RE: Industrial use and library costs

[Jan Velterop <velterop@biomedcentral.com>]

David,

You ask for verifiable percentages of industrial subscriptions and
downloads. I don't have them.

First of all, the 5% I mentioned was my *estimate* based on what I
remember from when I still worked in the conventional publishing
environment. That's a few years ago now, and things may have changed, but
somehow I find it implausible that the percentage would have multiplied by
a factor of 5. I'll be as interested as you, if not more, in Crispin's
justification of the 25% he mentioned in the House of Commons inquiry.

You may also remember that I applied my estimate to *primary research
journals* (not review journals, databases and the like, let alone
advertising and other products or services). BioMed Central does not sell
*any* primary research material to anyone, so not to industry, either.
It's all open access. There is, therefore, no answer to your question of
what percentage of the BioMed Central paid-for material we sell to
industry, in the context of primary research material (that being the only
material affected by open access), simply because we no not have any
paid-for material that can be classified as primary research.

With regard to downloads, we have statistics on totals, on downloads per
article, and on downloads per IP address (this we use to spot excessive
downloads from single IPs, in case there are attempts to artificially
boost download figures for a given article or journal; that very rarely
occurs and we remove those numbers from our statistics if it does). Only
for institutions who give us their IP ranges, do we know how many articles
they have downloaded, but those numbers constitute private information
only for them. Unless users ask us to keep a count of their downloads,
privacy considerations prevent us from keeping any specific statistics
relating to identifyable users. So there, too, I cannot give you a
percentage of the total number of downloads that's performed by industrial
users. Even if I could, one has to realise that downloads from our site
are just the bare minimum picture. Because all articles are available from
a number of known repositories and quite possibly from many more unknown
ones, we do not have, nor ever expect to have, a full picture of the
number of downloads of our articles. Judging from aggregate download
totals (not per article) we are given by the known repositories, we
estimate that it must be at least twice as many as the count from our own
website.

Best,

Jan
BioMed Central
www.biomedcentral.com

> -----Original Message-----
> From: David Goodman [mailto:David.Goodman@liu.edu]
> Sent: 04 May 2004 14:06
> To: liblicense-l@lists.yale.edu; liblicense-l@lists.yale.edu
> Subject: Industrial use and library costs
>  
> 1. As we still have no verifiable data for the proportion of industrial
> use, some estimates of the factors involved will help define the data
> needed: There would be expected to be variation by subject to subject. 
> I'd expect a higher industrial use in chemistry than in mathematics.  
> I'd also expect lower industrial use for less important journals: 
> universities have been by themselves carrying the burden of minor 
> titles.
> 
> 2. Can we get real data, not approximate and unverifiable percentages:  
> Jan: having said 5%, please verify it by publishing the figures for
> downloads to the BMC open access journals, and for subscriptions to 
> your paid titles. Crispin, having said 25%, please release figures 
> justifying it.

[SNIP]

> Dr. David Goodman 
> Associate Professor, 
> Palmer School of Library and Information Science, LIU
> dgoodman@liu.edu