Choosing a Hand-Held Inventory Device Previous   Contents   Next Issues in Science and Technology Librarianship Summer 2008 DOI:10.5062/F4251G4D URLs in this document have been updated. Links enclosed in {curly brackets} have been changed. If a replacement link was located, the new URL was added and the link is active; if a new site could not be identified, the broken link was removed. Choosing a Hand-Held Inventory Device Lois Green Arts & Humanities Library Assistant lag2@psu.edu Verne Neff Annex Library Supervisor van1@psu.edu Janet Hughes Biological Sciences Librarian jah19@psu.edu Trish Notartomas Access Services Information Technology Consultant tan3@psu.edu The Pennsylvania State University University Park, Pennsylvania Copyright 2008, Lois Green, Janet Hughes, Verne Neff, and Trish Notartomas. Used with permission. Introduction In spring of 2006, a task force was charged to look at the feasibility of acquiring hand-held inventory devices for the Pennsylvania State University Libraries (PSUL). The task force's charge was not to look at the whole concept of doing an inventory, but rather to focus on the feasibility of acquiring hand-held devices to use in an inventory. There were several criteria that needed to be considered for the recommendation. PSUL uses SIRSI Unicorn as the basis for its catalog, The CAT, and SIRSI Workflows for the underlying acquisitions, cataloging, and circulation modules. The devices would therefore need to be supported by SIRSI. They would also need to be compatible with PSUL work practices. Devices must be able to be loaned to units around the state so they could conduct physical inventories, meaning durability, portability, obsolescence and lifecycle of such equipment had to be considered. The task force also tried to identify a funding source, such as a grant, and possible collections to test the devices on before making them widely available. The WorkFlows Inventory Guide (Deuink et al. 2005) prepared in October 2005, discussed the basic reasons for doing a physical inventory. A literature search provided further justification for doing a physical inventory as an overall concept, and from informal discussions it was apparent that doing an inventory could enhance the usability of specific collections. Physical inventories of collections may provide a physical count of the number of volumes in a collection, may allow for examination of the condition of the collection, and most importantly, can reconcile the catalog records with the actual collection holdings (Deuink et al. 2005; Emery 1990; Wild 2002). During an inventory, exceptions in the collections can be identified and corrected. Exceptions include misshelved books, books belonging to other campuses or institutions, unbarcoded books, barcoded books not in the catalog, barcoded books not connected to item records, and books whose listed status does not match reality, such as books on the shelf that are listed as checked out or missing, missing books that are listed as being on the shelf but are not present, etc. The task force focused on the equipment itself, with the assumption that the obstacles to doing an inventory, such as the costs, the time commitment, and the additional workload to identify, reconcile, and correct, any omissions and errors in the catalog and item records, would be addressed elsewhere. The WorkFlows Inventory Guide did begin such discussions. This paper discusses the first phase of the task force's charge, investigating the feasibility of choosing and acquiring hand-held inventory devices. Results of the actual inventorying process have been delayed due to unforeseen problems in acquiring the devices that were recommended. Only recently have the devices been delivered and they are still not fully operational. After the devices have been tested and some small collections have been inventoried, further dissemination of the triumphs and pitfalls of doing an inventory will be pursued. Background The Pennsylvania State University Libraries' collections comprise over five million items distributed in many separate collections, in dozens of buildings, and scattered over several dozen campuses across the state of Pennsylvania. Inventorying even a small portion of such collections would be a daunting task. However, doing a physical inventory was a desire expressed by many of the PSUL faculty and staff. The CAT, the PSUL catalog, contains records for most of those items, but, as in any large collection, whether the CAT records reflected the actual holdings was unknown. An inventory would enable the PSUL collections to be reconciled with the CAT. The task force discussed how an inventory would benefit collections, and how hand-held inventory devices might be used during an inventory. How to handle problem items, such as unbarcoded resources, was discussed because such procedures might affect what kind of functionality would be desired in hand-held inventory devices. At PSUL, there is currently a project to shelf read the collections. This was discussed to determine how it might affect the desired functionality of inventory devices if the shelf-reading project were coordinated with an inventory project. Finally, the task force also discussed inventory projects done at Schuylkill and York campuses, as mentioned in the WorkFlows Inventory Guide (Deuink et al. 2005). Assumptions and Methods We performed a literature search to assess how other libraries have approached inventories, and to determine what equipment other comparable libraries have used to do inventories. Further, we sent out a query on several discussion lists to gather information on equipment, problems, timelines, etc. from comparable institutions. We conducted a web search to find information on possible devices, and to discover any other libraries that had conducted similar assessments. The phrase "supported by SIRSI" was interpreted in a very broad way because SIRSI recommends only one or two options. We thus included devices that could be made compatible with SIRSI. Having devices that would be compatible with normal work practices was interpreted to mean they would suit desired goals and purposes and could be integrated into the PSUL workflows and procedures. We did not address exactly how the inventory process would fit into the workflow of shelving, shelf reading, cataloging or other technical or access services processes, other than considering if the choice of device might have an impact on such processes. With those assumptions and interpretations, the task force addressed the options for hand-held inventory devices, outlining the advantages and disadvantages inherent in each option, and discussing variations within options. Costs, durability, ease of use, how well such devices would interact with SIRSI, and how well such devices would meet potential needs, were all considered. Part of the task force's charge was to find alternate funding sources, but thus far no specific grants that might be pursued were identified. Several libraries noted they used "special funding" but most indicated that it was internally allocated funding, not grants or other outside sources of monies (Emery 1990; Wild 2002). The task force made a few observations on aspects of the inventory process that could be promoted in finding sources of funding. Real Time Records Updating versus Batch Updating Inventories can be done in real time or in batch loads. Real-time inventorying allows exceptions to be noted immediately and pulled or corrected at that time. However, real-time updating has several limitations. It requires connections to the server through the local area network (LAN) or via wireless (WiFi or Bluetooth) data transmission. Connecting to the LAN via cables to accessible ports is not feasible considering the area that would be involved, and wireless connections have limitations on usability. Not all PSUL buildings are WiFi enabled, and some other modes of wireless data transmission, such as Bluetooth, are extremely limited in range. Furthermore, real-time updating of records might slow down the inventorying process, and requires greater vigilance on the scanning staff's part. Also, if the errors were to be corrected immediately, scanning staff would have to be trained in procedures such as adding barcodes to WorkFlows records, creating CAT records, adding volumes or copies to records, and other such editing procedures. Batch loading circumvents the problem of environments where wireless data transfer is not feasible. Batch uploading also allows quicker scanning in the stacks and less training of the inventorying staff. However, it forces other staff to upload the data, run reports, print out exception reports and go back into stacks later to locate and pull problem items. If the locating and pulling is not done promptly, exceptions might be moved or become otherwise difficult to locate. Thus, it would require a commitment to locate exceptions as soon as reports are run. Also, pulling items for later processing might create backlogs of exceptions that are sent off to various departments for processing. If no provision has been made to expedite processing, the backlogs might languish on book trucks, creating even more discrepancies between the actual collections and the collections according to the CAT. Real-time updating was the preferred option, but with the understanding that this would not be possible in many cases. Equipment Options There are four main equipment options for doing a physical inventory. A fifth option is to pull books from the collection and bring them to a workstation, as was done at the York Campus library (Deuink et al. 2005) and at Brooklyn College (Wild 2002), but this is not our preferred procedure and does not require hand-held devices. The main equipment options are using simple hand-held barcode scanners that can be wireless or can be synchronized with the system in batches; using hand-held Personal Digital Assistant (PDA) devices with attached barcode scanners; using mobile hand-held computers (also called portable data collectors) with integrated/attached barcode scanners; and using laptops with barcode scanners, although laptops are not, strictly speaking, hand-held devices. Within those options, there are variations in how they work and what other equipment and software are required. Some equipment does batch uploading of data, some can do real time updating, and some can do both. There are corded options and cordless ones. Corded ones attach directly to circulation terminals, laptops or other computers, and usually use those computers for their power sources. However, dragging hundreds of feet of cable into the stacks is not safe, nor desirable. Therefore, we considered only cordless options. There are plugged-in versions and battery-operated versions. However, carrying long extension cords into the stacks would not be feasible, so battery-powered versions were preferred. For units that are not simple barcode scanners, there are usually versions without barcode scanners that require inputting the barcodes through keypads, touch screens, or other means. However, such input methods are not efficient; thus units with attached or integrated barcode scanners were preferred. There are units that produce data that could interact directly with SIRSI, and ones that would need application program interface (API) programming to allow data to be used in SIRSI. Direct compatibility with SIRSI was preferred but not mandatory. Then there are options of size, weight, shape and other such parameters that were considered. Cordless Hand-Held Barcode Scanners Simple hand-held barcode scanners offer several advantages and many disadvantages. These are available in many portable configurations, shapes, and sizes, and from many different vendors. Some popular vendors include Symbol Technologies (now {Motorola Enterprise Mobility}), Hand Held, PSC and Intermec. Cordless hand-held barcode scanners can be batch scanners that work offline, scanning and storing barcodes to upload their data in batches. Offline scanners can upload via wireless connections (WiFi 802.11 b/g or Bluetooth 2.4 GHz radio frequency), docking stations, or cables, including USB connections. Alternately, cordless hand-held barcode scanners can be wireless, and work in real time with radio frequency (RF) data transmission that can range in distance from a maximum of 50' to a maximum of 150'. No units were found that used WiFi for wireless data transmission. Cordless hand-held barcode scanners have major advantages: Cost: relatively cheap and can be found for under $200 to over $2,000 depending on the features and configurations; batch scanners tend to be less expensive than wireless scanners. Durability: few moving parts; some gun-type scanners have triggers; others are constantly on; more durability usually increases weight. Longevity: most of these systems do not change much from year to year, and would not become obsolete quickly; also, many vendors mention the possibility of upgrades and improvements. Distance reading: most use lasers which do not require the scanner to touch or be within 1/2" of the barcode; can read at large range of distances from a few inches to several feet, allowing for remote scanning, although increasing the read distance increases the cost. Variety: wide range of shapes and sizes to suit different preferences, from ring scanners, to wand/pen scanners, to gun scanners. Ergonomic: gun scanners are the most common and are generally ergonomic and easy to use; however, wireless gun scanners can weigh about one pound. Use time: usually have long battery life and can often purchase quick chargers, extra batteries, and other accessories to prolong use time. Cordless hand-held barcode scanners have several disadvantages though: Batch uploading = Batch problem solving: batch scanners do not allow real-time updating of data; problems must be addressed after data are uploaded and reports are run. Compatibility: most scanners are used in retail or industrial applications so they are not directly compatible with SIRSI; sytems are often designed for data to be uploaded into spreadsheets so may require application program interface (API) programming to allow data to be used in SIRSI. Lack of functionality: most of these devices only scan and decode; they cannot be used to correct errors, run reports, or otherwise do any database maintenance. Limited range: real-time updating of data with wireless scanners is limited to range of data transmission (30'-150'). Error alerts: generally only sounds used to alert for errors, and even then, such alerts would not explain the errors; for real-time updating it would be preferable to have displays to note errors or else would require a person at the workstation to monitor any error that pops up on the host workstation and some means for the workstation person to instantly notify the scanning worker of any errors. Functionality tradeoffs: laser scanners allow distance reading, but are more expensive than charge coupled device (CCD) scanners and heavier than wand/pen scanners that require close contact with barcodes; displays allow user to see if errors occur but add weight and cost; color displays add considerable cost. Weight: wireless units weigh almost one pound, and may be too heavy to use without a harness or other carrying method. Offline cordless hand-held barcode scanners include: Telxon 64K model #747 with light pen scanners (price unknown); no longer manufactured -- Telxon (now Symbol); can store 4,000 barcodes; used at University of Waterloo (Emery 1990). Symbol P360 Scanner ($694); can collect up to 53,000 barcodes and runs for up to 12 hours on one charge; upload data via cradle; also works in corded mode; simple keypad and display. Symbol CS1504 ($110); holds only 150 barcodes but are cheap, small, and very lightweight (keychain sized); no keypad, no display; used at Purdue according to Judith Nixon (personal communication March 13, 2006). Wireless cordless hand-held barcode scanners include: PowerScan RF Industrial Bar Code Scanner ($548); radio frequencies to transmit up to 150' away; very rugged, operates for up to 8 hours per charge; weighs about 13oz (not including base station). Symbol LS 4278 ($350); Bluetooth (2.4GHz RF) for up to 50' (more likely 30'); low power consumption; no keypad, no display; weighs about 1/2 pound; can be used in batch mode. Intermec Sabre 1552 Scanner ($933-$2250 if sytem includes base station); uses 2.4GHz RF (Intermec PicoLink wireless personal area network) for up to 50' away from the MicroBar 9745 base station; use time is about 14,000 scans; recharge time is 4 hours. Intermec ScanPlus 1802 Vista ($690-$850) with MicroBar 9735 base station; send data via narrow band (433MHz and 908 MHz) to allow two-way communication. Summary Wireless barcode scanners allow real-time updating, which may be desirable. However, the effective range of most wireless units is only 50' and this would limit the usefulness of such a scanner. Also, the cost of both the equipment and the need to program for the results to be compatible with SIRSI make this option less desirable. The offline version is less expensive and is not limited in range, but it does only batch loading of data and has the same compatibility issues as wireless barcode scanners. Selecting a unit like the Symbol LS 4278 would give the best of both options, because it is relatively inexpensive, light, ergonomic, and it allows for batch uploading when wireless connections are not available or distances are beyond its range. However, making its data output compatible with SIRSI may be problematic. PDAs Personal Digital Assistants (PDAs) are another option that has emerged recently. Prices, durability, and functionality range widely. Some users have called palm-sized units with integrated scanners and keyboards PDAs, but those will be separated out as portable data collection units in this report. For this report, PDAs refer to simple units without keyboards, such as Dell Axim X30 units, Symbol SPT-1550 Palm Pilots, and HP iPAQs. Most PDAs work offline and upload the data in batches through a synchronization cradle, infrared connections, or other cables. There are also Bluetooth- and WiFi-equipped PDAs available, but they have somewhat limited range, especially Bluetooth-enabled PDAs. PDAs with attached barcode scanners have been used by several libraries recently to inventory collections (Chen & Ma 2004; King 2003; Sterback 2006; Urquiaga 2002). PDAs with integrated or attached barcode scanners have several good points. Cost: inexpensive and can be purchased from many vendors; prices range from $300-$1,000; however accessories add considerable cost. Free software: software is available for free (Observer Control software from Indiana State University) or at very low cost (On Hand software for $80). Convenience: very portable and easy to use; at Brigham Young University's law library, workers preferred PDAs to laptops with scanners (Urquiaga 2002). Ergonomic: lightweight, fits easily into hand, bright color displays (although many have battery-saving modes that dim screens after a few moments of inactivity). Standard WiFi: wireless versions use standard WiFi 802.11 b/g wireless routers or Bluetooth (2.4GHz RF). Flexibility: can be used for other applications such as calendar management, taking notes, doing calculations, etc. depending on the software loaded, although some users found the screens too small to use for "roving catalog lookup." PDAs with attached barcode scanners have several disadvantages though: Durability: commercial PDAs are not very durable and the stylus is easy to lose or break; one can buy protective cases but even these are not meant to withstand repeated drops. Limited input options: must use stylus unless purchase keyboard separately or have integrated barcode scanner Easily scratched screen: screen protectors are essential if a stylus is used to input data or to do other functions. Obsolescence: because PDAs are "cool" high-tech items, they evolve rapidly and so does the supporting software; many of the software upgrades, such as new versions of ActiveSync to upload data, are available freely, but the operating systems (O/S) are usually not freely upgradeable; whether the inventory software will be freely or cheaply upgradeable is not known at this time. Lack of standards: some PDAs use Palm O/S, others use Pocket PC software; these are not compatible with each other so if multiple units were purchased, they would all have to use the same O/S and could share other software; also, because operating systems change rapidly, the devices should be purchased at the same time to ensure interoperability. Use time: battery life with standard battery may not be long enough, especially if the user needs to keep display well lit to be readable; extra long life batteries would be needed and a separate fast charger may be desired. Limited memory: onboard memory may be insufficient without an added memory card such as a Compact Flash (CF) or Secure Digital (SD) card. Short read distances: attached barcode scanners tend to be simple wands or pens or integrated scanners that require a close distance to barcode although the Socket SDIO scanner suggests it could read up to 24" away depending on barcode type and size. Compatibility: free or cheap software tends to translate barcodes into formats suitable for spreadsheets or database management systems; would require programming to make it compatible with SIRSI or would need to buy expensive SIRSI PocketCirc software (1-4 units $600 per unit plus $108 maintenance; unlimited units $15,000 plus $2,700 maintenance) Examples of PDAs with attached barcode scanners include: Dell Axim X30 ($350) + plug in Socket SDIO scanner ($220); SIRSI sells kit for $650; weighs about 5oz; optional WiFi or Bluetooth although either will take the remaining CF expansion slot; compatible with SIRSI PocketCirc software. Symbol SPT-1550 ($400-$500); can hold up to 1,000 barcodes; read distance ~ 1"-24"; weighs about 6oz with battery; used by Purdue, Southern Polytechnic State University , and Rutgers (Sterback 2006). Palm III Laser Scanner ($1,095) has integrated Symbol SE 900 barcode scanner; works with Alexandria ILS HP iPAQ model 5555 ($200 although this has been superseded by newer models) with Symbol SPS 3000 barcode scanner ($400 discontinued); could substitute a Socket CompactFlash Card with Symbol Scanner ($310) wireless; compatible with SIRSI PocketCirc software; used by Kansas City6 Summary PDAs are convenient, flexible, cheap and easy to use. However, they are not generally durable and are quickly superseded by newer models. While we would not need the kind of durability against wind, dust, moisture and other such stressors that industrial users require, units should be able to withstand being dropped several feet a few times. Other SIRSI libraries such as Rutgers have used the more rugged Symbol SPT-1550 successfully, so there are resource people to assist in transforming the data to SIRSI-compatible information. The SPT-1550 allows only batch uploading. There is a newer model, the Symbol SPT-1800 (~$700) that allows wireless communications as well as batch loading. If PDAs were used, this model might provide the best of both worlds. Portable Data Collection Units Portable data collection units work in a manner similar to cordless hand-held barcode scanners, but they have keypads, large displays, and enough internal memory to not only store barcodes but also to run simple software and send information to a portable printer. Some use a stylus, like a PDA, or a touch screen, but most rely mainly on the keypads for data input. Many are approximately the same size as PDAs but they are more rugged than PDAs. However, they are also considerably more expensive. Although these can be used without attached or integrated barcode scanners, this would require keying in or using the stylus to input barcodes. Either method would be inefficient so integrated or attached barcode scanners were preferred. Offline portable data collection units have many of the same advantages and disadvantages as cordless offline hand-held barcode scanners. Online portable data collection units work in a manner similar to cordless wireless hand-held barcode scanners but have additional memory, software, and functionality. Some can almost act as self-contained circulation units and many are compatible with SIRSI. In offline environments, a static copy of Workflows could be loaded, and the unit programmed to print out a slip on a portable printer whenever an exception is noted, allowing the worker to pull the book and insert the error slip for processing. In wireless environments, the unit could update incorrect circulation records automatically, or print out a slip for any error that is too large to correct automatically. Portable data collection units have a few advantages: Durability: even the smaller ones are designed to be more durable than commercial PDAs; more durability though usually increases weight. Ergonomics: often have optional pistol grips for better ergonomics; some also have optional shoulder holsters or other carrying accessories. Distance reading: most have integrated laser scanners that can read at large range of distances from a few inches to several feet, allowing for remote scanning. Use time: usually have long battery life and can often purchase quick chargers, extra batteries, and other accessories to prolong use time. Large memory: internal memory allows some programs to run, such as a program to compare the barcodes to a static master shelf list for offline units. Printing capability: can connect to portable printers to print out a slip any time an error is noted. Flexibility: if there are sufficient keypads, online units can be used to correct records on the spot; some units can act as standalone circulation units. Upgrades: many vendors mention that upgrading units is possible as needs evolve, although the costs involved have not been investigated. Portable data collection units have many disadvantages though: Batch uploading = batch problem solving: batch scanners do not allow real time updating of data; even if slips are printed, records are usually reconciled later. Limited input: many have limited keypads that require considerable manipulation to input data. Compatibility: most portable data collection units are used in retail or industrial applications so they are not directly compatible with SIRSI; some work with SIRSI PocketCirc, but this is extremely expensive software. Obsolescence: because these devices have operating systems (often Pocket PC), these might require purchasing new O/S versions to postpone obsolescence. Weight: although larger displays and keypads make the units more usable, they also increase the weight; may require shoulder holsters or other means to alleviate the burden of carrying the units. Cost: portable data collection units tend to be quite expensive, especially if they are wireless. Examples of offline portable data collection units include: ERS Oyster RT90 with CCD scanner ($1,000 + accessories); no longer manufactured; used at Imperial College (Allatt et al. 1997). Symbol LDT 3805 ($1,816.75 + cradle $211.25 + accessories = ~$2,109.25); no longer manufactured; used at Lamar University (Allen 1998). Intermec 2410 hand-held computer ($1,600 + accessories); used by Round Rock Independent School District (Anonymous 2006) and Rice University, according to Elizabeth Baber (personal communication, November 28, 2005); weighs about 14 oz; battery lasts about 6-10 hours; recharge time 3-4 hours; full keypad; can be upgraded to wireless (Intermec 2415). Intermec CK1 hand-held computer ($850 + accessories); batch data collection terminal; integrated laser scanner; simple keypad; optional pistol grip for better ergonomics Examples of online portable data collection units include: Intermec CN2 Mobile Computers ($1,295-$1,595); PDA-sized but more rugged with limited keypads; WiFi or Bluetooth or batch loading. Intermec 700 Series Mobile Computers ($1,795 + accessories); slightly larger than usual PDA, more rugged; integrated barcode scanner; data uploads via Ethernet (wired) or WiFi; some models Bluetooth capable; limited keypad, touch screen; SIRSI has tested PocketCirc with Intermec 730. PSC Falcon 4220 ($1,539 + accessories); SIRSI sells kit for $2,886; weighs about 13oz with battery; rugged; limited keypad; touch screen; optional WiFi; can be used for batch uploading also; used by Rutgers (Sterback 2006); recommended by SIRSI for PocketCirc software. Summary Portable data units are rugged but expensive. The PSC Falcon 4220 allows both batch uploading and WiFi, enabling its use in both wireless and non-wireless environments. It is also recommended by SIRSI if we choose to use PocketCirc software. However, SIRSI's price seems high, even with all accessories included. Portable data units might have obsolescence issues if the operating systems and software need to be upgraded regularly. Laptops with Attached Barcode Scanners Laptops, although not strictly hand-held, are portable solutions that have many variations and options. Laptops may be work in real time or offline for batch loading into the production server later. Although some libraries do plug their laptops into wall circuits, dragging extension cords throughout the stacks is neither safe nor desirable. Therefore, battery operation is preferable. Laptops that work in real time may be wireless or may be connected to a LAN. However, the latter would require accessible ports and long cables that may be tripped on, pulled, or otherwise be a nuisance. Laptops that work offline may upload their data through a special docking device or simple USB, infrared or IEEE 1394 connection. Although keying in barcodes is possible on laptops, that method would be inefficient so attached barcode scanners would be desirable. The barcode scanners that attach to laptops may be wireless or wired. Wired ones tend to work off the laptop power source, while wireless units usually need battery packs. Laptops have several major advantages: Applicability: offline laptops can work in most environments, including areas that are not wireless enabled or where the signal is not strong; wireless-capable laptops can also work offline when necessary. Low training needs: most users are familiar with laptops so training is minimal if they are just scanning and pulling any exceptions. Flexibility: laptops can be used for many other applications when not being used for inventory. Ergonomic: laptops have large screens that are easily readable and keyboards that almost anyone can easily use. Real time = problem solving on the fly: wireless updating allows real-time updating and noting of problems. Compatibility: laptops connect or upload directly to WorkFlows so no extra software nor programming required; used by many libraries that already use SIRSI (Sterback 2006; Northside Independent School District n.d.) or Dynix (Kent School District Library Media Services 2003). Cost: could use laptops already available with barcode scanners from circulation stations such that no new hardware needs to be purchased; alternately, could use laptops slated for salvage because inventory would not require the greatest processing speed or other options. Disadvantages of laptops: Errors must be noted: if using real-time updating, the staff must carefully note the screen response for every item scanned or else errors will not be noted; also, the advantage of being able to note errors as they occur is that such errors could be fixed immediately. However, this would mean the scanning staff must be trained and authorized to edit WorkFlows records. Wireless limitations: not all buildings and areas are compatible with wireless. Cumbersome: laptops are not really hand-held and must probably be wheeled around on carts; this could be problematic in stacks with less room. Use time: laptops are extensive power users and if battery operated, this would require several battery packs that can be swapped out to allow for full day's use; fast chargers may be needed; battery packs may be heavy. Obsolescence: like any device with an operating system and software, these must be consistently upgraded to avoid obsolescence; fortunately, there are standards for wireless compatibility, laptop O/S, and other software that laptops require. Cost: Laptops cannot be borrowed or taken from the lifecycle stream, would need to purchase new laptops and these are expensive. Summary Laptops are readily available, easy to use and flexible. They could simply be borrowed from other departments, or older ones could be used, without the need for new equipment. They can work offline, with a static version of Workflows or a shelflist that could be used to note any errors to be pulled. In wireless environments, records could be updated automatically and the workers could either pull egregious errors or correct them on the spot. However their major advantage would be the ability to correct errors as they are discovered and this might require more extensive training of the scanning staff, and would slow the inventorying process. Funding Although several libraries mentioned "special funding," we could not determine where such funding was acquired. Many libraries seem to simply re-allocate stacks maintenance funds for the inventory process. There are several angles that could be explored for funding: preservation, resource sharing, and Penn State's relationship with SirsiDynix. Preservation: if the inventory staff were trained to recognize and pull any items that are in need of repair, funding might be obtained by arguing that this is a preservation issue. A grant asking for preservation-oriented funds might allow the inventory process at least once. Resource Sharing: Penn State acts as a resource library for the state of Pennsylvania, and is part of many cooperative collection development initiatives. It could be argued that these arrangements would be better served if the catalog and collections had more cohesion. That way other libraries would not request items from PSU that we no longer own, or they can see that we possess items that are not currently visible in the CAT. Resource sharing would be made more efficient if an inventory were done. Therefore, some grant from PALCI (Pennsylvania Academic Library Consortium, Inc), the Committee on Institutional Cooperation (CIC), or other such consortia might be appropriate. SirsiDynix: It could be argued that many of the discrepancies between the CAT, Workflows, and the actual collections, resulted during the migration to Unicorn. Therefore, it would be in the best interests of SirsiDynix to assist PSUL in eliminating those discrepancies, since it would promote greater appreciation of the system. While SirsiDynix would probably not give the PocketCirc software gratis, a reduced price might be negotiated. Possible Test Collections The task force discussed possible collections to serve as test beds for any device before allowing widespread use. Such collections would need to be discrete (that is, to have a separate location that could be distinguished), small, and should have some possibility of having exceptions. Three collections were identified as test beds, the Arts and Humanities Library ready reference collection, the Kneebone Mushroom Collection, and the off-site storage Annexes. All are discrete collections and relatively small. However, both Kneebone Collection and the Annexes are limited-access collections, such that missing items, unbarcoded items, and other such exceptions are unlikely. Also, the Annex collections are large and many items remain unbarcoded. Thus, the Arts and Humanities Library ready reference collection will be the first test bed used. Conclusions It was determined that it would be highly feasible for PSUL to purchase a set of devices that can be loaned to units conducting physical inventories. There are many choices, depending on the funding obtained, the manner in which inventorying would fit into normal workflows, and other such factors. Cordless hand-held barcode scanners, while cheap, durable, easy to use, and not likely to become obsolete quickly, are limited in range for wireless applications and would require programming to make them compatible with WorkFlows. They have very limited functionality, although they could be put to use at circulation stations when not used for inventorying. This seems to be the least desirable option. PDAs are cheap, light, ergonomic, and flexible, but have limited scanner-read ranges and have issues with obsolescence and durability. The Symbol SPT-1550 is fairly rugged and has been used by SIRSI libraries so that compatibility, while not automatic, is assured. If the SPT-1800 is similarly compatible with SIRSI, the added wireless capacity might make this a desirable option. Portable data collection units offer the best balance of functionality and durability, but the price could be a barrier. Such units offer ruggedness, flexibility, long battery life, and some are compatible with SIRSI. The PSC Falcon 4220 can be used, offline or wirelessly, directly with WorkFlows, or with SIRSI PocketCirc software, if desired. Assuming that PocketCirc is updated as WorkFlows is updated, this might provide the most integrated, compatible environment. However, it is an expensive option. Laptops are a viable option because they are readily available, flexible and easy to use. Laptops that have been cycled out of use in subject libraries could be put to use for inventory, with just a barcode scanner attached. There would be no need for added programming or software. This would be the simplest solution, requiring the least amount of equipment, software and training. The best options for hand-held inventory devices depend on how inventorying will fit into the workflow of other processes. If inventorying were to be completely separate from other projects and processes, almost any of the options would work. Choices would then depend on issues of time, price, ease of use, compatibility, etc. If inventory were to be coordinated with shelf reading, speed of scanning, ease of use and ergonomics would prevail. It does not seem feasible to train the entire shelving staff to be able to edit WorkFlows records, and editing on the fly would reduce shelf-reading speed. Thus use of laptops to do real-time updating is not optimal if inventorying were coordinated with shelfreading. However, it might be useful to have some alerts of exceptions generated in real time or against a static shelflist so that they could be pulled as they are discovered. Therefore, a wireless unit, that also allows batch data collection for non-wireless areas, with sufficient internal memory to allow a static shelflist to be loaded, would be suitable. Ease of use and ergonomics make the PSC Falcon 4220 or the Symbol SPT-1800 suitable options. If inventory processes were to be integrated into cataloging processes to prevent a backlog of items whose records need to be added, deleted or otherwise edited, on the fly record editing would be best. This would indicate wireless laptops as the optimal option. However, the lack of wireless capacity in many library buildings make this workflow less feasible. Therefore, the recommendation for purchase, assuming sufficient funds, would be for some portable data collection device, such as the Falcon 4220, with potential to purchase the PocketCirc software. Funding would have to account for the units, the software, accessories, including carrying devices or pistol grips, extra batteries and other such options, and upgrades to the operating system and other software, as needed. Note: Funding for purchase of five Falcon 4220 devices, with PocketCirc software, was approved in Fall 2006. However, SIRSI was unable to supply the devices and the purchase order went to an alternate vendor recommended by SIRSI. Turnover at that company caused miscommunications and delays, such that when the order finally went through, the Falcon 4220 was no longer available. Its replacement, the Datalogic Pegaso Industrial PDA with Windows CE was purchased instead. They are in the process of being configured and tested as of November 2007. Bibliography Allatt, P., Fisher, M. & Yeadon, J. 1997. Automated stocktaking at Imperial College. Program 31(2):161-169. Allen, V. 1998. Automated library inventory using bar codes. In: Data or information: the fading boundaries: Proceedings of the 23rd Annual Conference of the International Association of Aquatic and Marine Science Libraries and Information Centers. (ed. by J.W. Markham & A.L. Duda), pp. 101-105. Fort Pierce, Fla.: IAMSLIC. Anonymous. 2006. Round Rock Independent School District - Inventory. [Online] Available: {http://www.roundrockisd.org/home/index.asp?page=509#invoverview} [March 1, 2006]. Chen, L. & Ma, Y. 2004. Library inventory using Palm Pilot. Technical Services Quarterly 22(2):15-23. Deuink, A., Notartomas, T., & Oswald, D. 2005. WorkFlows Inventory Guide. [Online]. Available: http://www.libraries.psu.edu:8001/tas/access/council/InventoryGuide.pdf. [This is an internal library document that may be requested from the author]. Emery, C. D. 1990. The use of portable barcode scanners in collections inventory. Collection Management 13(4):1-17. Kent School District Library Media Services. 2003. Inventory Instructions Using Dynix. [Online]. Available: {http://www.kent.k12.wa.us/library/files/inventory_instructions10_03.pdf} [March 1, 2006]. King, D. 2003. Going Wireless in Kansas City: Testing Handheld Technology for Public Library Use. [Online] Available: http://www.infotoday.com/cil2003/presentations/king.ppt [Febrary 24, 2006]. Northside Independent School District. n.d. Wireless Library Inventory "How to" 2006. [Online]. Available: {http://www.nisd.net/sirsi/7_SIRSI_Inventory.doc} [February 24, 2006]. Sterback, C. 2006. So, you want to inventory your collection? SirsiDynix SuperConference 2006. [Online]. Presentation available to conference attendees only: http://www.superconference.info/Resources_2006/So_You_Want_To_Inventory_Chris_Sterback.pdf [received as Powerpoint e-mail attachment from Chris Sterback March 21, 2006]. Urquiaga, L. 2002. Library Inventory Project Description, Brigham Young University Law Library. [Online].Available: http://www.law2.byu.edu/Law_Library/Lnis/projects/projects_book_inventory.html [February 24, 2006]. Wild, J. W. 2002. Inventory at Brooklyn College, 1998-1999: an original method. Library Resources and Technical Services 46(2): 62-71. Previous   Contents   Next