From owner-chemistry@ccl.net Sat Apr 25 11:21:01 2015 From: "N. Sukumar nagams|a|rpi.edu" To: CCL Subject: CCL: DropBox / GitHub / etc for ESI Message-Id: <-51322-150425080222-13328-XunOmlIblBAHoklGfeC3+g[*]server.ccl.net> X-Original-From: "N. Sukumar" Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=UTF-8; format=flowed Date: Sat, 25 Apr 2015 17:32:11 +0530 MIME-Version: 1.0 Sent to CCL by: "N. Sukumar" [nagams|rpi.edu] The twin problems of ensuring data integrity and data longevity are non-trivial. Any solution has to make long-term economic sense. With the proliferation of pay-to-publish journals, I'm not sure of the longevity of the articles themselves, let alone the supplementary data. In a pay up-front model, how does the publisher ensure support for the data "in perpetuity"? It is no surprise that publishers may want the authors' institutions to support the ESI, even when the authors transfer the article copyright to the publisher. But what happens when the author moves to another institution or to industry or retires? Will the author's former institution continue to support the data repository "in perpetuity"? -- N. SUKUMAR Professor & Head, Department of Chemistry Director, Center for Informatics Shiv Nadar University, India On 2015-04-24 22:45, James Buchwald buchwj],[rpi.edu wrote: > Sent to CCL by: James Buchwald [buchwj-.-rpi.edu] > I think one of the challenges for such an approach will be long-term > availability of the ESI data. The current model of "publisher hosts > the ESI" has the advantage that the publisher is responsible for > ensuring that the data continues to be available for as long as the > article is available. Moving to an external host could (would?) lead > to situations where the site goes out of business/suffers > irrecoverable data loss/etc and the ESI is potentially lost (and then > you get broken links in the original paper). > > Obviously this is less of a potential issue (in the short term) for > large sites such as GitHub, but there is still a question of how such > methods will persist in the long (30+ years) term. I think there is a > need for publishers to develop their own new ESI storage and retrieval > mechanisms for large data sets, etc., such that the publisher can > ensure the long-term availability of the data on the same timeframe as > the manuscript itself. > > Best, > James > > On 04/24/2015 09:51 AM, Brian Skinn bskinn(_)alum.mit.edu wrote: >> CCL-ers, >> >> Provision of sufficient electronic supplementary information for >> published articles (molecular geometries, etc.) appears to be an >> ongoing challenge for computational chemists. All too often there are >> interesting results worth repetition and/or extension, but which are >> challenging to perform because the source computational files are >> unavailable. Further, the current journal-site-based ESI publication >> model is quite confining, as (I assume?) there are size limitations on >> what publishers wish to host on their servers. >> >> So, an idea for discussion: What about posting ESI on third-party >> services such as Dropbox, Github, etc? ESI is meant to be public >> information, so security needs are minimal. Such services are well >> equipped to handle the file transfer loads that ESI would likely >> represent, and storage costs would either be nil, or comparatively >> small. Further, it would be possible to deposit binary files >> (wavefunction information, electron density files, etc.) that might be >> of significant use to outside researchers wishing to analyze, repeat >> and/or extend a given work. >> >> >> Pros? Cons? I'm curious what you all think. >> >> >> Best regards, >> Brian > > -- > James R. Buchwald > Graduate Student, Dept. of Chemistry and Chemical Biology > Rensselaer Polytechnic Institute > Graduate Research Assistant, Dinolfo Laboratory > Teaching Assistant, Experimental Chemistry IV> To recover the email address of the author of the message, please > changeConferences: > http://server.ccl.net/chemistry/announcements/conferences/