The views expressed by Ms. Schroeder come as no surprise to us folks in the Library field.
Here's how University Libraries (which are the really the one's being targeted in the article, not Public Libraries) usually deal with electronic material.
Purchasing: We buy access to electronic journals for a yearly fee (usually 10% above what we would pay for the print). This grants us access to the current issues as well as whatever backfiles the publisher makes available. Some publishers have a scale of fees to determine how much of the backfile we get access to. If, for some reason, we have to cancel our online subscription, we lose all access, including for years that we paid for (there are a few publishers, usually scholarly societies that promise perpetual access for years paid for, but they are the exception.)
This access is controlled by an ip based authenication scheme on the publisher side. Anybody can come into the library and get access and members of the University community can get acces off-campus through our proxy servers. In general there are no limits on number of simultaneous users.
So far the model for electronic books is similar. There are several e-book services available: Books 24/7, ITKnowledge, NetLibrary, etc. Books 24/7 and ITKnowledge work like electronic journals - we buy access and our ptarons can use it whenever and for however long they need to. NetLibrary requires use of a non-browser based reader and requires books be 'checked out' for periods ranging from 2-6 hours or so. The length of time a book can be checked out is negotiated for in the contract and the library pays more for a longer check out period.
Document Delivery/Inter-Library Loan: Whether or not we can provide copies of an article to another library depends on the license terms of our access contract. In general it is allowed. However we are not allowed to provide an electronic copy. We must print off the article then fax or mail it to the requesting institution.
For e-books, ILL is pretty much straight out, except for providing a few pages or a chapter.
In either case the requesting library has to pay two fees: they pay us for our labor/costs and they pay the publisher the copyright fee for the copy. Typical document delivery costs are 25-50 $US per article. Some publisher forbid document delivery of copies even from print journals and other s charge copyright fees in excess of 100 $US.
Indexing: Just having electronic access to journals is useless unless our patrons have the ability to search for articles. This entails the library purchasing access to electronic databases that index a particular discipline for an additional yearly fee.
So if we consider the case a mid-sized American public research university and look at what we pay yearly for the Computer Science field:
Now publishing in the computer science field is dominated by two scholary associations: the IEEE and the ACM, which explains why these costs are relatively low.
Biology, by contrast, tends to be dominated by the large commerical publishers like Elsevier, Gordon and Breech, Springer-Verlag, Kluwer Academic etc. The yearly costs in biology would probably be approximately 3-5 times those in computer science.
Quite frankly, the commercial publisher do have a problem in so far as most libraries have inflation limited budgets while commerical publishers tend to increase the costs of their journals by about 10-25% per year. This leads to libraries having to engage in 'cancellation projects' every 3 years or so in order to stay on budget. This lowers the publisher's profits and forces them to raise prices while simultaneously forcing more libraries to rely on document delivery for access to the research literature being produced by the faculty at their own University.
Librarians are most assuredly not taking this sitting down. We have our own political attack dog, the American Library Association (whose views I can't say I always agree with, but I am a member and do support their lobbying efforts.) Also library's are funding projects like SPARC: Scholarly Publications and Academic Resources.
Hopefully this has been more or less on topic.
Stephen W. Fairfield
Engineering/Mathematics/Statistics Librarian
DjVu was actually developed by AT&T Labs - Research. http://djvu.research.att.com/ well over a year ago. They've released it to various other companies to comercialize it. Since the folks at AT&T Labs are (not surprisingly) big Unix users there is support for the various Unix flavors, including Linux. You can get more technical information than you want by going to their website. (one note: Also not surprisingly, most of their technical documents about DjVu are in DjVu format, so you'll need to get the plugin, either from AT&T or LizardTech.)
Disclaimer: I used to work at AT&T Labs as an outside contractor in their library.
Slashdot Mirror
Here's how University Libraries (which are the really the one's being targeted in the article, not Public Libraries) usually deal with electronic material.
Purchasing: We buy access to electronic journals for a yearly fee (usually 10% above what we would pay for the print). This grants us access to the current issues as well as whatever backfiles the publisher makes available. Some publishers have a scale of fees to determine how much of the backfile we get access to. If, for some reason, we have to cancel our online subscription, we lose all access, including for years that we paid for (there are a few publishers, usually scholarly societies that promise perpetual access for years paid for, but they are the exception.)
This access is controlled by an ip based authenication scheme on the publisher side. Anybody can come into the library and get access and members of the University community can get acces off-campus through our proxy servers. In general there are no limits on number of simultaneous users.
So far the model for electronic books is similar. There are several e-book services available: Books 24/7, ITKnowledge, NetLibrary, etc. Books 24/7 and ITKnowledge work like electronic journals - we buy access and our ptarons can use it whenever and for however long they need to. NetLibrary requires use of a non-browser based reader and requires books be 'checked out' for periods ranging from 2-6 hours or so. The length of time a book can be checked out is negotiated for in the contract and the library pays more for a longer check out period.
Document Delivery/Inter-Library Loan: Whether or not we can provide copies of an article to another library depends on the license terms of our access contract. In general it is allowed. However we are not allowed to provide an electronic copy. We must print off the article then fax or mail it to the requesting institution.
For e-books, ILL is pretty much straight out, except for providing a few pages or a chapter.
In either case the requesting library has to pay two fees: they pay us for our labor/costs and they pay the publisher the copyright fee for the copy. Typical document delivery costs are 25-50 $US per article. Some publisher forbid document delivery of copies even from print journals and other s charge copyright fees in excess of 100 $US.
Indexing: Just having electronic access to journals is useless unless our patrons have the ability to search for articles. This entails the library purchasing access to electronic databases that index a particular discipline for an additional yearly fee.
So if we consider the case a mid-sized American public research university and look at what we pay yearly for the Computer Science field:
Print books: ~5000 $US
Electronic books: ~6000 $US
Print Journals: ~36000 $US
Electronic access to journals: ~20000 US$
Electronic indexes: ~50000 $US
Total: ~117000 $US
Now publishing in the computer science field is dominated by two scholary associations: the IEEE and the ACM, which explains why these costs are relatively low.
Biology, by contrast, tends to be dominated by the large commerical publishers like Elsevier, Gordon and Breech, Springer-Verlag, Kluwer Academic etc. The yearly costs in biology would probably be approximately 3-5 times those in computer science.
Quite frankly, the commercial publisher do have a problem in so far as most libraries have inflation limited budgets while commerical publishers tend to increase the costs of their journals by about 10-25% per year. This leads to libraries having to engage in 'cancellation projects' every 3 years or so in order to stay on budget. This lowers the publisher's profits and forces them to raise prices while simultaneously forcing more libraries to rely on document delivery for access to the research literature being produced by the faculty at their own University.
Librarians are most assuredly not taking this sitting down. We have our own political attack dog, the American Library Association (whose views I can't say I always agree with, but I am a member and do support their lobbying efforts.) Also library's are funding projects like SPARC: Scholarly Publications and Academic Resources.
Hopefully this has been more or less on topic.
Stephen W. Fairfield
Engineering/Mathematics/Statistics Librarian
Disclaimer: I used to work at AT&T Labs as an outside contractor in their library.