[Prev Page] [Next Page] [Contents]
Agreements should be as simple and clear as possible, and preferably capable of being understood by the lay-person; clauses should be headed so that their content is obvious; and the clauses of most concern to subscribers (i.e. those containing the issues covered below) should precede, or be distinguished from, the routine ones that fill most agreements and have little bearing on the key issues for suppliers and subscribers, namely those of pricing and control.
Existing agreements are often very long, very detailed, frequently have no clause headings, and the key issues, especially those dealing with networking, are often buried or not even present. It is all too easy to miss, or misunderstand, important clauses, unless the agreement is examined by lawyers. This may be practicable at present, but will represent a nightmare in the electronic future, particularly if the second recommendation is ignored.
Libraries, publishers and/or information providers should use existing or new (subscription) agents or form consortia for the distribution of their products, rather than supply directly.
For administrative purposes, librarians (and users) would like to minimise the number of agreements that have to be signed, particularly in the case of electronic books and journals, which, potentially, far outstrip the number of existing bibliographic electronic products. It would also be administratively easier to deal with several agents or consortia, rather than a multitude of different publishers/information providers, on a day-by-day basis, especially concerning technical matters.
Suppliers should offer libraries a range of different networking licences to suit their individual circumstances. These options should not be price-biased in such a way as to force libraries towards one or another option. Suggested options are: consortium licence; organisational licence; site licence (with site carefully defined); concurrent users licence; specified workstations licence. Licences for potential users or potential workstations will not be appropriate in most cases.
Colleagues appear to favour the concurrent users option for CD-ROM services, at present, probably because this results in a predictable annual subscription and because individual CD-ROM database use is relatively low. However, the organisational or site(s) licence option appears to be favoured for tape-based services. This is somewhat paradoxical as CD-ROM databases may be loaded onto hard-disk and networked too. Almost certainly, it is the predictable-budget linked to expected level of use that has been the determining factor in this choice. Therefore, if suppliers offer reasonably-priced organisational or site licences for CD-ROMs, it is likely that subscribers will choose this option if usage is sufficiently high.
The charging mechanism for local area networking should not be dependent upon the medium on which the database is supplied, or networked locally, but rather upon the way in which the database is used. The corollary of this is that there should be no prohibition by supplier on the loading of CD-ROMs onto hard disk for local area networking.
It seems inequitable that a database delivered on CD-ROM and then networked from hard disk, for example, should be charged at a different rate to the same database delivered on CD-ROM and then networked from a CD-ROM optical server, all other factors being equal. At least one supplier - Silverplatter - has adopted medium-independent charging for some (most?) of the databases it supplies. Charging for these is banded by number of concurrent users, whether the database is networked from CD-ROM or from hard disk.
This option is not favoured by librarians and they will resist its introduction. However, if suppliers insist on introducing it, whether or not it is coupled with encryption, it is essential that network operating systems are capable of disconnecting inactive sessions after a specified time.
This option is not favoured because it destroys the main advantage of CD-ROM and other subscription-based services over online pay-as-you- go (PAYG) services, namely predictability of costs, and thereby opens up the spectre of the "bottomless pit" in the mind of library managers.
Disconnection of sessions is important if the metering software monitors connect time as well as portions of the database printed or downloaded.
This option should not attract any additional charges; instead remote access by legitimate users should be included in the existing licence, whether the latter is for concurrent users, site(s) licence, or whatever.
There seems to be no justifiable reason to penalise an institution because its users are working remotely. After all, this is the way in which education and working practices are developing in the electronic age. Suppliers can be protected from illegitimate third-party use by PIN-controlled log-in.
The practice of leasing software and information supplied to a subscriber on some physical medium, e.g. CD-ROM, should be discontinued. If a subscription is cancelled, the current version of the software and the information backfiles should remain the property of the subscriber.
If the subscription to a printed journal or abstracting service is cancelled, subscribers are not expected to return the backrun. Even in the case of leasehold houses and flats, the law has been changed so that the property does not automatically revert to the leaseholder when the lease expires! The subscription to an electronic product should be considered to be payment for updates of the information and upgrades of the software only, and not payment to retain the product during the lifetime of the subscription only.
If the product continues to be used, once the subscription is cancelled, no fees should be charged for networking, provided that it is not networked to more than one concurrent user.
If the product is not networked to more than one person at a time, then the situation is analogous to the consultation of a single volume by one person, albeit that that person is saved a trip to the library! If the product is networked to more than one concurrent user, then the supplier can argue that the subscriber would probably have had to have bought more than one subscription to the printed product to have achieved the same result. It could, of course, be argued that even in the latter case, the subscriber does not have to continue to pay fees once the subscriptions have lapsed, but it is probably better to recognise the degree of value added by the networking capability to multiple users.
Royalties for downloading or printing, if in force at the time the subscription is terminated, may continue to be made.
It seems reasonable for subscribers to expect to have to continue to pay downloading or printing royalties, if they already exist, even if the subscription is cancelled, as in this way the supplier continues to be compensated for the copying of copyrighted material.
Suppliers should replace lost, stolen or damaged media for the current price of the medium, plus the cost of administration, postage and packing, if a copy is in stock. If a copy is not in stock, then the full costs of re-mastering may have to be borne by the subscriber.
The subscriber has already paid for the information once, and a lost or stolen disk is unlikely to deprive the supplier of revenue that s/he would normally have expected. The main cost to the supplier in producing the product is in acquiring the information and processing it to the master stage, rather than in copying it or supplying an existing copy.
It should be the subscriber's decision whether or not to insure against such risks and not a condition of the agreement, provided that the product is purchased not leased.
If the subscriber is purchasing the product, rather than leasing it, then it is unreasonable for the supplier to insist that the subscriber insures it. It must be at the subscribers own discretion (and risk) to decide whether or not to do so.
Users in subscribing organisations should be allowed to download bibliographic references into personal bibliographic information management packages, retain this information indefinitely, and subsequently edit it for inclusion in article bibliographies, provided that these are not resold.
This is already happening without permission and is virtually impossible to stop without auditing users' personal software for signs of infringement. It might be inhibited by encryption to prevent electrocopying without a fee, but it would still be possible for users to transcribe references from screens and re-key the information, as they have been doing for some time from printed sources. Therefore, it would be far better to legitimise it and increase income for the supplier.
Subscribers may expect to have to pay an additional fee to retain downloaded references indefinitely. To avoid monitoring of use, a pre-purchased limit might be negotiated with the subscribing organisation.
This is analogous to the downloading option in online databases, where an additional fee is paid if references are displayed in the downloading format. However, as librarians prefer predictable budgets and want to avoid the administrative inconvenience of monitoring use, they would probably prefer to be able to pre-pay for specified amounts of downloading.
Suppliers should be prepared to negotiate supplementary licences with educational organisations, so that the latter can make academic-discounted products available to third parties, at appropriate fees.
Many products stipulate no third-party use. However, many academic libraries have active connections with local industry and commerce, especially Science Parks. These users, who, previously, were able to use all the printed products in the library are now often barred from using their electronic equivalents. This represents income that is currently lost to suppliers, as such users cannot usually afford to pay the commercial rates for full subscriptions to the kind of products that the library can afford, but would be prepared to pay small amounts on a PAYG basis, which could be passed onto the suppliers.
Agreements should clearly state what discounts over the list price are available, or negotiable, and to whom.
Discounts may be seen by suppliers as another way of losing revenue. If, however, they increase the exposure of the product they are likely to result in increased market penetration in the long term.
There is a case for public libraries to receive some form of discounts, as they are not-for-profit organisations and do support the educational needs of an increasing number of students, outside their "home" institutions.
Commercial subscribers may wish to negotiate volume (of use) discounts.
Suppliers should maintain telephone help desk support for their products, during office hours at least.
Help desks are de rigueur in the online industry and have proved to be extremely useful in that context. It would be difficult for every individual information provider to fulfil this obligation, which lends weight to the argument for them to use agents or form consortia.
Suppliers should encourage the formation of user groups, perhaps by the offer of discounts, to provide feedback on their products, especially new releases. Certain sites, by agreement, should be designated reference sites where potential users can see the product in operation.
This can only be of benefit to the supplier, but may be onerous for the subscriber who could expect some form of recompense for assisting the supplier in this way.
Software upgrades should not be released until fully tested by the supplier and, preferably, by members of the user group as well. Pro-rata refunds should be provided, if software upgrades, or other changes, part-way through the subscription period, prevent the product from being used as previously, either in stand-alone mode or across a network.
In the past year, there has been at least one case of an "upgrade" that failed to work on networks as previously, because it required more memory, rather than less as was claimed. This resulted in much wasted effort on the part of subscribers, not least in communicating with other subscribers via the electronic mailing lists to share experiences about the problem. In Aston's case, at least, a pro-rata refund of the networking fees was negotiated.
Suppliers should state in annexes to their agreements (and in their literature) what network operating systems and workstation specifications the product is known to work with, and give details of reference sites operating the product with that equipment and software.
It is impossible for suppliers to guarantee that their product will work initially at a given site, even if it appears to meet the specifications given. In a networked environment, there are too many variables that may influence the end result. Nevertheless, potential subscribers would find it useful to visit sites with similar operating profiles that are using the product.
Subscribers should be required to destroy superseded discs rather than have to return them to the supplier.
It is administratively time-consuming, and therefore costly, for both subscribers and suppliers to have to track superseded discs and ensure that they are returned.
Subscribers should be permitted to retain some superseded discs for training purposes, on condition that this is the only purpose for which they will be used.
This can only benefit the supplier in the long term, as librarians are helping to increase the (effective) use of their products by training.
This allows suppliers to vary prices in case of inflation, and subscribers the opportunity to cancel a subscription in case of budget constraints or dissatisfaction with the product.
Suppliers should offer the option of the term of the agreement running either from the date of signing of the agreement, or for a calendar year expiring on the 31st December. In the latter case, the first year's subscription should be pro-rata depending upon when the agreement is signed.
Some libraries may always wish to enter into agreements at one particular time of the year, depending upon their budget cycle. Others may choose different times to enter agreements, yet wish to synchronise their review times.
The period of notice for termination of the agreement should be standardised at three months from the expiry of the term.
This seems to be a reasonable amount of time for either side to make other arrangements.
If your browser supports tables click here to see a Licence Agreement checklist.
[Prev Page] [Next Page] [Contents]
[UKOLN Home Page] [Papers and Reports] [British Library Papers and Reports ]