Loading ...
Sorry, an error occurred while loading the content.

Re: {MPML} Discovery credit rule change proposal

Expand Messages
  • Dave Herald
    There is another dimension to this whole discoverer status issue. It is something to be recognised as the discoverer of something rare. But who cares when the
    Message 1 of 28 , Sep 1, 2006
    View Source
    • 0 Attachment
      There is another dimension to this whole discoverer status issue.

      It is something to be recognised as the discoverer of something rare. But who
      cares when the thing is very common. We have already seen the situation where
      the allocation of proper names to asteroids is losing significance/importance
      due to the sheer numbers (it is one thing to have one of 4000 objects bearing
      your name; it is another if its merely 1 in 400,000).

      Surely the same thing is going to happen with new minor planet discoveries. If
      we are heading to a period of a major increase of detections, then surely a new
      minor planet discovery will become a bit like finding a new pebble in a paddock
      that no-one else has seen - basically, who cares! As the numbers grow,
      significance will be attached to the population of minor planets more so than
      individual minor planets. The impact on amateurs, of course, is that the
      likelihood of us finding a new minor planet in the future may be seriously
      reduced - and perhaps there is a grieving process surrounding that fact.

      In this vein, perhaps discoverer status and proper naming should only be
      considered significant for 'special' or unusual minor planets - such as
      potential Earth impactors. ....although I wonder how much you would want your
      name associated with an object that was actually going to impact the earth.
      Imagine the popular press - 'next year "Fred's asteroid" is going to wipe out
      all of humanity....'

      Dave Herald (3696)
      Canberra, Australia
    • Reiner M. Stoss
      ... ...*and* reported the object first. Full ACK. I am happy you have changed your opinion on this since we were debating about who is the discoverer of 2003
      Message 2 of 28 , Sep 1, 2006
      View Source
      • 0 Attachment
        Tommy wrote:
        > in a paper. But discovery (which in my view is a very overrated
        > credit) should go
        > to the one that saw and reported the object first.

        ...*and* reported the object first. Full ACK.

        I am happy you have changed your opinion on this
        since we were debating about who is the discoverer
        of 2003 EL61... at that time you supported MB's
        way of claiming discovery for things he never reported!

        If he really has a dozen more of these, I hope he reports
        them soon. Or he might see again that someone else
        finds one or the other and reports it. Several more surveys are
        underway meanwhile.

        R.
      • Dave Tholen
        ... Some people feel that way. If the IAU ever stops naming the bulk of the asteroids, it won t be long before the International Asteroid Registry steps in to
        Message 3 of 28 , Sep 1, 2006
        View Source
        • 0 Attachment
          > Or should we stop naming asteroids?

          Some people feel that way. If the IAU ever stops naming the bulk
          of the asteroids, it won't be long before the International Asteroid
          Registry steps in to make a buck.
        • Ladislav Nemec
          There are entrepreneurs who, for a small fee, name a star after you or Aunt Adelaida. Why not asteroids? Nobody pays attention to it at all and these
          Message 4 of 28 , Sep 1, 2006
          View Source
          • 0 Attachment
            There are entrepreneurs who, for a small fee, name a star after you or Aunt
            Adelaida. Why not asteroids? Nobody pays attention to it at all and these
            entrepreneurs are fairly harmless. If we try to stop them they may turn into
            mugging old ladies on the streets.
            LN



            -----Original Message-----
            From: mpml@yahoogroups.com [mailto:mpml@yahoogroups.com] On Behalf Of Dave
            Tholen
            Sent: Friday, September 01, 2006 11:14 AM
            To: mpml@yahoogroups.com
            Subject: Re: {MPML} Discovery credit rule change proposal



            > Or should we stop naming asteroids?

            Some people feel that way. If the IAU ever stops naming the bulk
            of the asteroids, it won't be long before the International Asteroid
            Registry steps in to make a buck.





            [Non-text portions of this message have been removed]
          • Andrea Milani
            We are glad to have contributed to MPML record month, and we thank all the people who have given their contribution to the discussion on the new discovery
            Message 5 of 28 , Sep 5, 2006
            View Source
            • 0 Attachment
              We are glad to have contributed to MPML record month, and we thank all
              the people who have given their contribution to the discussion on the
              new discovery rules. All the suggestions will be taken into account in
              the formal discussion within the IAU committees (Commision 20,
              Division III). We obviously cannot answer to each comment
              individually. However, we would like to make some general comments.

              1) Please note that we have taken the initiative of consulting you. In
              the past, the rules about discovery credit have been adopted by the
              Director of the MPC without any consultation whatsoever, and even
              "adapted" case by case by discretionary decisions.

              2) We have been surprised that many successful asteroid discovereres
              do actually know neither which are the rules currently in effect
              nor who has decided them. E.g., the rule is not and has never been
              "the one who sees it first is the discoverer", for the simple
              reason that this is not a logically consistent rule: what does it
              mean "see"? what does it mean "it"? Enough information has to be
              accumulated to make sure that one and the same object has been seen
              several times, to establish its nature and that it is not the same
              object already discovered by someone else.

              3) We respect and appreciate the work of the small observatories,
              including the amateur ones, and this is why we are consulting the
              subscribers of MPML (as nobody else does; please compare with
              the recent example of the definition of planet). On the other hand,
              this is not a good reason to despise and refuse to recognize the
              work of the professionals. If everybody was willing to respect the
              contribution of the other participants to the discussion, we would
              all be better off. As an example, the sub-thread on the "$500
              computer" was really insulting. Both the computer costs, and even
              more the manpower costs, of our research are quite large. Try to
              put together a 120,000 lines software system like OrbFit and you
              will understand it not a trivial task, and this work has to be
              acknowledged too. Of course we are no heroes, we are paid to do
              this, but this implies that we have to pass very high quality
              standards to get a job in the first place, and we need to have our
              work credited and published to keep the same jobs in hard times of
              cuts like the current ones.

              4) We are not entitled to speak on behalf of the Pan-STARRS project,
              but we can even less allow that, as a result of misunderstanding of
              what we have said, false rumours are spread on that project. The
              astrometry (and photometry) from PS will all be made public. The
              satellite trails have nothing to do with our discussion: indeed,
              they are removed by the USAF before entering the data processing
              pipeline (foreigners like ourselves will never be allowed to look
              at one raw image; does this disqualify us as observers? :-). The
              Memorandum Of Understanding between the IAU and the Smithsonian for
              the running of the MPC allows no secret data to be submitted to the
              new MPC, no exceptions are envisaged. The only issue is WHEN the
              data will be submitted to the MPC, thus made public. While many
              issues have yet to be decided, one think is clear because it was
              one of the basic design choices of PS: the data will be processed,
              nut just dumped raw for some external center to find out if
              anything has been discovered; in other words, the people working
              for PS are astronomers, not technologists interested just in
              showing they can produce many raw data. PS has invested $millions
              in developing the software of the processing pipeline, and dozens
              of people have been working to this already for years. In
              particular, suggesting that all the PS data will be processed by
              some Italian astronomers is an insult, especially to the MOPS team
              who has been working full time to handle the moving objects. The
              processing will be performed by using the know-how from a large,
              world-wide consortium, and the software will be later used by LSST,
              DCT, even by a proposed NEO space survey. This implies that the
              data of PS will not be made public BEFORE being processed. On the
              other hand if they were, the MPC would need the computer hardware
              and software being put together for PS and for the other next
              generation surveys, which is quite unrealistic.

              5) The problem to be solved is that the deeper a survey, the more
              complicated is the problem of identifying the observations from the
              same object; looking one by one with human eyes the candidate
              discoveries to validate them is out of question. Follow up of
              possible discoveries is going to be the central problem: the
              surveys, including PS, need to be encouraged to do as much as
              possible their own follow up, but even if they succeed in this at
              the 90% level, there will be much more need for follow up than
              there is now. Thus discovery credit rules need to encourage and
              reward, although it is just a moral reward, the work which is
              needed, namely follow up and identification. If there are no
              changes to the rules, our estimate is that the big new surveys will
              simply take all the credit, and then they will be left with
              insufficient follow up. In our opinion, the astronomical community
              needs new rules, and we want an open discussion based on a
              scientific study of the effect of new rules.

              Having said all this, we are glad if you can propose a new set of
              rules, or even some consistent modification of the rules proposed by
              us; to be honest, we have not had much of this so far in this
              thread. Please refer to the last version of the rules, to be found at

              http://adams.dm.unipi.it/~milani/preprints/disc_rul_iau.pdf

              and please take into account that to read what you want to criticize
              is a good way to start.

              Yours

              Andrea Milani, Giovanni Gronchi and Zoran knezevic

              ================================================
              Andrea Milani Comparetti
              Dipartimento di Matematica
              Piazzale B. Pontecorvo 5
              56127 PISA ITALY

              tel. +39-050-2213254 fax +39-050-2213224
              cellular phone +39-349-4482751
              E-mail: milani@...
              WWW: http://copernico.dm.unipi.it/~milani/
              ================================================
            • Tommy Grav
              I have already made known my position on what I consider a discovery, but since Andrea is asking for comments on his proposal :) First of all I think this
              Message 6 of 28 , Sep 6, 2006
              View Source
              • 0 Attachment
                I have already made known my position on what I consider a discovery,
                but
                since Andrea is asking for comments on his proposal :)

                First of all I think this proposal is terribly self serving for
                Andrea's group and
                red flags go up for me when I see that. Second the proposal gives no
                argument
                (as far as I can see) why the orbital computer should have discovery
                rights. It is
                clear to me that the orbital computers of the world feel that they do
                not get enough
                credit for the work that they do. However, the question has to be
                asked, why steal
                from someone, the observers, that is actually on the same team as
                you? Is it not
                possible to come up with a system where the discoverer (the person
                who identified
                the object in an image) gets listed as that, and the linker/orbit
                computer gets listed
                as just that. For example if an observer observers the object on two
                nights and
                correctly identifies the object as the same:

                2007 AA1
                Discovered by: A. Observer
                Linked by: A. Observer

                While if the observer(s) observers the object but not realizing that
                it is the same object
                we get:

                2007 AA2
                Discovered by: A.Observer and B. Observer
                Linked by: L. Inker

                I just don't see the need for the orbital computer to call himself
                the discoverer. Is
                being the linker not "cool" enough? Maybe it is time for the orbital
                computers to
                seek other ways of publishing their results to get recognition,
                instead of trying to
                steal it from the observers.

                As the proposal reads now it is very hard to make sense of the tests
                that
                have been done. There is no description of the values in the tables
                so how
                can one really comment on the validity of the claims?

                Also how does the numbers work out for things like TNOs or even other
                Sednas?
                How far apart do the observations need to be before your proposal
                calls the object
                a discovery (Arc Type 3)? It seems crazy to me that you can see the
                object move,
                I can do spectra on it and other cool stuff. But I won't be the
                discoverer because
                I didn't get all the observations needed to get the object to a arc
                type 3? I would
                like to ask why dynamics in this case is so special compared to other
                types of science
                (remember that in the planet discussion the dynamics community was up
                in arms
                about the fact that only one type of science was used to define the
                term)?

                Finally I really think that pushing the big surveys as a reason for
                why a new set of
                rules have to be adopted is really construed. I see no reason why the
                current system
                will not work perfectly fine with all the new major surveys.

                Tommy
                tgrav@...
                http://homepage.mac.com/tgrav/

                "Any intelligent fool can make things bigger,
                more complex, and more violent. It takes a
                touch of genious -- and a lot of courage --
                to move in the opposite direction"
                -- Albert Einstein
              • Tommy Grav
                ... Just so that it becomes clear. Who are in Commision 20, Division III? And what is its charter? ... Since that came from me, I would like to point out that
                Message 7 of 28 , Sep 6, 2006
                View Source
                • 0 Attachment
                  On Sep 5, 2006, at 11:17 AM, Andrea Milani wrote:
                  > We are glad to have contributed to MPML record month, and we thank all
                  > the people who have given their contribution to the discussion on the
                  > new discovery rules. All the suggestions will be taken into account in
                  > the formal discussion within the IAU committees (Commision 20,
                  > Division III). We obviously cannot answer to each comment
                  > individually. However, we would like to make some general comments.
                  >
                  Just so that it becomes clear. Who are in Commision 20, Division III?
                  And what is its charter?

                  > 2) We have been surprised that many successful asteroid discovereres
                  > do actually know neither which are the rules currently in effect
                  > nor who has decided them. E.g., the rule is not and has never been
                  > "the one who sees it first is the discoverer", for the simple
                  > reason that this is not a logically consistent rule: what does it
                  > mean "see"? what does it mean "it"? Enough information has to be
                  > accumulated to make sure that one and the same object has been seen
                  > several times, to establish its nature and that it is not the same
                  > object already discovered by someone else.
                  >
                  Since that came from me, I would like to point out that I never said
                  that it
                  was the current rule. That was what I *think* the rule *should* be.
                  And I
                  disagree that the rule is not logically consistent. If you can prove
                  that you
                  were the first to observe it and there is enough evidence that you
                  realized
                  what it was you are the discoverer. Yes there has to be some judge to
                  handle the few hard cases, but it is still logically consistent. I
                  however would
                  point out that the orbit is not the only thing to make sure that it
                  is the same
                  object seen over and over again. Examples are a very slow moving object
                  that will not move far in days or weeks. Another is the case of a
                  space craft
                  passing a small asteroid that can not be detected from earth. Is the
                  object
                  not discovered in either case, even though an orbit those not exist?

                  > 3) We respect and appreciate the work of the small observatories,
                  > including the amateur ones, and this is why we are consulting the
                  > subscribers of MPML (as nobody else does; please compare with
                  > the recent example of the definition of planet). On the other hand,
                  > this is not a good reason to despise and refuse to recognize the
                  > work of the professionals. If everybody was willing to respect the
                  > contribution of the other participants to the discussion, we would
                  > all be better off. As an example, the sub-thread on the "$500
                  > computer" was really insulting. Both the computer costs, and even
                  > more the manpower costs, of our research are quite large. Try to
                  > put together a 120,000 lines software system like OrbFit and you
                  > will understand it not a trivial task, and this work has to be
                  > acknowledged too. Of course we are no heroes, we are paid to do
                  > this, but this implies that we have to pass very high quality
                  > standards to get a job in the first place, and we need to have our
                  > work credited and published to keep the same jobs in hard times of
                  > cuts like the current ones.
                  >
                  There are however many other avenues to go to get your work credited
                  (as others has pointed out). You have not (in my view) argued
                  succesfully
                  why getting discovery credit is the right one. It is certainly the
                  easy one,
                  but not necessarily the right one.
                  > 5) The problem to be solved is that the deeper a survey, the more
                  > complicated is the problem of identifying the observations from the
                  > same object; looking one by one with human eyes the candidate
                  > discoveries to validate them is out of question. Follow up of
                  > possible discoveries is going to be the central problem: the
                  > surveys, including PS, need to be encouraged to do as much as
                  > possible their own follow up, but even if they succeed in this at
                  > the 90% level, there will be much more need for follow up than
                  > there is now. Thus discovery credit rules need to encourage and
                  > reward, although it is just a moral reward, the work which is
                  > needed, namely follow up and identification. If there are no
                  > changes to the rules, our estimate is that the big new surveys will
                  > simply take all the credit, and then they will be left with
                  > insufficient follow up. In our opinion, the astronomical community
                  > needs new rules, and we want an open discussion based on a
                  > scientific study of the effect of new rules.
                  >
                  I really disagree in this assertion. Follow-up has never given much
                  credit but a lot of people have done it anyways. Why should this
                  suddenly change. And I really don't understand how this argument
                  justifies giving discovery credit to orbital computers?

                  > Having said all this, we are glad if you can propose a new set of
                  > rules, or even some consistent modification of the rules proposed by
                  > us; to be honest, we have not had much of this so far in this
                  > thread. Please refer to the last version of the rules, to be found at
                  >
                  Proposing a new set of rules takes time. Yours were not done in a
                  few weeks either. So if you want other suggestions you are going
                  to have some patience :) When will commision 20, division III going
                  to handle this issue?

                  Cheers
                  Tommy
                  tgrav@...
                  http://homepage.mac.com/tgrav/

                  "Any intelligent fool can make things bigger,
                  more complex, and more violent. It takes a
                  touch of genious -- and a lot of courage --
                  to move in the opposite direction"
                  -- Albert Einstein
                • Tommy Grav
                  Hi Andrea, I re-read your proposal and had a few more questions. How do you envision that the IAU Data Center handle the submission of least square solution of
                  Message 8 of 28 , Sep 8, 2006
                  View Source
                  • 0 Attachment
                    Hi Andrea,

                    I re-read your proposal and had a few more questions. How do you
                    envision that the IAU
                    Data Center handle the submission of least square solution of orbital
                    computations? It seems
                    to me that adding this to the things that the IAU DC has to be able
                    to process automatically
                    would further increase the pressure of the center in a time when they
                    are already working hard
                    on improving their system. Do you have a specific plan in mind for
                    implementing this? Also, what
                    is the envisioned policy of submissions that due to format error can
                    not be processed automatically?

                    Also how do you envision the dessemination of data from the IAU DC to
                    the public? One of
                    the fundamental flaw, in my mind, of your proposal is that it is the
                    person that downloads the
                    available data first and has an automated pipeline to handle the
                    orbital linking and computation
                    that will win. This means that everyone being in on the game will try
                    to figure out when the data
                    is being released and have bots getting ready to download it. This
                    could easily lead to denial
                    of service problems as the IAU DC servers could be flooded with
                    requests for the newly published
                    data? Someone could even conceivably create a program that access the
                    server and starts the
                    download while at the same time denying access by all others. How do
                    you envision a fair handling
                    of this issue?

                    Also as Mr. (Dr.?) Dixon pointed out, what is the envisioned plan for
                    handling disputes. What if two
                    people send in an orbital linkage or computation at the same time.
                    And how good does the fit need
                    to be to be accepted? How do one handle borderline cases were on fit
                    is just higher then the accepted
                    fit, while another is just lower?

                    It seems to me that the rules you propose will lead observers to not
                    send in their data to the IAU DC
                    in hope that they can do their own follow-up and compute their own
                    orbit? Is this secrecy of data an
                    intended consequence? If not how do the community still encourage
                    people to submit their data
                    quickly under the new rules?

                    Cheers
                    Tommy
                  • Andrea Milani
                    Please note I never intended to answer one by one the messages of this thread: this is not a discussion with Milani , but a discussion among list members
                    Message 9 of 28 , Sep 9, 2006
                    View Source
                    • 0 Attachment
                      Please note I never intended to answer one by one the messages of this
                      thread: this is not a discussion "with Milani", but a discussion among
                      list members "about the Milani et al. proposal".

                      Nevertheless I would like to answer to this mail because it asks relevant
                      questions (also to reward the effort by Tommy in avoding name calling :-).

                      On Fri, 8 Sep 2006, Tommy Grav wrote:

                      > Hi Andrea,
                      >
                      > I re-read your proposal and had a few more questions. How do you
                      > envision that the IAU Data Center handle the submission of least square
                      > solution of orbital computations? It seems to me that adding this to the
                      > things that the IAU DC has to be able to process automatically would
                      > further increase the pressure of the center in a time when they are
                      > already working hard on improving their system. Do you have a specific
                      > plan in mind for implementing this? Also, what is the envisioned policy
                      > of submissions that due to format error can not be processed
                      > automatically?

                      To CHECK a proposed identification, with a good orbit, and to perform
                      standard statistical quality control on the residuals requires of the
                      order of 1 millisecond CPU time on any PC with a standard software. If
                      there is the list of observations but no orbit, a few hundreds of a second
                      should be enough. Thus, if say Pan-STARRS will submit after a lunation of
                      observations and processing 100,000 discovery claims complete of orbits,
                      the MPC will check them in less than 1 hour (CPU time, almost zero human
                      time).

                      On the contrary, if someone else (Pan_STARRS will not do this) submits
                      400,000 ONS and the MPC has to FIND the identifications, the CPU time
                      required is of the order of 1,000 times longer, and still this requires
                      smart mathematical methods, which cost a lot in terms of human time to
                      select the algorithms and write the code. The MPC has neither the
                      computers nor the software to be able to cope with this. (I am not saying
                      they could not do it, but they would need by far more resources). If a
                      small observatory submits 1,000 ONS there is no problem.

                      Thus the proposed rules try to encourage the big surveys to send FULLY
                      PROCESSED data, containing the identifications and the orbits, but they
                      also allow the smaller observers to send only the data with just the
                      indication they belong to the same object (no orbit required) and still
                      they can get full credit.

                      Pan-STARRS has decided, since the first request of funding, to fully
                      process the data, and the same will do LSST and DCT, thus there is no
                      problem expected from the next generation surveys. It would be good if
                      the current surveys were do do some effort in this direction. Indeed, as
                      shown by the examples discussed on this list, already now the MPC takes
                      days to weeks to process the data, does it in an order which we may call
                      illogical and still largerly relies on external orbit computers for
                      identifications (e.g., in her presentation at IAUS236 in Prague Genny
                      Sansaturio said that our group has submitted more than 10,000
                      identifications accepted by the MPC, and Doppler et al. did get even
                      more).

                      As for format error, this is off topic and not my responsability, but I
                      understand Tim Spahr thinks the MPC may in the future provide free
                      software to assist the observers in sending well formed datasets.

                      > Also how do you envision the dessemination of data from the IAU DC to
                      > the public? One of the fundamental flaw, in my mind, of your proposal is
                      > that it is the person that downloads the available data first and has an
                      > automated pipeline to handle the orbital linking and computation that
                      > will win. This means that everyone being in on the game will try to
                      > figure out when the data is being released and have bots getting ready
                      > to download it. This could easily lead to denial of service problems as
                      > the IAU DC servers could be flooded with requests for the newly
                      > published data? Someone could even conceivably create a program that
                      > access the server and starts the download while at the same time denying
                      > access by all others. How do you envision a fair handling of this issue?

                      This is not about the discovery rules but about the operations of the MPC.
                      The MOU between the IAU and the Smithsonian which regulates the MPC from
                      now on states that no secret data will be accepted. I interpret this in
                      the sense that the submitted data will be made public (on the web) as soon
                      as technically possible, should be minutes, even seconds. However, this is
                      not my job, ask Tim. The only thing our proposal asks to the MPC is that
                      the submissions are TIME STAMPED. As the examples shown by Stoss prove,
                      there is no rational way to assign discovery credit, whatever the
                      definition, if the information on WHEN each data have been submitted is
                      not available (as it is not available now).

                      You apparently think that there is a huge (and nasty) competition among
                      orbit computers to submit identifications, but you are wrong. The only
                      time when there was some competition was when, a few years back, both
                      Sansaturio et al. and Doppler et al. were working full steam on this, and
                      indeed there was a race to submit first after each monthly update: they
                      did almost always beat us, because their computer system was by far more
                      efficient :-( . Still we never used hacker attacks against each other!
                      Maybe, if the rules 6 and 7 in our proposal which give priority (not full
                      discovery credit) to orbit computers in some extreme cases (linking of ONS
                      all from different observatories) are approved, there will be more groups
                      in the game, but I doubt there will be a crowd.

                      Both the MPC and NEODyS have been targeted by hacker attacks in the past,
                      and they have to use standard cautions. However, I am not aware of any
                      attack by astronomers willing to sabotage the work of these and other
                      similar web sites. Maybe astronomers are nicer than the average people?
                      Can you believe this? :-)

                      > Also as Mr. (Dr.?) Dixon pointed out, what is the envisioned plan for
                      > handling disputes. What if two people send in an orbital linkage or
                      > computation at the same time. And how good does the fit need to be to be
                      > accepted? How do one handle borderline cases were on fit is just higher
                      > then the accepted fit, while another is just lower?

                      A simple, back of the envelope computation shows that the maximum
                      admissible "dispute rate" when Pan-STARRS is in full operations is of the
                      order of 1/10,000. Otherwise, the people at MPC, Pan-STARRS and in various
                      IAU committees would be working full time to handle discovery credit
                      disputes. Thus the problem must be handled by discovery_credit.exe, a
                      computer program running in batch with extremely rare human intervention
                      by the MPC personnel, who then may consult IAU committees.

                      This implies that it is necessary to use sharp rules, with precise
                      numerical boundaries, even if this may imply messages (from software
                      robots, of course :-) like this: "1 second too late, sorry, already
                      discovered", or " 1.01 arcsec RMS no good, 0.99 good, sorry". Honestly, I
                      see no other way: computer programs can use only binary logic and
                      mathematical formulae, unless you would like unpredictable outcomes.

                      > It seems to me that the rules you propose will lead observers to not
                      > send in their data to the IAU DC in hope that they can do their own
                      > follow-up and compute their own orbit? Is this secrecy of data an
                      > intended consequence? If not how do the community still encourage people
                      > to submit their data quickly under the new rules?
                      >

                      This is the most interesting point, on which we should discuss in depth.

                      The big surveys MUST do their own follow up for 90% or more of their own
                      discoveries, also because nobody else could do this on such a large scale.
                      If they are doing the full processing, this implies that they cannot
                      submit night by night, but only as soon as the automatic follow up is over
                      (e.g., at the end of the observing nights dedicated to moving objects in
                      one lunation) and the computations are concluded (between a few hours to
                      and couple of days later). Large surveys can afford the risk of losing
                      some discoveries credited to observers submitting before.

                      On the other hand, surveys should be encouraged to schedule their self
                      follow up in such a way that the final product are good orbits, not
                      rubbish designations with two nights like the ones encouraged by the
                      current rules (there are now some 100,000 designations with either no
                      orbit or useless orbits allowing neither recovery nor identification).
                      Thus the minimum requirement should be 3 nights of observations, which
                      correspond to the type 3 arcs of our proposal if the astrometry is very
                      accurate, as it should be with the next generation surveys.

                      In the multi-purpose surveys like Pan-STARRS the people involved in
                      processing the moving objects (like, e.g., you and me) have to argue with
                      the rest of the project that the minimum requirement is 6 images in 3
                      nights per lunation: if the two night rule was still valid the gamma ray
                      bursts people could leave us only 2 nights :-). Luckily, this is not going
                      to happen with Pan-STARRS thanks to Nick Kaiser, who is a cosmologist but
                      has studied seriously the problem of NEO discovery and since the very
                      beginning of the project got to the conclusion that the self-follow up
                      should be such that good orbits are computed. However, this could happen
                      in some other survey.

                      Thus my direct answer to your question is: the proposed rules encourage
                      the observers to submit the data as soon as they are ready, which means
                      after completion of the astrometric reduction for the observatories not
                      doing orbits, and after the orbits have been computed for the ones who do.
                      This because of rule 8: no prior submission, no credit. The rules do not
                      encourage to set up new surveys operating like some of the old ones, just
                      as producers of raw data, and I think this is right.

                      The main exception to rule 8 is rule 11, which was introduced taking into
                      account the discussion at the Division III meeting in Prague. It is
                      possible to "reserve a discovery", by promising to do the follow up later.
                      This can be used only for rare and important cases (because it requires
                      manual processing), e.g. for "dwarf planets" :-). Of course the promised
                      follow up must be done really to later claim full credit. We would very
                      much like to have some feedback about these new rules 11 and 12, but to be
                      frank, if the only answer comes form someone who anyway rejects entirely
                      rules 1-10, this is not very useful.

                      If you think that rule 8 is a "pro-Ortiz" rule and rule 11 is a
                      "pro-Brown" rule, well, you are right, this is how hard we are trying to
                      be balanced and fair to everyone. Just to conclude with a joke (which
                      could become serious, if you want) I have consulted my former student
                      Alessandra La Spina who happens to be the webmaster of the next DPS
                      meeting. I asked whether it would be possible to add to the abstract
                      submission form an fully automatic program asking the author if he/she
                      would like to submit a "discovery reservation" for some of the objects
                      mentioned in the abstract, and upon positive answer send the appropriate
                      form to the MPC. She said she knows how to do this.

                      More serious conclusion: the purpose of this discussion is not to convince
                      me and my coworkers, we are not the ones who will decide. The purpose
                      should be to reach a consensus, if not unanimous at least with a strong
                      majority, on either the rules we have proposed or some modification of
                      them or even on a very different, consistent set of rules. Then this
                      result would have a significant influence on the IAU official decisions.
                      Unfortunately it looks like you are very far from achieving this.

                      The most clear indication we got is that there are a few (but graphomanic :-)
                      members of the list who would like to keep the previous rules. This is
                      simply not possible, because the current rules, even apart being
                      rated unfair by many, are by no means automatic, transparent and
                      objective, and therefore have a "dispute rate" which is orders of
                      magnitude above the level which can be tolerated with the next generation
                      surveys. This is not my opinion, but the one of Tim Spahr, who actually
                      said in his IAUS236 presentation in Prague that he is quite fed up even
                      now with the amount of time he consumes in disputes.

                      I would like add some comments about comet discovery, but I have to go
                      now, will answer on Monday.

                      Yours

                      Andrea Milani

                      ================================================
                      Andrea Milani Comparetti
                      Dipartimento di Matematica
                      Piazzale B. Pontecorvo 5
                      56127 PISA ITALY

                      tel. +39-050-2213254 fax +39-050-2213224
                      cellular phone +39-349-4482751
                      E-mail: milani@...
                      WWW: http://copernico.dm.unipi.it/~milani/
                      ================================================
                    Your message has been successfully submitted and would be delivered to recipients shortly.