Change page style: 

FT FAQs and Answers

You are here

On this page we attempt to answer the questions that have been raised as we have set up the Fast Turnaround program and discussed it with members of the Gemini community and their representatives. Please contact us as Fast.Turnaround at gemini.edu if you have questions that are not addressed here. In case of any conflicting information, the rules should be taken as definitive.

 

 


What is the overall workflow and timeline?
This graphic illustrates what will happen during one cycle in the FT program. Red boxes show fixed events (note that FT observations are no longer scheduled on fixed nights as shown in the figure), blue shows tasks done by the PI/co-I, and green shows work done by the FT support team at Gemini.


  • The last day of each month (Month #0) at 11:59 pm Hawaiian Standard Time (HST): Proposal deadline.
  • 1st of Month #1: Assignment of proposals for each PI or designated co-I to scientifically review.
  • 14th of Month #1, 11:59pm HST: Deadline for PI/co-I to submit all reviews.
  • 21st of Month #1: Gemini FT team completes technical assessments of scientifically highly ranked proposals, constructs the final lists of accepted programs, and advises each PI of the status of his/her proposal.
  • 10 days after the 21st of Month #1 (i.e. end of Month #1 or very early Month #2): Deadline for completion of PhaseIIs.
  • Months #2-4: Accepted proposals are "active" until the final night of Month #4; are inactive thereafter.
  •  

    What if I submit after the deadline?
    Immediately after each proposal deadline, the proposal handling software takes the submissions and begins the process of assigning reviewers. Proposals that are received after this process has started will take part in the next proposal cycle, at the end of the next month. If you submit just after a deadline and do not wish to take part in the next cycle, please contact us (fast.turnaround at gemini.edu) so we can delete your proposal. Proposals cannot be inserted into the active cycle after the deadline.

    What if I re-submit the same (but edited) proposal before the deadline?
    The FT team will manually remove older versions of the same proposal before beginning the review process. Repeated submissions can cause delays and problems, however, and we cannot guarantee that the right version will be used. We urge PIs to carefully review their proposals before submission so that re-submission is not needed.

    How does the review process work?

  • Immediately after each deadline, the proposal handling software assigns proposals to the reviewers named in the proposals (specified in the Phase 1 Tool). A keyword matching algorithm is used to attempt to assign proposals to reviewers working on related subjects, as far as possible. Ideally everyone will review eight proposals, but that number will be lower if fewer than nine proposals are received.
  • Within a day or two after the proposal deadline, each reviewer is emailed the link to a page on which he or she must accept or decline the conditions of the FT program: confidentiality and using the proposals only for the purposes of the review. (If the reviewer does not receive that email by two days after the deadline, he or she should contact FT asap.) Once they have accepted the terms, reviewers are each taken to a page showing the titles, abstracts, and investigators of the proposals they have been assigned. They must then declare whether or not there is any reason why they cannot provide an unbiased review of each proposal. See this blog post for screenshots of this and other web pages/forms
  • If any conflicts are declared, the reviewer is assigned replacement proposals. If not, the reviewer is sent the links to the proposals themselves, and to a review form. The review form asks for a grade and brief written assessment, along with the reviewer's assessment of her or his knowledge of the field of the proposal. The reviewer's self-assessment is not currently used to weight the scores, but may be used to assess how such a weighting would have changed the final grades and whether it should in fact be implemented.
  • Immediately after the review deadline (currently the 14th of the month), the software looks for the review forms that have been submitted, takes the mean of the scores for each proposal, and creates ranked lists for each telescope that are then available to the FT support team. A simple mean grade is used; we have investigated the effect of using a clipped mean and found that it would not have changed the outcome of any of the proposal cycles so far.
  •  

    How will my proposal be judged?
    Reviewers should assess proposals according to these assessment criteria, giving them an absolute grade as described on the program overview page (see the screenshot of the proposal review form linked in the previous question). Please note that, although we attempt to assign proposals to reviewers working in related fields using the keywords selected by PIs in their proposals, it is very likely that at least some reviewers will have little knowledge of the subject of your proposal. Proposals that are written for a broad audience have a greater chance of being successful.

    Will my proposal really be rejected if I don't submit all my reviews on time?
    Yes.

    Who will see my proposal?
    Your proposal will be read by 5-8 reviewers (PIs or co-Is of other proposals submitted for the same deadline, plus members of a standing, on-call panel when few proposals are received, see Rule 13). If any of the reviewers is a PhD student, the proposal will also be accessible to a PhD "mentor", as described below. All of these people will be required to agree to the terms of the FT program. The FT support team at Gemini will have access to the proposals, to enable technical reviews to be performed and the final queue to be constructed. Finally, during the initial trial periodaccepted proposals were read by 1-2 members or former members of the Gemini National TACs. This is so they could judge how the quality of the proposals compares to those awarded time through the regular, semester-based process, which formed part of the assessment of the FT pilot program.

    How can you be sure that the review process is fair?
    Not surprisingly, this was the subject of much debate as we designed the FT program. We read the Merrifield & Saari (2009) paper on distributed peer review, followed the discussion of that paper on the Facebook Astronomers group, and dipped briefly into psychological studies of lying and cheating. We had much discussion with Gemini's users and advisory committees, tested the system on volunteers from the Canadian community (see these SPIE proceedings), and the program was reviewed by a panel of internal and external experts. In the end, we decided not build in numerical safeguards (hard to calibrate until we have some experience, may penalize inexperienced reviewers), and to start with a relatively simple system. During the initial trial period we monitored the review process closely and saw no obvious signs of bias or undesirable behavior. We continue to monitor the program. The FT program's peer review system will probably not appeal to everyone, but it should be seen as just one of a number of ways of gaining access to telescope time.

    What determines which proposals are accepted?
    The FT team selects the programs with the highest rankings from the peer-review process that are technically feasible and that will fit in the available time. We attempt to roughly fill the likely weather conditions and avoid accepting more observations than can be executed at a given RA. For example, if the two highest-ranked programs both require all of the 20 percentile image quality that is likely to be available (statistically speaking), then only one will be accepted. This does require some subjective decisions on the part of the FT team, but so far we have not had to reject highly-ranked programs for these reasons.

    How can graduate students be involved?
    We encourage graduate students to be involved in the FT program, and students may submit proposals as PI. When an FT proposal is submitted through the PIT, the PI or a co-I must be nominated as the person who will review the other proposals. If the reviewer is a graduate student, then a "mentor" must also be selected. The mentor will receive the same conflict of interest form as the student, and will also receive access to the proposals. The mentor is expected to provide whatever guidance is necessary for the student during the review process.

    Can I submit a proposal that requests really good conditions?
    You can, subject to the restriction that you don't request more time than is likely to be available in your most restrictive observing condition bin (see the Rules and the current Call for Proposals). However, given that the FT program can only allocate up to 20 hours per month, and does take the likelihood of various conditions occurring when allocating time, you will need to decide whether the risk of those conditions not occurring is worth the time you will put into writing the proposal, reviewing others, and setting up the observations.

    How strict are the RA limits in the call for proposals?
    We view the RA range as a guideline rather than as strict limits. If your observations are short and could be executed in the first or last few FT nights of the cycle, then targets a little outside the RA boundaries will be OK. On the other hand, if you submit many observations right at the edges and requiring very good conditions, that would count against your proposal as we attempt to select the highest-ranked programs that are actually feasible. We are deliberately not setting strict RA limits while we gain experience constructing a feasible FT queue each month; we’d rather leave it to users to decide what they think is worthwhile proposing for. We will adapt the program according to users’ feedback as we go along, though, so please contact us if you wish to provide feedback on this process.

    How much time can I ask for?
    The only restriction is that you don't request more time than is likely to be available in your (single) most restrictive observing condition bin (see the Rules and the current Call for Proposals). For example, if your proposal requires 85%-ile cloud cover but all other conditions (IQ, water vapor, sky background) are unrestricted, you could in principle request 85% of the time advertised in the CfP.

    What constitutes a "non-standard" instrument configuration or observing mode?
    This would include techniques like using the acquisition cameras to observe an occultation, or drift-scanning over a globular cluster with GMOS. While the Fast Turnaround team was set up partly to include a wide range of experience and expertise, we won't be able to support every conceivable observation on the short timescales involved. Rather than attempt to provide an exhaustive list of what can and cannot be proposed for through the FT program, we encourage users to contact us (fast.turnaround at gemini.edu) if they are considering anything out of the ordinary.

    What if someone proposes for observations that are already in the queue?
    As stated in the Rules, accepted programs have priority over new FT proposals. The phase I tool will warn if targets have already been observed by Gemini. This information will be in the pdf files that are available to reviewers, so proposers should give a clear justification of why the "duplicate" observations are necessary, or how they differ from those in the archive. Planned observations are trickier. We cannot make the target lists of accepted queue programs public, so the FT team will check for potential duplications in FT proposals before they are accepted. The titles and PIs of accepted Gemini programs are available here. If you are thinking of writing an FT proposal but suspect there may already be a program doing very similar observations, feel free to get in touch (fast.turnaround at gemini.edu).

    How is it determined whether my accepted FT program is in Band 1 or Band 2? What about Band 3?
    In the OT, you will see that your FT program has been assigned a band (1 or 2). In order that the queue coordinators have some way of distinguishing the relative priorities of FT programs (which will not be obvious from each program's ID number), we have appropriated the usual ranking band system. FT programs are not assigned Band 3 (once FT is selected in the Phase I the Band 3 option disappears from the Time Request options). The top-ranked ~1/2 of each month's programs will be assigned band 1, the next ~1/2 band 2. We aim to accept FT programs that stand a good chance of being completed, and we will attempt to complete all of them. However, in case of poor weather etc. using the ranking band system will allow us to prioritize programs according to how they were rated by their peers. The usual queue program completion rate goals by science band are not relevant to the FT program.

    How does Fast Turnaround differ from Director's Discretionary Time?

  • Currently, DD time is specifically aimed at urgent, high-impact, and risky proposals;
  • DDT proposals are scientifically and technically assessed by the Gemini Director of Science or her/his designees;
  • Accepted DDT programs become part of the normal queue observing process;
  • Accepted DDT programs usually have Band 1 priority and are almost certain to be observed (or attempted);
  • The timeline for a DDT proposal to be assessed and (if accepted) its Phase II to be prepared can be significantly shorter than the FT timeline.
  • Can Target of Opportunity (ToO) observations be applied for?
    Both standard and rapid ToO observations can now  be applied for through FT proposals. 

    How much time is being used for the FT program?
    Each partner except Australia has contributed 10% of their Gemini North time to the pilot. Using the science time given in the 2015A call for proposals (for example), this translates to 153 hours in 2015A, or about 3 nights/month (before weather loss). Each call for proposals gives the total time available for each partner.

    Should my publications that include FT data acknowledge the FT program, and if so, how?
    Not specifically. However, the publication should include the program ID, which includes an FT identifier, as well as one of the standard Gemini acknowledgements. If you would like to refer to the FT program in a publication, please use these SPIE proceedings.

    How can I give feedback?
    We would really like to hear your opinions about what does and doesn't work. Feedback forms will be sent to all applicants, but (potential) users should also feel free to contact the FT team directly (fast.turnaround at gemini.edu)


    Gemini Observatory Participants