17th ACM Symposium on Operating Systems Principles December 12-15, 1999 Kiawah Island Resort, near Charleston SC http://www.diku.dk/sosp99 http://www.cs.dartmouth.edu/SOSP99/ SOSP99 paper submission instructions

SOSP99 submissions are now closed.

See the call for papers to get an idea of what we are looking for. This page provides instructions on how to submit your paper for consideration.

Important dates

• Deadline for receipt of submissions: May 17, 1999
• The real, absolute, final, honest deadline for receipt of submissions: 23:59 (Pacific Daylight Time), May 24, 1999.
  • You may take an automatic extension until May 24th without asking anybody. (In particular, please don't send email to the program chair asking for one!)
  • No further extensions will be granted under any circumstances, and no paper that is not completely submitted by the real, final deadline will be considered. Really.
  • Submissions that are "in progress" at the deadline will simply be discarded: the machine will be set to automatically disconnect itself from the network. Submit early! Avoid the rush!
• Acceptance notification: Aug 20, 1999
• Deadline for camera-ready final papers: Sep 24, 1999

To submit a paper

1. Heed the following words of advice:
   • Roy Levin and David D. Redell, An evaluation of the ninth SOSP submissions -or- How (and how not) to write a good systems paper ACM SIGOPS Operating Systems Review 17(3):35-40 (July, 1983).
   • Alan Snyder, How to get your paper accepted at OOPSLA OOPSLA '91 Proceedings, pp. 359-363.
   • Ralph Johnson et al, How to get a paper accepted at OOPSLA Panel at OOPSLA'93, pp 429-436.
   • Alan Jay Smith, "The Task of the Referee", IEEE Computer 23(4):65-71 (April 1990).

2. Generate your paper using the following guidelines:
   • No longer than sixteen (16) US letter sized (8.5"x11") or A4 pages including everything: references, title page, figures, appendices, etc.

   • A typeface no smaller than 10 point, preferably in 2 columns.

     Although we impose no rules about things like margins and line spacing, you can be sure that the reviewers will be thinking "can a good paper of the final appropriate length be culled from this?" if it looks as though too much has been squeezed on to the page. And if their eyes hurt, they'll be less inclined to be generous ...)

   • Generate either PostScript or Adobe PDF. No other formats will be accepted.

   • Note: you will have the opportunity to revise your submission after you have sent it in, up until the submission deadline.

   • Warning: the submission deadline applies even for papers that fail to print. You will not be able to fix a non-printing paper after the cut-off time, to ensure a uniform, fair process for everybody.
     • If you are unsure about whether you can generate something that can be printed, please send something in early, to allow time to sort out the snafus. Early submissions will be tested for printability on an HP LaserJet 4SiMX using US letter-size paper (8.5"x11"), and you will be notified by email if your submission has difficulties.
     • Make sure that any PostScript file you submit starts with the magic string %!PS-Adobe, not some wierd string of control characters. (This seems to be a particular problem for Windows users.)
     • Make sure that you have not asked for a printer-specific option like a paper tray or output bin.
     • Make sure that you aren't using any oddball fonts (the "standard 35" set of PostScript fonts is a good one to limit yourself to).
     • Since the reviewers are scattered across two continents, they will want to print your paper on both A4 and US letter paper sizes.

   • Substantially identical papers must not have been published elsewhere or be under consideration for publication elsewhere.

   • Authors must not be identified in the paper itself, either explicitly or by implication (e.g., through the references or acknowledgments).

     The idea here is to leave reasonable doubt in a reviewer's mind that the work is yours - at the least, muddying the waters about precisely who is in the author list. If you succeed, there'll be nagging doubts about whether the work was done by you or by somebody who has seen your body of work and cleverly built on it.

     It is usually impossible to completely disguise who you are - but it is not acceptable to throw up your hands and say "I shan't even try". Here are some ideas that people have used in the past:

     • If you have a particularly catchy, well-known project name, use a different name in the submission.
     • Don't be afraid to cite your own work. But at least disguise who you are, and where it came from. At an absolute minimum, hide all the author names and affiliations of the work emanating from you and your team. (If nothing else, this disguises precisely who the authors of the submitted paper are. Creating doubt as to whether a particular individual was involved isn't as good as completely hiding the originating group, it's a step in the right direction.)
     • Refer to your own earlier work as if it was written by somebody else - that is, include it in the Related Work section, and discuss its differences from the current submission as if it was work done by The Competition. (This is a variant of the prior strategy - and rather better, in many ways.)
     • Omit the acknowledgements section.
     • Leave off those funding agency contract numbers.

   The Program Chair will do something creative (but unpleasant!) to papers violating these rules, and you risk them not being further considered. (I've been considering things like randomly deleting pages until the paper is "small enough", for example, or dropping every nth byte from the graphics until the paper reaches the median paper byte-length.)

3. Submit it. [note: SOSP99 submissions are now closed.]

4. Wait patiently.

Questions may be addressed to the program chair:

John Wilkes <wilkes@hpl.hp.com>
Hewlett-Packard Laboratories, mailstop 1U13
1501 Page Mill Road
Palo Alto, CA 94304-1126, USA.
Tel: +1 (650) 857-3568

Important Terms

• "Reviewer" shall mean an SOSP program committee member, or somebody asked to perform a review of a paper for SOSP by such a committee member or another Reviewer. Such Reviewers are automatically deemed to accept all of these Important Terms and be bound by them.
• "Held in confidence" and all similar terms shall mean to apply a reasonable degree of care to maintain secrecy, in no event less than those measures a Reviewer or his or her employer would apply to material of a similar nature that they consider confidential.
• Please be aware that the submissions will be kept on a server machine that is accessible from the Internet. Neither the submissions nor their reviews will be held in encrypted form, and although reasonable care will be taken to limit the likelihood of these being accessed inappropriately, no guarantees can be offered, and a submitter runs the risk of hackers and errors in applying that care.
• Reviewers will be expected to hold accepted submissions in confidence from the time of submission up to the start of the first scheduled event of the conference. (For SOSP, that'll be the reception on the night before the first session.)
• Reviewers will be expected to hold rejected submissions in confidence until they are published elsewhere or the authors' explicit permission has been obtained to release them to others.
• Reviews and notes written by Reviewers are intended as private communications to the SOSP program committee and the authors of the submissions. The SOSP program committee and the Reviewers will be expected to hold such reviews and notes in confidence until the start of the Symposium, and will be expected to provide an appropriate degree of care for them after that point. (The intent is to allow a Reviewer to reuse a review for a resubmission of the same paper, or for them to communicate with the authors of a submisison about it, but not to allow them to discuss the contents of the review or notes with others.)
• The SOSP program committee will be expected to maintain the identities of the Reviewers in confidence until such time as the Reviewers provide explicit authorization otherwise.
• A submitter may waive their right of confidentiality at any time, either explicitly by communicating with the Program Chair by email or other written communication, or implicitly by allowing their submission to become visible outside their own organization without a burden of confidentiality imposed on it. In such cases, the Reviewer(s) shall no longer be required to hold the submission in confidence. Any obligation of secrecy or confidence shall also not extend to information which is in the public domain, or which is independently developed without direct or indirect access to a submission.
• Importantly, this conference is given for academic purposes, and no individual is receiving any money for being a Reviewer. It is consequently understood and agreed by all that no provision has been made for any financial recourse against the symposium (SOSP99), the SOSP program committee, the Reviewers, their employers, or anyone connected with them. Under no event will any of these entities or individuals working for them or their employers be financially liable for any reason whatsoever, based on any legal theory whatsoever.
• These Important Terms may be revised at any time. If a "substantial change" (as defined by the Program Chair) occurs, authors of submitted papers will be given a chance to rescind their submissions, and the Reviewers will be asked to destroy any copies they may have of the submissions and any related notes. (No such substantial changes are anticipated, but this gives the Program Chair an escape clause if it is needed.)
• Submitting a paper or participating as a Reviewer constitutes acceptance of these terms.

Last modified: Tue May 25 02:05:37 PDT 1999