Journal Print-E mix: one, the other, both, or neither?

Sometimes so many things come up at the same time it becomes difficult if impossible to ignore. Here's just a brief list:

• An oceanographer came to me and asked to see a print copy of an AGU journal article. If you've followed me here from elsewhere, then you'll know my place of work was mandated to discard all print materials (we did actually make the case for maybe 4 journals that are both not available online and are not widely held - there was a 5th but it got discarded by accident). Turns out that the entire point of the article was to show two color graphics on the second page. Well this journal is widely held, but we weren't sure who could actually do a color scan. You see, AGU scanned these articles from that time period in black and white.  I'm sure if I actually asked AGU, they could probably have provided, but my awesome colleagues on my listserv came through.
• Journal of the American Chemical Society became just the most recent in a string of publishers to prefer e over print, make e the copy of record, or end print publishing. (yes, chemists, JACS is not the first!). This caused Nature Chemistry to run an editorial which was then picked up on one of the Nature blogs, which was then commented on in friendfeed.
• Recently Ben Wagner suggested on a list that someone in a region should agree to keep the last print copy - this generated some discussion.

You can see from my comments in friendfeed, that I'm not only not impressed and not surprised, but I think most scientists in most cases won't notice. In my library, we did have a steady stream of scientists and engineers who would browse journals. But then they would go back to their desk, pull up the electronic copy and print and save that!  Actually, I would say that the Phys Rev, the AIAA, and the IEEE journals really haven't been browseable for years.  Too big, too general in some cases.... I've also noticed people have fewer personal copies of journals, too, and this is borne out by studies (search for Tenopir or Tenopir and King).

AAS (publisher of AJ and ApJ) is really smart in this. They were the first ones I heard talking about a couple different versions of print on demand. They mentioned that you might just order the print as a pre-bound set at the end of the year for archival purposes!!  Isn't that sweet? Seriously - no tracking down or claiming copies, paying to send to the bindery... We get EOS that way, and it's pretty cool.

With the proliferation of journals and the increase in the number of articles in each, I'd say that it's not all that realistic to keep up by looking at them. I recently got a copy of the member journal for the Society for the Social Studies of Science - I had no idea what to do with it! JASIST I only get in e because I'm a student member.  I browsed it- but I get the RSS feed from Sage so, eh. I guess I'll throw it out.

So what's the controversy - and there are legitimate concerns, it's not just nostalgia or fear of change (as I was accused of at work).  There are problems that just apply to scans and then other issues that also go with things born digital.

First, sometimes there are problems with the scans including, but not limited to:

• black and white when the original is color
• missing articles or pages out of articles or even issues
• fuzzy or illegible scans
• omitted front matter, ads, editorials, etc.

Some of this is planning and reflects the evolution of the whole process. Older scans were just of the research articles. Most publishers have doubled back and now go from cover to cover. I always find it amazing that publishers don't have complete runs of their publications. They seriously don't. Some of the missing issues are because they couldn't get a copy to scan. Some are just quality control issues.  Black and white scans were clearly a management decision - someone did a cost-benefit analysis and decided this was the way to go. On the other hand, one of the aggregators does everything in really crappy color so even articles originally in black and white aren't very readable.(if you have an interest in digitization, go visit Jill!)

Second, you have the whole issue of preservation. Digital preservation is tricky - pdf/a? migrate? keep old computers? Microfilm lasts a long time and archival paper lasts even longer. There's LOCKSS, CLOCKSS, PORTICO - there are dark archives, light archives, geographically distributed mirrors, off site backups. Yeah, it's really lousy when you don't have internet access or if your authentication system breaks down, but most journals have some sort of plan, even if it might not be as well developed as it could be.  A similar issue is what you do in time of a crisis, when you need scientific information to save lives and your connectivity is down. We don't have any print indexes anymore, but if you knew a citation, you could go pull an article.  Some disaster plans and continuity of operations plans store copies of handbooks at the backup site, but you can't do that with the journals.  There are a couple of government organizations that pay to locally load journals and indexes for various reasons you can probably guess - so they might be ok and they might be able to provide services to responders.  Well, if they haven't fired their librarians.

Third, there's serendipity. I, for one, do better with new articles trickling in via RSS as they go up on early view.  I also like special issues that gather a bunch of related articles together.  At my place of work, the Kalman filtering article was nearly shredded. There was a note on it not to photocopy from the bound issue - there was a yellowed and much-photocopied copied version in a folder next to it you were supposed to use. Copies that are routed or passed around a lab probably indicate which articles are read... in a library, though?  And with such quantity?  I totally believe in serendipity in the stacks - but then books are co-located by subject. Maybe if it were a very specialized journal you might get this effect, but otherwise?

Last do we need "issues" when what we already have is a stream? I've been talking about disaggregating journals for a few years now.  I think special issues have value. I think virtual journals have value. I think there's a need for someone to manage editing, copy editing, peer review, and provision - I think journals should continue. But maybe a journal is a database?

Libraries can save a ton of money by going e-only, but they are very, very concerned about preservation and access. What I've been hearing is that they go e-only when

• the e is a good substitute for the print (or is better than)
• the publisher has a good policy for digital preservation
• they can buy perpetual access
• it's the copy of record
• it's that or nothing because there are massive budget cuts or, like us, they were mandated to go e-only

So this is all for science and engineering, actually. Humanities are a different thing and something I know little about.