The SDSS-III Publication Policy

Approved Oct 9th 2009 by SDSS-III Advisory Council

[Types of Papers] [Publication Coordinators] [Participants] [Architects] [Authorship] [Publication Procedure]

1. Introduction

This document describes the policies and guidelines governing the publication of scientific and technical results from the Sloan Digital Sky Survey III (hereafter SDSS-III). The SDSS-III Publication Policy is designed to promote the scientific and technical accuracy of SDSS-III publications and ensure that fair credit is given to the authors and to other individuals who have contributed to the SDSS-III. In addition, SDSS-III participants should be encouraged to carry out excellent science with the database and to publish their results in a timely manner. By agreeing collectively on how publication issues will be handled, we aim to minimize future disagreement among SDSS-III participants. Unless explicitly stated otherwise, this document applies to publications based on data taken with the SDSS-III telescopes and instruments, which are not yet in the public domain, as well as data taken with the survey equipment during commissioning periods.

This policy applies to: papers submitted to electronic archives and refereed journals; scientific or technical books or book sections about the SDSS-III and its results; graphical or tabular materials or discussions of results, in electronic or hard-copy form, that are based on analysis of unreleased SDSS-III data and are generally accessible to non-SDSS participants (e.g., via a public WWW site or a publicly distributed preprint). The special cases of colloquia, conference presentations, conference proceedings papers, and theses are discussed in Section 7. Communications which are not directly covered by this document, but the dissemination of which should be consistent with it, include: popular articles and books, circulars, telegrams, electronic announcements, press releases, and press conferences.

Responsibility for oversight of SDSS publications rests with the SDSS-III Scientific Spokesperson. The SDSS-III Collaboration Council (hereafter CoCo), which serves as an advisory body to the Spokesperson, has been charged with formulating this Publication Policy and ensuring that it is efficiently implemented to the benefit of the collaboration. The CoCo consists of one member from each Participating Institution with three or more Participants, plus one member who represents, and is elected by, the Associate Institutions with fewer than three Participants.

As the Survey progresses, the Spokesperson may revise this Publication Policy from time to time in what he or she perceives to be the best interests of (and in consultation with) the collaboration.

2. Types of Papers

We distinguish three types of publication: "scientific papers", "technical papers", and "data release papers". Scientific papers are based on analyses of, or presentations of, the SDSS-III data. Technical papers describe the SDSS-III instrumentation, calibration, software, strategy, and targeting algorithms. Technical papers may include some SDSS-III data for illustrative purposes. Data release papers describe the contents of SDSS-III data releases to the public.

3. Publication Coordinators

Logistical responsibility for handling scientific, technical, and data release papers will rest with the Scientific Publications Coordinator (SPC), Technical Publications Coordinator (TPC), and Scientific Spokesperson, respectively. Unresolved disputes about publication matters will be settled by an Ombudsperson. The SDSS-III Director, in consultation with the Management Committee, will appoint the SPC and TPC, while the Ombudsperson is appointed by the ARC Board as outlined in the SDSS-III PoO. The Publications Coordinators will ensure that publications follow the publication process outlined in Section 7 below. To aid in this process, they will maintain a common electronic WWW-based archive of all scientific, technical, and data release publications of the SDSS-III, hereafter the Publications Archive, with accompanying graphical and tabular material. The Publications Archive shall be accessible to the participants only. The Publications Archive will contain the following proprietary information and links to public documents:

projects under investigation or in preparation,
papers available for review by the collaboration
papers approved for public distribution and journal submission
other SDSS-III documents of general interest, such as related papers by SDSS-III participants not based upon SDSS-III data, common proposals, etc.
the SDSS-III acknowledgement script required on publications (Section 7.6)
the current list of SDSS-III Architects and participants (Sections 4 & 5)
the current version of this Publication Policy.

The papers in this archive that have been approved for public distribution can therefore be used by any SDSS-III participant for purposes of public presentation, review articles, etc.

The Scientific Spokesperson will have overall responsibility for the Publications Archive. The TPC will maintain a complete list of technical publications describing the hardware, software, etc on the Publications Archive. The SPC is responsible for assuring that all SDSS-III papers (scientific, technical, and data release) reference the appropriate technical papers. The TPC will help coordinate publication of these technical papers to ensure that technical documentation of the project is disseminated efficiently and promptly. The Scientific Spokesperson will coordinate the publication of the data release papers.

4. Participants

Participants are defined in Section 3.1 of the SDSS-III PoO. An updated list of participants shall be maintained by the Spokesperson, with input from the participating institutions. All participants must read and agree to the contents of this document before accessing the SDSS-III data. All participants must follow the rules of authorship and publication described in this document. Those who violate these rules are subject to losing their participant status.

The Spokesperson will keep the list of participants up-to-date and posted on the Publication Archive.

4.1 Non-Participants

According to Section 3.1 of the SDSS-III PoO, Participants may request permission to share a limited and approved portion of the SDSS proprietary data archive with non-Participants of the SDSS-III for collaborative research on a specific topic or project.. The policy regarding use of proprietary SDSS data by these non-participants, or external collaborators, can be found in the

Policy for external collaborators

Once granted limited data access, non-participants must read and agree to the contents of this document and follow the rules described herein. Non-participants violating these rules are subject to losing their data access.

5. Architects

In order to ensure that proper credit is given to those who were responsible for the construction and maintenance of the hardware, software, and other infrastructure of the SDSS, a list of "architects" will be maintained by the Spokesperson and posted on the Publication Archive.

5.1 Construction of the Architect list

Architects are those who have made substantial contributions to the SDSS-III hardware and software as a whole, including (but not limited to) the areas of optics, telescope, infrastructure, calibration, camera, spectrographs, data reduction and archiving software, commissioning, management, and major fundraising. The definition of "substantial contribution" is a total of one year of effort across the whole SDSS-III collaboration (see SDSS-III PoO as well). This can include multiple small (part-time) contributions to all the SDSS-III surveys that total to a year of effort spread over several years. Work done for SDSS-I/II does not count to this total.

To request Architect status, an individual (or someone acting on behalf of the proposed Architect, e.g., a CoCo representative) should submit a brief summary of their contributions to the CoCo for assessment. The justification must extend beyond merely being on the SDSS-III payroll, or being involved in SDSS-III science activity, for the appropriate length of time.

The request should also include which of the four main SDSS-III surveys (MARVELS, BOSS, SEGUE, APOGEE) the Architect status is requested for. Requests for multiple (or all) SDSS-III surveys are allowed and encouraged, but should be justified, e.g., people making contributions to the general SDSS-III infrastructure (telescope, data archives, etc.) that aids all four of the SDSS-III surveys. The CoCo will then make a formal recommendation to the MC (who makes the final approval) on all requests for Architect status and may restrict that recommendation to particular SDSS-III surveys based on the details of the request and the CoCo discussions.

The Scientific Spokesperson will keep the Architect list up-to-date (with details of which surveys apply) and posted to the Collaboration web page. Since individuals become members of the above lists by dint of work already performed, they cannot be removed unless they explicitly request removal. Architects remain on the Architect list when they leave an SDSS-III participating institution, and even in the unlikely event that they commit sufficient infractions to have their data rights revoked.

Once approved, an Architect can add their names to any science paper from the surveys they have been approved for. This benefit does not extend to technical papers as the author list of such papers will be directly related to the people working on a particular component or subsystem (see Section 6). As stated in Section 6.1.1, Architects adding their names to SDSS-III paper should have read and broadly understood the main contents and results of these papers, especially items close to their expertise and contribution(s).

6. Authorship

6.1 Scientific papers

6.1.1 Author inclusion

At the very least, any scientific paper will include in its author list any individual who has contributed to the scientific analyses presented, along with all architects who have requested authorship. In addition, in Section 6.2 of the SDSS-III PoO, it states that "Any member of SDSS-III who is authorized to have access to the Science Archive may, at any stage of a research project within the SDSS-III collaboration, request that his/her name be added to the list of authors, with the presumption that permission will be granted if he or she has made any significant contribution to that specific research project.” With the exception of architects, those requesting co-authorship should include a brief description of their contributions to the project. Authorship requests should be made to the paper's team contact (see Section 7). Project teams are expected to err on the side of inclusion in response to such requests.

Anyone who has been placed on a paper's author list may make a request to the paper's team contact at any time that her or his name be removed from the paper and it shall be removed.

We stress that all authors should have read and broadly understood the content of all papers they are co-authors on. This is statement reflects the spirit of the professional guidelines outlined by the APS , as cited in the PoO-III.

During the three-week period that the paper is available to the collaboration (see Section 7.1.ii), the SPC will check that the author list on each scientific paper is appropriate and that the appropriate technical papers are cited. Disputes about author inclusion will be settled by the Ombudsperson, whose decisions will be final.

6.1.2 Two-group author order

The scientific achievements of the SDSS will truly be collaboration achievements, because the SDSS hardware and software make possible projects that would in many cases be inconceivable without them (or without some similarly ambitious undertaking). However, an authorship policy that strictly recognizes all scientific papers purely as products of the collaboration (e.g., alphabetical authorship on all papers) runs the risk of making the individuals in the project effectively anonymous, and it provides little incentive for project participants to analyze the data quickly rather than wait for data to become public. Therefore, for most SDSS-III scientific publications, the order of authorship will follow a "two-group" system. The first group of authors, the "analysis group", will consist of those who were heavily involved in the specific analysis described in the paper and in the writing of the paper. The other group will consist of all other authors on the paper and will be ordered alphabetically after the analysis group authors.

Membership in the analysis group and order of authorship within the analysis group should be decided by those involved in writing the paper, just as they would decide on membership and order in the author list if they were writing a non-SDSS science paper. In the event of a disagreement about membership in the analysis group or order of authorship, the authors may present their cases to the Ombudsperson, whose decisions will be final.

The analysis group is always permitted to merge the two author groups and make the author order alphabetical.

6.2 Technical papers

Those who have worked on a particular component or subsystem will author the appropriate technical paper, with author order decided by the authors. Disputes may be taken to the Ombudsperson, whose decisions will be final. See Section 7.2 below.

6.3 Data-release publications

The Scientific Spokesperson and the Chairs of the Survey Science Teams (SSTs) shall be responsible for the production of journal papers that describe all of the SDSS-III data releases. The Spokesperson and SST Chairs, in consultation with the CoCo, shall also determine the author list, which for these papers will be alphabetical. Those who have contributed to the writing of the data release paper, or who have contributed in a substantive way (typically one month contribution) to the creation or science validation of the data described in the paper, are eligible to be authors.

6.4 Conference proceedings

Because of the practical constraints of page limits, submissions to conference proceedings are not required to follow the authorship policy above. However, any conference paper that does not follow this authorship policy may only include quantitative results (tables and figures, for example) based on papers that have been published or submitted for publication, and it must reference those papers as the primary source for the results. The rationale for this requirement is that conference papers authored by a single author or a small group should not become the primary reference for results that are properly products of the SDSS-III Collaboration. Conference proceedings which are primarily overviews of the SDSS-III project status or scientific results should include the byline "for the SDSS-III Collaboration" after the list of authors.

6.5 Theses

Whether they would otherwise qualify as scientific or technical papers, some undergraduate or graduate theses will be authored by a single author. However, it is required that the author will acknowledge the contribution of the SDSS-III collaboration and reference the papers describing the thesis results that have been published in or submitted to research journals.

7. Publication Procedure

7.1 Scientific papers

To ensure that SDSS-III publications fairly represent the contributions of those involved and that publications are vetted in a timely and efficient manner, scientific papers based on unreleased SDSS-III data will be handled as follows:

As soon as a specific scientific investigation begins, a project team of SDSS-III participants should identify a team contact who must then submit to the SDSS-III Projects Web Page (see link below) a project title, a list of current project members, and a brief abstract of the intended project. An electronic announcement of the project should then be automatically sent to the collaboration (e.g. via the sdss3-general email list). At this, and any other stage of the process, other SDSS-III participants can join the project.

Follow this link for details of how project announcements can be posted. This page requires the "standard" SDSS-III username/password.
When a draft of the scientific paper has been written, the project team must post the paper to the Publication Archive (see below) as a "pending" paper, and an electronic notification is sent to the collaboration (e.g., via an automatic message sent to the sdss3-general email list). This announcement must indicate the journal to which the paper will be submitted. SDSS-III participants then have three weeks to make comments, request co-authorship, etc, to the project team contact (with copies sent to the SPC). To streamline the process, an automated system to query architects or respond positively to their authorship requests should be used. The team contact is required to circulate substantive comments received on the paper's content, along with the team's response. The project team revises the paper in response to the comments and other requests it deems valid. The SPC, or a representative of the SPC, also checks that the authorship, references, and acknowledgements fairly represent the contributions made to the publication, in particular checking that the appropriate SDSS-III technical publications are referenced. The SPC carries out this duty within the same three week period. A publication can be withdrawn at any stage of this process.

Follow this link for details of how to announce SDSS-III publications. This page requires the "standard" SDSS-III username/password.
The authors revise the paper according to comments and suggestions from the collaboration. At the end of the three week period, or at the end of the revision process (whichever is longer), the final version of the paper is posted to the Publications Archive and, as a professional courtesy, a final version of the paper should be sent to all co-authors by the lead author(s) 48 hours before submission to a journal (and/or the arXiv archive) to solicit any final critical comments or issues. After this final stage, the paper, its results and accompanying material may normally be submitted for publication in a journal and to the arXiv archive (the latter is left to the discretion of the lead author(s)). The paper is not considered public until its embargo is lifted by the lead author(s), as indicated at its posting in the Publications Archive (or until its appearance in the arXiv archive, or the publication to which it was submitted, should the lead author(s) neglect to lift the embargo before then). Until the embargo is lifted, the paper and its results cannot be quoted in public, referenced in other publications, etc. without the permission of the lead author(s).
When a project team makes revisions in response to an external referee report, it should post the final accepted or published version on the Publications Archive. In cases of disputes about credit, authorship, or scientific results, the SPC shall work with the author(s) and SDSS-III participants to resolve them. In cases where a dispute cannot be resolved, the matter may be appealed to the Ombudsperson, whose decisions will be final.

Publications based on projects started with proprietary SDSS-III data should follow the publication procedure outlined in this section, even if the data being used has become public during the project and before any publications have been finished and posted to the SDSS-III publication page. This was standard practice in SDSS-I/II.

7.2 Technical papers

As technical papers have a special status within the SDSS-III, prospective authors of any technical papers should first consult with the chair of their appropriate SST (or TPC if equally applicable to all four SDSS-III surveys) for confirmation of this special status. If this pre-approval as a technical paper is granted, then the paper should follow the procedure below. In addition, the TPC can suggest technical papers that should be written and work with the Chair of the SSTs to ensure they are published in a timely fashion.

When a technical paper is drafted, it is posted to the Publications Archive. This announcement must indicate the journal to which the paper will be submitted. The TPC, or a representative of the TPC, will review the paper to ensure that the authorship, references, and acknowledgements fairly represent the contributions made to the publication.
The TPC and SDSS-III participants will have three weeks to send comments to/request changes from the lead author. Copies of participant comments should also be sent to the TPC.
The lead author posts the final draft on the Publications Archive and submits the paper to the appropriate journal and the arXiv archive. The contents of the paper may thereafter be quoted in public presentations, etc., by SDSS-III participants.
When a technical paper is revised in response to an external referee report, the lead author will post the final version on the Publication Archive.

In the case of disputes about technical publications, which the TPC cannot resolve, the matter may be appealed to the Ombudsperson, whose decisions will be final.

7.3 Data Release Papers

Major releases of SDSS-III data will be accompanied by a paper (or papers) describing the contents of these data releases. Such papers will be coordinated by the Scientific Spokesperson and the SST Chairs, and posted to the SDSS-III archives for three weeks for comment by the collaboration. The Spokesperson and SST Chairs will construct the author list(s), which should contain all appropriate SDSS-III participants and be in alphabetical order.

7.4 MARVELS Planet Detection Papers

The credibility and scientific success of MARVELS hinges on the reliability of its planet discoveries and characterizations, which could be easily undermined by erroneous claims. Therefore, a MARVELS paper that presents the first public announcement of a planet candidate must be subject to higher level of internal scrutiny than discussed in Section 7.2 above. Therefore, the Chair of the MARVELS Survey Science Team, in consultation with the MARVELS PI and Project Scientist, is allowed to delay the submission of any MARVELS paper that presents the first public announcement of a planet candidate if he/she considers the strength of the evidence for this candidate to be insufficient. Note that this also applies to papers that are derived directly from planet candidates discovered from unpublished MARVELS data, even if the MARVELS data are not included, as this effectively represents the first public announcement of the candidate.

Any imposed delay may not exceed six months from the time that the paper is first posted in complete form to the Publication Archive. Also, the delayed paper may be submitted as soon as any other SDSS-III paper that includes the candidate is submitted, and it may also be posted to ArXiv simultaneously with the ArXiv posting of any other paper that includes the candidate, or upon acceptance by a journal if that happens earlier. The decision by the Chair of the MARVELS Survey Science Team to delay submission of a paper can be appealed to the Spokesperson first, and then the Ombudsman, whose decision is final.

7.5 Colloquia, conference presentations, and conference proceedings

In oral or display presentations, colloquium or conference speakers and presenters may, upon consultation with the project team contact, make use of unpublished SDSS-III results (with the exceptions noted in 7.1.iii above) and are expected to give appropriate credit to the SDSS-III Collaboration and to those involved in producing the results shown or described. This policy also covers digital presentation slides published on conference websites soon after speakers have delivered their presentations. In such cases, only material presented in public by the speakers can be posted to conference websites (i.e., extra slides containing unpublished results must be removed if not presented in public as part of the talk).

Conference proceedings are required to follow the publication procedure described in this Section. In special cases, given the realities of conference proceedings submission deadlines, the authors may request from the Spokesperson an exception from some or all of the publication process requirements. Such requests should be submitted at least three weeks prior to the submission deadline.

Applications for talks/presentations at the annual American Astronomical Society (AAS) meetings should be announced to the appropriate SST mailing lists at least two weeks before formal submission of any abstract. Any conflicts that arise from such postings will be resolved by the SDSS-III Spokesperson, but in cases where a dispute cannot be resolved, the matter may be appealed to the Ombudsperson, whose decision will be final.

7.6 Theses

All theses in which SDSS-III data play an essential role must be announced to the collaboration when they are begun, just as with other publications (see Section 7.1 above). It is the responsibility of each SDSS-III institution to work out a process for publishing theses which is consistent with both SDSS-III and departmental policies.

Theses based on SDSS-III data should not be distributed to the public via the WWW or other electronic archives until either the data on which it is based are public or journal articles describing its main results have been published or submitted for publication. Any such distribution should note the journal articles describing the results, so that they can be appropriately cited by others.

The SDSS-III Spokesperson will maintain a list of approved SDSS-III theses in the Publication Archive. The contents of this list are included in quarterly and annual reports to demonstrate the contribution of the SDSS to graduate education.

7.7 Standard acknowledgement

All SDSS papers that fall under the purview of this policy must contain the standard SDSS-III acknowledgement, which is maintained by the Spokesperson and posted on the Publications Archive. If a conference proceeding is subject to a tight length constraint (four pages or less), the author may submit a request, to the appropriate publications coordinator, to include only part of the acknowledgement.

7.8 Dissemination of Time-critical Information

The SDSS-III will detect time-variable events and/or objects for which timely dissemination to the broader astronomical community will pay big scientific dividends. Participants who carry out programs with SDSS-III data that detect time-critical events are encouraged to report these to the collaboration as a whole, and to the broader community through IAU Circulars and the like. Given the space constraints and expense of IAU Circulars, the attribution for the report should read "The Sloan Digital Sky Survey III (see http://www.sdss3.org/) reports...". Relevant finding charts, positions, and photometry drawn from the SDSS-III data set can be released. Publication of time-critical results can also request to the Spokesperson and CoCo for a shorter review period (outlined in Section 7.1) if the scientific dividends are high and/or there is a fear of being scooped by others not part of the SDSS-III collaboration. Such cases will be dealt with on a case-by-case basis by the CoCo and Spokesperson.

Participants who wish to regularly release substantial amounts of time-critical data, such as positions and photometry of asteroids, must request approval from the SDSS-III Management Committee. Upon approval, the project must be posted to the Publications Page under the category of "On-line Catalog". At this point the project falls under the rules governing Scientific Publications (three week posting period, authorship rules). Since some (if not all) of these publications will be "living documents", individuals who contribute to the publication but were not part of the project when it was originally posted can, with approval of the lead author, be added to the list of authors at any time.

8. Page Charges

SDSS-III funds will not be used to pay page charges for any scientific paper. The SDSS-III will pay page charges for approved technical and data release papers, subject to the budget.