the .

hbrarlanby m arr.hI _r_p• own

Direclor foi Innovative ýechnologies adele,r I,

VandprbO Unversity bbianLs

The Elusive Cost ofLibrary Software

;TANTIAL A question that I'm often asked involveshow much a library should expect to pay forautomation software. As much as I try to un-derstand all the various aspects of the libraryautomation industry, I have never been able

ARIES IN to discover a great deal of data describingwhat libraries actually pay for their automa-tion products. While I have an informal senseof initial license payment and annual supportfees paid by libraries of various types and

)NMENT sizes, it's based only on sketchy information.

VE HAVE Software Pricing: A Complex Matrix

Software pricing isn't a straightforward is-"CESS TO sue, since each procurement involves a spe-

cial business arrangement between a libraryand its chosen vendor. I think that it's rea-sonable to scale the cost of a product to suchfactors as the size of the library, the complex-

I LS THAT ity of the installation, the number of simulta-neous users, or the quantity of resources in-

PI N THE volved. While some may feel that it's odd fordifferent libraries to pay different amounts forthe same software, adjusting the cost by these

LIBRARY factors generally allows libraries with moremodest needs and more modest budgets to payless than those with more complex needs andlarger budgets. The service and support needsfor large and complex organizations cost more

IDUSTRY. to fulfill than smaller-scale installations. Li-brary automation isn't in the realm of shrink-wrap software. Rather, each installation is

unique, and it's advantageous to both the li-braries and the companies to peg the price ofthe software to appropriate indicators.

In the library automation industry, pricingis not only scaled according to multiple fac-tors, it's also a result of a private negotiation.A vendor will propose a price in response to arequest for proposals based not only on itsstandard formula of size and complexity fac-tors but also on the specific competitive situ-ation. A procurement with highly competitivebidding may result in a different price pro-posal than might otherwise be offered.

The broader economic climate also plays arole. Inflation, for example, will naturally betaken into consideration in comparing soft-ware prices over time. Prices may also varyaccording to international regions, scaled notonly to currency exchange rates but also ad-justed in tune with local business realities.

It's my understanding that most compa-nies have some kind of general formula thatthey use for calculating the base price fortheir software that includes such considera-tions as the modules required, options se-lected, the size of the bibliographic database,number of patrons served, the number ofbranches or facilities, and the number of staffusers. Yet this base price does not necessar-ily become the final bid, as other competitiveand economic factors come into play.

Hosting options invoke entirely differentpricing models. Software as a service (SaaS),

28 1 SEPTEMBER 2009 N wwwinfotodaycom

I SEE SUBS

ADV/

FOR LIBR

FOSTEF

ENVIRC

WHERE '

MORE AC

THE Fir,

DETAI

UNDEF

AUTO

IN

SVSIP_'1MS

th®

COMPUTERS IN LIBRARIES_1

the systems librarian

where the vendor hosts the software onits own servers, involves an annual sub-scription fee pegged higher than annualmaintenance on a software-only instal-lation, with or without an initial licensefee. SaaS has become an ever-increasingportion of the library automation econ-omy. This approach allows the library toreduce local hardware and technical sup-port costs involved in operating its au-tomation environment. SaaS, as a formof outsourcing, usually involves a re-duced overall cost for automation whileresulting in a higher concentration of thelibrary's technology budget flowing tothe automation vendor. While turnkeyarrangements that include both hard-ware and software are not as common asin previous times, it is important to fac-tor out the hardware component in anycomparisons with software-only deals.

As commercially supported opensource ILS products grow as a compo-nent of the library automation industry,a number of pricing questions arise. Thecomponents of public open source con-tracts cover a variety of services, in-cluding data extraction and migration,installation, configuring, hosting, andongoing technical support. Some mayalso include sponsorship of the develop-ment of specific items of functionalitythat will be incorporated into the soft-ware. All of these factors must be takeninto consideration as part of any com-parative study of pricing among opensource competitors and relative to con-tracts with traditional proprietary soft-ware vendors. I observe that the pur-veyors of open source and proprietarysoftware each make claims regardingthe lower overall technology costs. Asthe competition heats up among thecompanies providing support for opensource library automation products,comparative data also help to under-stand this niche of the industry.

AVeil of Secrecy

Many, if not most, software vendorsroutinely include confidentiality clauses

in license agreements and contractsthat prohibit the library from revealingthe price paid for the products and ser-vices procured. The prevalence of confi-dentiality requirements impedes theability of other libraries to have even ageneral sense of what the various prod-ucts on the market might cost them.

IT ONLY SEEMS FAIR

FOR LIBRARIES TO

KNOW THE PREVAILING

MARKET PRICE FOR

SOFTWARE PRODUCTS.

The absence of public information re-garding what libraries pay for softwareproducts and the associated annual feeshas a negative impact for libraries. Itleaves the library community as a wholewithout the ability to assess the impactof library automation software pricingoptions. This situation also leaves indi-vidual libraries in a weaker position fornegotiating with companies for thebest terms. Knowledge of what other li-braries of similar profiles historicallyhave paid for a similar package of soft-ware and services provides a bench-mark to assess any given price proposal.

Public Accessto Contract Terms?

It only seems fair for libraries toknow the prevailing market price forsoftware products. These representmajor investments for a library andvery long-term commitments. Giventhat libraries tend to operate a givensoftware product for at least a decade,the price negotiated at the time of pro-curement has incredible ramificationsfor the library's budget.

The Association of Research Li-braries recently issued a resolution thatencourages a more open approach to theterms contained within the agreementsbetween libraries and their suppliersfor content and software:

The Association of Research Li-braries (ARL) Board of Directorsvoted in support of a resolution in-troduced by its Scholarly Com-munication Steering Committeeto strongly encourage ARL mem-ber libraries to refrain from sign-ing agreements with publishersor vendors, either individually orthrough consortia, that includenondisclosure or confidentialityclauses. In addition, the Board en-courages ARL members to shareupon request from other librariesinformation contained in theseagreements (save for trade secretsor proprietary technical details)for licensing content, licensingsoftware or other tools, and fordigitization contracts with third-party vendors. (www.arl.org/news/pr/nondisclosure-5june09.shtml)

There has been recent activity in therealm of pricing for econtent packages.Library Journal, for example, recentlypublished a story describing efforts togain access to the terms of Elsevier'scontract with Washington State Uni-versity through a request for publicrecords. Elsevier filed a motion in courtto block access, which the court denied.In the same vein, many believe thatejournal pricing options need more pub-lic scrutiny so that libraries can assesstheir broad impact. I believe that sim-ilar benefits can be obtained by thestudy of the pricing of library automa-tion software.

Gathering Data onPricing and Expenditures

It seems that it should be possibleto gather pricing data for libraries that

wwwvinfotoday.orn x SEPTEMBER 2009 129

I COMPUTERS IN LIBRARIES

the systems librarian

rely on public funding. Even thoughthe contract itself may prohibit disclo-sure of pricing and terms, the budgetsof public libraries fall within the realmof public information. I've occasionallycome across the amount that a publiclibrary paid for its automation systemand the associated annual mainte-nance fees in the minutes of city coun-cil meetings or other public documents.It's my experience that this kind of in-formation can only rarely be discov-ered on the web. Specific budget re-ports may be available upon writtenrequest. Such data should also beavailable through Freedom of Infor-mation Act provisions. But obtainingpricing data in this way would not nec-essarily be feasible for any kind ofbroad research study.

I have often considered the possi-bility of providing a repository wherelibraries could voluntarily contributepricing information for their automa-tion products to the extent that theymay be legally allowed within theagreements that they made throughtheir contracts or license agreementsand by the requirement to make ex-penditures publicly available. Themore that such data could be gatheredcomprehensively, the more that itwould be possible to reveal trends inexpenditures on library software andthe financial impact of the competingbusiness models.

UnderstandingPricing Benchmarks

My view that there needs to be moretransparency in pricing for softwaredoes not necessarily imply that I be-lieve that libraries have been histori-cally overcharged. On the contrary, itseems that libraries sometimes haveunrealistic expectations regarding sus-tainable costs for the software toolsthat automate their operations in away that supports efficiency. I occa-sionally hear complaints regarding thehigh cost of library software, but it's

very difficult to provide perspective inthe absence of financial data. In mostcases, libraries validate that they re-ceive sufficient value for their invest-ments, as seen in the routine renewalof software support contracts.

IN MOST CASES,

LIBRARIES VALIDATE

THAT THEY RECEIVE

SUFFICIENT VALUE FOR

THEIR INVESTMENTS.

I generally believe that the pricesthat libraries pay for automation soft-ware would be proven as fairly modestcompared to those paid in other sectorsinvolved in similar kinds of activities.It would be an interesting study, forexample, to compare software costsbetween libraries and bookstores orother segments of retail. The lack ofsystematic data on the library sidemakes such comparisons impossible.It's also likely that gathering data onthe comparative sectors would beequally problematic.

Pricing and ExpendituresData Benefit Decision Making

It wasn't that long ago that a li-brary's key automation strategy in-volved simply selecting one of the ever-narrowing menu of proprietary ILSproducts. But today, a number of com-peting scenarios exist in the library au-tomation arena: open source versusproprietary, local installation versussoftware as a service, and OCLC'sforthcoming cloud-based model. Advo-cates of each of these models makeclaims of benefits in terms of overalllower technology costs.

In today's environment, where li-braries not only face this new slate ofautomation models but also face ex-tremely challenging budget scenariosdue to the difficult economy, we needall the data we can find to help makeinformed decisions about our technol-ogy strategies. A body of systematicpricing data might help guide librariesas they evaluate the options availableso that they can make informed deci-sions regarding which offers the bestvalue in short- and long-term technol-ogy costs.

Yet the creation of any resource thatinvolves the pricing of library automa-tion software must be done carefully.We've noted the complexities involvedin negotiating prices between librariesand vendors. The price that librarieshave historically paid for a given set ofsoftware and services represents justone of many elements to be consideredin making technology decisions. As faras my personal research, I'm inter-ested in assembling aggregate data ina way that will reveal trends related tothe competing implementation models.As I've mentioned in previous columns,I believe that the library automationindustry will benefit from a higherlevel of transparency. So among theother movements afoot to enable moreopenness, I see substantial advantagesfor libraries in fostering an environ-ment where we have more access to thefinancial details that underpin the li-

brary automation industry. U

Marshall Breeding is the director forinnovative technologies and researchfor the Vanderbilt University Libraries,the executive director of the VanderbiltTelevision News Archive, and the

founder of Library Technology Guides(www.librarytechnology.org). His emailaddress is marshall. breeding@vanderbilt.edu.

30 I SEPTEMBER 2009

COPYRIGHT INFORMATION

TITLE: The Elusive Cost of Library SoftwareSOURCE: Comput Libr 29 no8 S 2009

The magazine publisher is the copyright holder of this article and itis reproduced with permission. Further reproduction of this article inviolation of the copyright is prohibited. To contact he publisher:http://www.infotoday.com/