in the UK, CHIN in Canada, or CIDOC at ICOM internationally. However, the Museum Documentation Association and CHIN were the earliest, as I recall. After numerous local, national and international attempts, all of which failed, there was created enough of a consensus, or at least enough of a consensus that could be imposed, to build some more or less stable CMSs. With these "standards", a number of software companies build some rather clunky CMSs using out-of-date or aging database methods and interfaces and began to sell. Oh, how they sold!
You see, museums then, and to a large extent now, are not very technologically sophisticated. I know that there are a number of exceptions, but the exceptions prove the rule. Museums, like so many large public institutions (health services, governmental agencies, security services, etc.) like big, expensive and largely unworkable systems to run their organizations. Museums are no different, so they bought these CMSs like they were hot cakes. Now, everywhere you go, at least practically every museum you go to, has one of these absurdly complex, expensive and clunky systems "managing" their data. The problem is that these systems manage very little. In fact, they provide very little service at all, often for thousands of dollars every year.

Here are a few deficiencies that I have found with these products.
1) When you buy one of these systems, you buy the structure of the data, "as is." This is regardless of the kind of museum you are, the size, the mandate, the audience and the collections. Of course these companies provide you with a service where you can customize your structure, to a degree, but only if you are willing, and able, to spend a vast amount of money. How much money? Often tens of thousands of dollars just to get a few fields to be added or changed.
2) The whole point of a CMS is to help the institution organize its information. That means that you will need not only some sophisticated means finding and organizing the vast records that these systems create, but also reporting the results in an impressive array of reports. Most of these systems, however, provide you with anywhere from two to a dozen fixed reports, mostly just lists, that you can't change either. Again, if you need another one, you pay, and you pay a lot.
3) We live in an age of the Web and sharing of our information. We all use, daily, systems that help us organize and share our information on-line with a whole host of different communities. One would think that our CMSs should do the same, but no. These leading CMSs provide only the most basic web modules and output systems. None can effectively link to on-line services, few have APIs and all cost thousands of dollars to add.
4) Finally, though I could go on and on, is the user interface. We have had over 40 years of HCI expertise built up in the software industry. We are all use to using quite sophisticated user interfaces on line and in our applications. So why are we, as museums, asked to pay tens of thousands of dollars, or its currency equivalents, and almost that much again each year, for a user interface that, frankly, I could have pulled out of my arm-pit in the 1980s?
Tens of thousands of dollars for what? An embarrassing user interface to an out-of-date application that hasn't even realized that the rest of the world is well into Web 2.0 and moving towards the Cloud? No thank you. I have been building CMSs for small to medium museums for over 30 years. They are easy, they are simple, and they can do a huge amount more than you imagined. More than that, they are very cheap and easy to implement and manage. You can go out tomorrow, buy yourself a copy of FileMaker Pro 10 Advanced, give it to your 11 year old son or daughter and let them build you a powerful, sophisticated, user friendly and web-savey CMS over the weekend. If that is too easy for you, join one of the two open-source CMS projects that are now approaching completion (CollectionSpace or CollectiveAccess).
I am calling on the museum workers of the world, "Cast off the shackles of exploitation." "Unite around the simple, the usable, the effective, and the cost-effective." "Free our documentation so that it can serve our audiences." It is not a dream, it can be done.


  1. Well done, Robin! I have been wanting to say much of the same about library CMSs for a long time. I am shocked and appalled by not only what libraries and museums will put up with, but what they pay for.

  2. Many many thanks Christine. My bile too has been bubbling up for a long time. I thought that these systems were an embarrassment when they came out back in the 1990s, but now they are just a disgrace.

    What worries me the most about these systems, however, is the institutional culture they represent. They are the system manifestation of a culture of centralization and control that is no longer acceptable. I think they remain popular because they maintain the museums, libraries and archives as gatekeepers not just of the collections, but of knowledge itself. This is now unacceptable and has to change. I think that a good start would be changing our CMSs so that they are actually accessible.


  3. As a developer who has been building collection management systems for the last 20 years, I think your post over-simplifies the history of museum software and the value of different models. I build museum software because I’m passionate about museums, not because I want to ‘make a packet’.

    Software development evolves over time. New products grow with the benefit of standards and functionality already developed and with the hindsight of lessons learned from the previous generation of software. The Internet did not exist when many of the current collections management systems were built and it is having a huge impact on the way software is developed.

    Museum systems do have some complex problems to tackle, such as the need to catalogue a diverse range of material (books, archives, biological specimens, photographs, equipment), manage related digital files, intelligently handle imprecise dates, and browse structured terminology lists. Museums range from small volunteer run organisations to those with hundreds of staff. One solution doesn’t suit all of their needs.

    Open Source is here to stay and is a paradigm shift for software development. Open Source Software as we know it today wasn’t possible in the 1990s – the web is an essential part is the making Open Source Software work because it provides ways to collaboratively share, discuss and contribute to the source code. Open Source Software isn’t free and it isn’t a magic bullet. With both proprietary systems and open source the museum needs to consider the long-term viability of the vendor or community, what support they will have access to, and whether they have the technical resources to manage the system.

    Some new CMS packages, including CollectionSpace and CollectiveAccess, still have detailed functionality. This goes beyond what an individual museum can knock together in Filemaker Pro. The functions are needed for larger organisations and do provide a service. I doubt that large museums making use of CMS software (open source or otherwise) to manage tasks such as loans, reproduction requests or exhibitions find they’re getting little or no service out of these features. Like many existing CMS packages, CollectionSpace is designed using the Museums Documentation Association’s SPECTRUM standard. The ability to tailor the data structure per museum and easily build ad hoc reports are problems that also need to be addressed by web based systems. Being Open Source or web based is no guarantee that your concerns will be any better addressed.

    Existing CMS packages need to update their user interfaces and offer a range of ways to share information, especially on the web. Making these changes is harder with a large base of existing users on a variety of environments. However, developers that fail to update their products in the light of newer standards or methodologies, or charge unfairly for making these changes, are unlikely to survive.

    Museums also need to be cautious when thinking of building their own custom systems. Who will maintain these systems? Is software development part of the museum’s mission? If a museum wishes to tackle software development, contributing to an open source project would be a better way to go.

  4. Dear Paul,

    Thank you so much for your long comment. I knew that this would raise a stink, and I am glad that it has. I accept that my post has over-simplified a complex history. I began my career with GOS and MUSCAT and have lived through that history. I also am happy to accept that the people developing, and buying, these systems are sincere in what they are doing. However, sincerity and passion are not excuses if we are doing things wrong.

    I am also happy to accept your point that museum systems are complex and diverse. Having been a contributing author for all three SPECTRUM standards and the current British Documentation Standard, I know all too well what those complexities are. I have also argued for over 20 years that the goal of a single, or even dominant, documentation standard is a false god and will actually hinder access. I agree, further, that the two open source systems under development (CollectionSpace, of which I am a member of its Executive Advisory Committee, and CollectiveAccess) are also grappling with these issues. They too have a long way to go and are moving, in my mind, too far down the all-seeing all-dancing system.

    However, I do not agree that that the problem for existing vendors, proprietary or open source, is accommodating the variety of environments. This simply perpetuates an assumption, that was always contentious and is now becoming untenable, that museums are uniform institutions with uniform information management needs. Or worse, it assumes that information, or information use, is uniform.

    The real problem is not simply that the existing systems are clunky and not fit for purpose, but also that the very principles by which we document and manage our collections are proving untenable. The assumption that management of information is singular, stable and optimal is false, as is the assumption that standardised descriptions enhance access. Some of us have been saying this from the beginning, but I am happy now that this is proving to be true. You need only to look at what is happening in Libraries (e.g. Christine Borgman, Geof Bowker and Susan L. Star), and also to see the new SPECTURUM standard and its abandonment of data models.

    I know that my final statements about FileMaker, and the open source systems was largely a rhetorical one, I assumed that that was obvious, but my point was not altogether rhetorical. For most museums, especially ones with small collections, their information needs are relatively simple. A simple, and cheap, proprietary system, or even a free service, would do them just fine, and would provide them with far more than they have now. Further, it is misleading to suggest that buying into these proprietary systems provides necessary support. You and I both know that to buy into such a system requires at least one full-time post, if not more. Enormous levels of local support is needed simply to implement and run such systems.

    I think, and it is but my informed opinion, that in an age of open source and cloud based apps, proprietary vendors have to do much better than they are to justify the investment by museums into the costs of their systems and the costs of their maintenance. However, I would also argue that museum themselves have to do much better at understanding and becoming expert at our own diverse and distributed information needs.

  5. Thanks for your thoughtful response.

    When I referred to supporting a variety of environments, I was meaning the technical environments which are currently in use. For example, the various versions of Windows and different approaches to implementing a wide area network. Many of the changes in computing have simplified these problems. A year ago we released a software as as service system for small museums - eHive - and we have found that running a hosted web-based system makes it much easier to support a variety of desktop systems and allow access from more locations.

    An all-encompassing system is something which many vendors have strived for, but it's very difficult to achieve a result that even all similar sized museums are satisfied with. Newer approaches such as plug-in/module driven architectures (looking to successful examples in content management such as Drupal and Joomla for example) or smaller systems that do once function very well and interconnect through web services or web hooks, seem to have more potential.

  6. OK Paul, now the penny drops. You had me worried at first. I thought you could be from another CMS developer. No, I am still not going to name any names, but I am happy to say that I know of eHive and I wouldn't include it in the group I discussed above. Don't get me wrong, I have issues with eHive and your CMS, but they are just issues. eHive isn't going to make me break out in a rash and want to break something, which many of the others do.

    Of course as an apps developer you have to worry about a whole lot of things that people don't normally think of like versions of Windows (horror!). And eHive has gone down the right direction in thinking about a web-based service, in my mind. I am also glad to hear that you are thinking about the potentials of web services and web hooks. I have been involved with Jeff Lindsay for some time and we are implementing web hooks on our CMS this spring.

    My reason for taking such a hard line in the above blog was, as I think you might agree, because there remains so many CMSs that have not moved beyond the 1990s (and early 1990s at that). The reason, I think, is not just that the companies are uninterested in change, but because museums are uninterested in change. As I travel around and talk to colleagues in museums, I can see the frustration of many, but others actually relish in the antiquated rigor and opacity. I think that this is a serious obstacle to engagement with museums on-line. Only when we get some CMS developers who actually care about access, use and diversity will thing begin to change. I am happy to tell you, as the eHive developer, that I do see this happening. However, I only know of 4: you, Madrona in Victoria, CollectionSpace and CollectiveAccess. Perhaps you know of some more.

  7. Dear Robin
    I know what you feel. And also all of you.
    As a visitor I feel the same anger each time I get in a Museum. That's why I invented Percipio, a behavioral peripherial.
    You do not need to do nothing. Just walking as you wish, in and out, looking as you wish, how you wish and the system is following you in the real sense of the word.
    It will reconstitute for you, your own trip with your preference, your "visual behavior"

    This system is the answer to your dream.
    Have a look.

    And if you want to try it, be my guest.
    It is fully operational

    I will be delighted to answer all question about this revolution.
    yves APELOIG / eshkar