Monday, November 28, 2016

FOLIO, acronym for “Future Of Libraries Is Open”? I’d suggest: “Fantasy Of Librarians Inflamed by Organizations”

Introduction

There is a feature of librarian groupthink that is both a great asset and a tremendous liability to the profession. The asset is that Librarians have big and often inspiring hopes/dreams. The liability is that they don’t have the resources to achieve all of them. Nor do they have a good mechanism for synchronizing the two so that their hopes/dreams are in line with their resources. This can be inflamed by a willful refusal to examine the historical record, to extract lessons learned and apply them to the future. Worse yet, it can be further inflamed by organizations with vested interests.  

Such is the case with the OLE, now recast/renamed the FOLIO project. Now, let me say, as I have before, that I’m a huge supporter of open source software.  As the Chief Technology Officer of an R1 research university, I understand the advantages and disadvantages of it, and have worked with it in libraries and companies. In my library, we run a lot of open source software, including DSpace, Fedora Commons, Drupal, Omeka and many, many more core infrastructure packages.  It’s a wonderful platform and can offer outstanding value.  


So, it’s with dismay that I watch what is happening with FOLIO. Because we’re simply betting a great deal on a bad bet, as it is presently framed.  We need to inject some reality into what is being said and done. We need to separate fact from fiction and the associated marketing hype. I hope this post will help move the discussions in that direction.

I recently keynoted the 2016 University of Houston Discovery Camp and had the opportunity to hear a presentation on FOLIO. I also attended the Charleston Conference and heard a presentation there. One of the talks was very well crafted, but, like many marketing talks, glossed over the realities in order to paint the picture the speaker wanted to paint. Some of that talk aimed to address issues I raised in my keynote talk from earlier that day, including:

  • Understanding that the OLE code base is essentially dead and the at least the Library Service Portion (LSP) portion is being restarted from scratch.
  • Questions about the suitability of micro-services as the core architecture.
  • Questions about the governance structure.
  • What is the true target market?
  • Is the core architecture being used really multi-tenant?  What are the implications if it isn’t?
  • Is it realistic to imply the product delivery date will be 2018? 
These are all topics covered in my previous blog post on OLE. However, as a result of the presentations I heard at the Discovery Camp and the Charleston Conference, some new questions and points need to be raised and added to the list. These include: 


1.  Can the library community support the vision this project entails?  

At the Discovery Camp, there were over 130 academic librarians attending. I did a quick poll asking how many of those librarians worked in libraries which had programmers on staff? Less than 12 people (less than 10%) raised their hands (and note that some of those might have been from the same institution, but for the sake of discussion, let’s say they were not). I then asked how many of them had programmers waiting for projects to perform. ZERO hands were left in the air. I know I’m in that situation at my library. If I wanted to, it would be 1 - 2 years before I could assign anyone to this project for any significant amount of his or her time, and I have a moderately sized programming team for a library.   

When the FOLIO rep took the stage, the response to this issue was: “250 libraries indicated interest in providing development assistance, and there were a total of 750 attendees at 4 recent seminars.”  OK, let's take this apart and inject some facts and reality into that response.   


> First, librarians are usually spending frugally and they feel an obligation to investigate options so they can make informed choices. Open source software is often believed to be a lower cost option (although it’s certainly not always true,) so a library wants to evaluate the option. Their attending a session is clearly not the same as saying they’re definitely going to download the product and put it into production! It simply means they’re interested. 


> Second, 250 libraries saying they want to provide development assistance can mean very little. While it seems wildly out of fashion these days, let’s bring some facts into the discussion and look at the size of the development teams of companies/organizations that have built a Library Service Platform from the ground up. Let’s do this by using Marshall Breeding’s Library Technology website data on staffing levels for the only two organizations that have successfully developed, from the ground-up, Library Service Platform (not ILS!) products. Those would be OCLC and Ex Libris (ProQuest), so let’s focus on those. Next, using numbers from that chart and the number of products they’re supporting (from the same chart), one can do a rough calculation on how many hours were likely devoted to the Library Service Platform software development.  Then using the year those organizations have publicly reported they started their work on the Library Service Platforms, calculations would show that in total person years of development of these products, OCLC = 542, Ex Libris = 447.  

If there are 250 libraries saying they could provide development assistance (which again, doesn’t necessarily mean a programmer), and if 15% of those could provide a .25 FTE programming position on a continuing basis (I’d really like to know who those are as I can’t find any of them in my informal surveys) to help develop the product, and even if the commercial organizations involved devote another 3 FTE just to this development effort -- even if all that were true -- it would take decades of effort to match either Ex Libris’s effort or OCLC’s effort to date.  Now, it is true that libraries could proxy, by financial contributions, development resources that might mitigate this timeline.  However, given the state of higher education funding in the United States today, unless the private institutions take on this load fully (also unlikely) and even looking into the future, this still seems a very long shot.


Which is a nice segue to pointing out that, during the time FOLIO is being developed to bring it to the functional level of the current Library Service Platform products, those products won’t be sitting still. They’ll continue to move forward; and, therefore, FOLIO will always, always be behind the competitive offerings unless a LOT more resources are devoted to it!   


Let’s remember, OLE went on for eight years (starting in 2008) using this same proposed resource model and never did produce a complete production ready system that could be installed by all types of academic libraries.  As I pointed out in my last post, the OLE software only went into semi-production status in three research libraries because it was still missing a lot of functionality.  


Enterprise software requires enterprise level development teams, and we simply do not have that available to us in the form of library part-time open source software developers.   Enterprise development teams need programmers who can focus full time on the extremely complicated nature of the work they're doing in our libraries.  They also have the advantage of work towards filling a common defined product definition (although this is a point of contention for many librarians, as we'll discuss below).   

If you like challenges, just try to get 10 agreed upon functional definitions for any one major library function, out of a group of 250 libraries.  And if you think that's good, then be sure to think about why you're frustrated with your current ILS, which ties directly back to what it takes to maintain all those versions, to coordinate software development between them, to run them on different hardware platforms, to synchronize testing, documentation, releases and installations. In other words, yes, think back to the days of locally installed ILS's and back up the vision for your library by 25+ years. Tell you software developers to emulate that in the open source code they write/contribute because you enjoyed it so much. Oh yes, and be sure to call that new open source software you develop FOLIO.  Because that's what you will be creating.

It is, quite simply, insanity that for as long as libraries have been using automated systems, we can't find common vision or "shared-best-practices" for our workflows.  You can find it at the Dean/Director level, but it rapidly starts disintegrating as you get to the lower levels of the Library. It is truly a failure of the profession and its leadership.

Does anyone besides me see the vicious cycle here?  Quite simply, the answer to the question I posed at the start of this section is: “NO, no, we can’t support this vision.”  

2. Listen to what you’re REALLY being told by the “organizations” involved!  


I found it very interesting to note, that in the Charleston Conference presentation, done by an EBSCO rep, the following was said: "The code would be released in 2018." But they also said this: “The list of things that can be developed are (you caught that right? “can”!):

  • E-Resource management
  • Acquisitions
  • Cataloging
  • Circulation
  • Data Conversion
  • Knowledge Base
  • Resource Sharing 
And "that you will be able to expand" (again, note the highlighting here!) the Library Service Platform to include:
  • Discovery
  • Open URL Linking
  • Holdings Management
  • ERP integration
  • Room booking
  • Analytics and student outcomes
  • Linked Open Data
  • Data mining
  • Research data management
  • Institutions Repositories
  • And ….more!….
Once again, analyze what they’re saying. They’re telling us that the release date of 2018 is actually just going to be the kernel of the system, not the complete system. The good news?  That might even make that date doable. But it also confirms the calculation above, that if you want to have a complete system, you need to add on the years of development required to put all these other modules in place, which will mean a date more like what I’ve shown above, decades from now.

Now ask yourself a really hard question: Do you really believe that this list describes the functional system you’ll want, or even need, decades from now??  I doubt it. I know it doesn’t for me, not even if the numbers were far less, like a single decade from now.  

At the Charleston Conference, I heard James Neal, incoming President of ALA, say in his opening talk:  “I propose that by 2026, there will be no information and services industry targeting products to the library marketplace. Content and applications will be directed to the consumer, the end user." While I might argue his date, I will not argue the direction of his statement. It’s not that far in the future.  Think about that and the implications it has for FOLIO (and really, a lot of library software). 


3. “Follow the money” 

This is a saying that is frequently used to remind people that, in analyzing a situation, they should look at the financial benefits (and to whom they flow) in order to understand why something is happening. Let me suggest we do that here.  EBSCO is providing most of the financial backing for this project, which on the surface level seems quite laudable. They’re a good company with long history in libraries and are privately owned by a family with a long history of philanthropic activities. BUT. They’re also a major company selling a discovery product without an ILS or Library Service Platform that they own (unlike both OCLC and Ex Libris), thus they can’t provide the tight integration that those solutions provide. That puts them at a severe disadvantage in the marketplace.  So they have a vested interest in diverting you from buying one of those solutions, because they don’t want to lose discovery system sales or content sales. Even if they wanted to try to buy another LSP vendor/product to offer, they’re too late. If they were for sale at all, they’ve already been bought. So, EBSCO is in a squeeze. Again, this is a big disadvantage for them and is why they're driving the FOLIO product to move towards the Library Service Platform when, in fact, it should be moving to address market segments where we really do need new tools, i.e., research data management systems, knowledge creation platforms, open access/OA publishing systems, GIS systems, etc., etc. But that’s not where EBSCO wants to go and so since they’re providing the funding, the Library Service Platform is the focus.  


We also had an EBSCO VP standing on stage in Charleston, telling us: “The vendor marketplace is consolidating therefore we must develop alternatives!”  Really? It seems to me we DID develop alternatives (look at this chart Marshall Breeding compiles on this topic), and over a long period of time, the market has acted to consolidate and focus on the solutions that best meet its needs. In so doing it has narrowed and selected two superior products to fill the Library Service Platform niche i.e., WorldShare from OCLC and Alma from Ex Libris.  We don’t need more alternatives, the market has already spoken, and, for the foreseeable future these are the successful and widely adopted products.   Alternatives at this point are only going to further fracture a profession that needs to be spending its efforts on new areas which we allow librarians to add new value.  Simply put, at this point in time, that is not in reinventing the ILS as an open source software solution!


4. What is the governing structure going to be for this effort?  


Have you noticed how little is being said about this topic? Ask and they’ll point you to this webpage. Except there are really only two pages here, neither of which address the specifics of how this project will be governed. Things like:


  • Who sets code directions?  
  • Who decides and commits what gets into the code base? 
  • How are conflicts resolved? 
  • How is Open Library Foundation funded? 
  • By whom and in what amounts? 
  • Who are the board members? 
  • How do they get on the board? 
  • Who names them (EBSCO)?? 
  • How will they make strategic decisions for the organization? 
Without this information, there is cause for concern. Lest we forget, OLE went through multiple governance structures, all of which, after Duke University Library’s initial work on the project, ultimately failed.  A large part of this was documented well in this post. I’m truly unclear why we want to relive this, but apparently some of us do. Again, it seems to me we should learn from history here and, as a result, this whole governance issue should be cause for deep, deep concern.

Conclusion


I’ll say what I’ve said before. There is a real place for a FOLIO concept (i.e. mass community collaboration around open source software) in the library profession. The concept needs far more work in defining governance, sustainability and the guidelines for mass collaboration BEFORE it takes on the effort of developing any products.  What it will not be successful in doing at this point in time is producing a Library Service Platform. (And please remember, the concerns above are my latest ones in addition to all those expressed previously in my last post and only briefly summarized above. 


At the Charleston Conference, one of the librarian presenters, speaking about FOLIO, proudly said: “If FOLIO is a Unicorn, then I believe in Unicorns.”  I listened intently and thought: “You’d better, because that’s exactly the concept you're buying into here…”


There are a lot of good people involved in this effort, and for the sake of all concerned, we need them to do something that will be wildly and broadly successful for libraries. It's not in doing a Library Service Platform. We’ve already put somewhere between $5-10M in the failed venture of trying to build OLE for this purpose.  Now we’re queuing up, like tourists at a DisneyWorld ride, to do it all over again. Let’s stop fawning over FOLIO, as was done recently in Information Today, apply some of the vast facts we have at our disposal in our libraries, and realize we simply CAN’T afford to waste this money or time on this particular idea. 


Let’s stop this fantasy, pause, redirect and get pointed in a direction that offers real value-add and solves some of the many real problems that we really need to be solved.