Comment Re:Can anyone point out (Score 1) 293
These things improve the reading experience, and I appreciate them, but I don't think they're truly necessary. The arXiv does just fine with an almost fully-automated system.
I think the arXiv project has been highly successful, and is a great example of what instituional repositories should look like. But like it or not, there is still a market and expectation among many academics for glossy-looking journals. They explicitly want the things that improve the reading experience - in particular, they want to be able to print good-looking paper (ie. PDF) versions of papers, and are willing to pay extra for that. Maybe things will develop where the journal as medium becomes extinct, but not ay time soon.
Right now, the arXiv's goal is not to replace journals, but to be as fast and as efficient as possible at hosting preprints. If they set out to replace journals outright, and added a slighly more sophisticated review system and automated validation of input, they could do what most journals do with about 90% of the quality for about 10% of the cost.
What you're describing is exactly what OJS does: sophisticated review/workflow system with a little automated validation. It provides about 90% of the quality of the journals you're describing with almost zero cost. It has been wildly successful, especially in the developing world, where starting and hosting a journal is expensive. If we're going to really break it down, how would this review/validated arXiv system you describe differ from a journal? Not much, it seems.
I don't know too much about XML journal article formats, and would appreciate more info. PubMed's explanation didn't really help much. Why bother with XML? Why not just require authors to submit in LaTeX? ... Why re-invent the wheel for publishing? My best guess so far is to accomodate authors who can't be bothered to learn LaTeX and insist on submitting Word docs, but that's about all I can think of. Why not just tell the Word users to suck it up and learn LaTeX?
Authors submit to specific journals for two main reasons: exposure and prestige. Whether this is a flawed system or not (I think it is), the fact is that, if you're a large successful journal, you can force authors to submit in whatever format you want - but then of course, you probably don't have money/cost problems. Smaller journals trying to survive or start-up journals don't have the luxury of telling authors to "suck it up". The authors will simply submit to another journal of the same class that's more lenient about their submission requirements.
Like I said, whily you may be comfortable with LaTeX and willing to take on much of the layout, etc. the vast majority of authors are not - whether through being too lazy, too busy, or simply not technically-inclined enough. Bear in mind that there are plenty of social science researchers out there who have a hard enough time with Word, much less LaTeX. If you make submission too difficult, authors will go elsewhere.
XML - and in particular the NLM format (by far the most comprehensive, in my experience), is a format that lends itself much better to multi-layout transformations than LaTeX. It also has the advantage that it can be fairly easily generated from source files, such as .DOC, via things like OpenOffice and XHTML. There is still some manual work required, but I've spent the past 3 years working on automated ways to convert articles in formats authors are comfortable with (like .DOC, but some even use .RTF or weirder) into XML. It's doable, and we're getting much closer.
The simple fact is, journals still take work, and work costs money. The Internet (and in particular, the Web) has brought an alternative for academics to expose their research other than journals (eg. IR like arXiv), but the journal paradigm still exists, and for the most part, it is the main venue for academic publication. All the social structures built around journals, like peer-review, editorial boards, etc. aren't going to disappear overnight.
If you want to submit all your papers in LaTeX to arXiv, you go for it. Unfortunately, many, many authors don't have that option.
I think the arXiv project has been highly successful, and is a great example of what instituional repositories should look like. But like it or not, there is still a market and expectation among many academics for glossy-looking journals. They explicitly want the things that improve the reading experience - in particular, they want to be able to print good-looking paper (ie. PDF) versions of papers, and are willing to pay extra for that. Maybe things will develop where the journal as medium becomes extinct, but not ay time soon.
Right now, the arXiv's goal is not to replace journals, but to be as fast and as efficient as possible at hosting preprints. If they set out to replace journals outright, and added a slighly more sophisticated review system and automated validation of input, they could do what most journals do with about 90% of the quality for about 10% of the cost.
What you're describing is exactly what OJS does: sophisticated review/workflow system with a little automated validation. It provides about 90% of the quality of the journals you're describing with almost zero cost. It has been wildly successful, especially in the developing world, where starting and hosting a journal is expensive. If we're going to really break it down, how would this review/validated arXiv system you describe differ from a journal? Not much, it seems.
I don't know too much about XML journal article formats, and would appreciate more info. PubMed's explanation didn't really help much. Why bother with XML? Why not just require authors to submit in LaTeX?
Authors submit to specific journals for two main reasons: exposure and prestige. Whether this is a flawed system or not (I think it is), the fact is that, if you're a large successful journal, you can force authors to submit in whatever format you want - but then of course, you probably don't have money/cost problems. Smaller journals trying to survive or start-up journals don't have the luxury of telling authors to "suck it up". The authors will simply submit to another journal of the same class that's more lenient about their submission requirements.
Like I said, whily you may be comfortable with LaTeX and willing to take on much of the layout, etc. the vast majority of authors are not - whether through being too lazy, too busy, or simply not technically-inclined enough. Bear in mind that there are plenty of social science researchers out there who have a hard enough time with Word, much less LaTeX. If you make submission too difficult, authors will go elsewhere.
XML - and in particular the NLM format (by far the most comprehensive, in my experience), is a format that lends itself much better to multi-layout transformations than LaTeX. It also has the advantage that it can be fairly easily generated from source files, such as
The simple fact is, journals still take work, and work costs money. The Internet (and in particular, the Web) has brought an alternative for academics to expose their research other than journals (eg. IR like arXiv), but the journal paradigm still exists, and for the most part, it is the main venue for academic publication. All the social structures built around journals, like peer-review, editorial boards, etc. aren't going to disappear overnight.
If you want to submit all your papers in LaTeX to arXiv, you go for it. Unfortunately, many, many authors don't have that option.
