V.S.O.P. talk:About

Moved from article page
This is the old "overview" section: The Variable Star One-shot Project aimed at providing the spectral type of many variable stars of the southern sky, and finding strange objects along the way. This project has been first accepted with and  instruments in La Silla Observatory for ESO period P77 (ID 077.D-0085). For P78, it has been granted more HARPS time (ID 078.D-0245).

The Vision
I am going to use this space as a "sandbox" for the Vision before going public with it on the main "About" page. Please do comment!

The problem statements

 * The technical: Large telescopes operate in service mode, which relies on scheduling assumptions.  Often, no observations will fit the conditions, resulting in lower observing efficiency.
 * The scientific: Stellar variability types are usually assigned on the basis of photometric lightcurve appearances. Often, subsequent spectroscopy reveals that the assigned variability type is wrong.


 * I would say: (technical) Astronomical observatories allocate time by chunks. Either night-scale chunks in case of visitor mode, dedicated service nights, and GTOs. Or by, in case of ESO, one-hour OB chunk for service mode programs, adding the possibility to specify many additional constraints (moon, transparency, seeing...). However, between chunks, there is small holes. Cumulating the little holes ends up with a lot of time. Especially if you relax all weather constraints. "VSOP wants to use your observing holes..."
 * (scientific) same as you. Very few objects would benefit from repeated one-shot observations over the long-term. There are of course some special cases, like HD5980 in the SMC for Wolf-Rayet star guys, or eta Carinae, or beta Pictoris and stuff like that. But no matter what, there are of dedicated interest, not community-wide. Non-stellar objects would not benefit neither from this mode of observations. The other side is obvious: what kind project can be build with extremely simple and simplified observation mode with no weather constraints and can still be useful. The answer is the combination of stars (from what is said above) and spectroscopy (since photometry/imaging is difficult, and weather-dependant). One might of course imagine to define another class of stars that deserve observations. You proposed variables stars because there is a need for that. Go for it.
 * Just to say, this project satisfy your wish of exploring new ways of operating La Silla. There is no other projects like that. We have no dedicated telescopes. We want to use official time (this is a way to ask for a a priori and not a posteriori legitimation of the project). But we make the best of the time usually considered as non-interesting. ESO being so obsessed by efficieny, we give them what they want in a plateau, although they considered "non front research" at the first time...
 * (I'm just speaking loudly, in order to find simply the better words to express the evolving goal of VSOP). -- C édric [[Image:kmail.png|16px]]T alk  05:28, 29 October 2006 (CLST)

Guidelines
I imagine vision statements along the lines of the following (based on the above bullet points):
 * VSOP will be the fastest publishing collaboration in stellar astrophysics for years to come.
 * VSOP will obtain data from all the worlds major observatories; reduce, process, and make the data available - all automatically.
 * Give me more time to think about all the above, but the last statement is very very strong. I like it very much. We must build on that. -- C édric [[Image:kmail.png|16px]]T alk  04:57, 29 October 2006 (CLST)

The VSOP Vision
''this is sketchy, and not a final list. A lot more thought and refinement will have to go into this. Some of them may be combined. All of them should be rephrased.''
 * first spectroscopy
 * spectra of uncertain variables
 * fast pipeline-fueled dataflow.
 * when observed, automatic reduction, upload to server, spectral typing, notification.
 * The VSOP wiki and its organization is a shining example of how to manage large collaborations.
 * provide spectra to everybody/community service
 * assign spectral types and revise variability types
 * generate serendipitous discoveries
 * any researcher is free to use the data in any way he wants.
 * open database of research opportunities.
 * facilitate the follow-up
 * facilitate collaborations
 * involve astronomers across all fields of stellar astrophysics
 * involve all the major observatories
 * resolve all the unstudied variables
 * UV, optical, NIR, MIR, submm, ...

Many of the above points can certainly be combined, or are essentially the same.


 * Ok, let me try to reorganize this. Maybe one way to do it is to compare what we have today and what this could become.
 * Today, we have:
 * * First spectroscopy of about 200 uncertain variable stars of the southern hemisphere. More are coming, slowly.
 * * When observed, the data are locally reduced by dedicated pipeline, and transferred automatically to our data/ directory, and we are notified.
 * * We assign, by hand, star by star, the spectral and variability types of each star observed.
 * * We have a powerful wiki website, where each star has its own page.
 * * We have some robot scripts that can perform large-scale work on the wiki, automatically.


 * Now the real whole sequence we are trying to automatize is this: selection of targets -- creation of OBs+creation of wiki pages -- observations -- local reduction -- transfert to data/ -- notification -- (local? remote?) spectral type assignement -- (local? remote?) variability type assignement. Publishing require us, but at least, if we reach the last point, we have raw meat to discuss. This sequence is (almost) hemisphere- observatory- telescope- and instrument-independent. :-)
 * Our main technical problem so far are: automatic spectral and variability types assignements. In any case, at some point, we will need some manual checking...
 * IMHO, we must stay focused on one simple and useful goal. Trying, today, to imagine VSOP doing anything, accepting data from anywhere, in any wavelength is nice, but irrealistic and out of scope. VSOP will become what it will become. We have so far one goal: providing spectroscopy of many variable stars.
 * Before getting anywhere high, we could imagine first rendering VSOP even more useful by automatizing the logg/Teff/binary status determination. This would make VSOP for a lot of people. And then (only then), we could imagine people fueling their own spectroscopy to be processed by VSOP and made available through the wiki.
 * Bottom line. We must secure the sequence. If it works, we could even imagine a "VSOP Bulletin"...
 * Our strength: We provide reduced data of very good quality.
 * Our (becoming) strength: We provide data for many stars.
 * But I repat, for my POV, the most important is: securing the sequence. I think we cannot succeed at making VSOP bigger, if we cannot at least do one simple thing fast, reliable and working well. :-) -- C édric [[Image:kmail.png|16px]]<font color="#006AD5">T <font color="#46A3FF">alk  12:26, 30 October 2006 (CLST)

The Vision, 2nd try
Below text added by Tdall 04:08, 1 November 2006 (CLST) I like the 'problem statements' as they are, i.e. short and concise! But I agree we should have a general structure of a few bullet points plus some text that goes a bit deeper. So... how about something like this (mixing & editing our texts):


 * Sorry, Thomas, but I don't agree much. It is more a fantasm than a vision. :-) . I added some comments. I just put my black hat for a while.-- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric [[Image:kmail.png|16px]]<font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  16:48, 1 November 2006 (CLST)


 * Black hat, huh... well, the vision is what you read with your red hat on and it makes you go "yes!" -- Tdall 04:21, 2 November 2006 (CLST)

The problems

 * Scientific: Stellar variability types are usually assigned on the basis of photometric lightcurve appearances. Often, subsequent spectroscopy reveals that the assigned variability type is wrong.
 * Operational Large telescopes operate in service mode, which relies on scheduling assumptions.  Often, no observations will fit the conditions, resulting in lower observing efficiency.


 * Scientific: On the order of 500 new variables are identified every year, usually being identified from broadband photometry. However... There are numerous examples of complete misunderstandings of the physical processes responsible for the variability. Usually because of misinterpretations of, or ambiguities in the light curves. Most of such cases are only properly revealed following spectroscopic observations.
 * This must be stronger! We must state that the lack of spectroscopic observation for variability type has a bad impact on stellar studies, or something like that.-- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric [[Image:kmail.png|16px]]<font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  16:48, 1 November 2006 (CLST)
 * Operational: Astronomical observatories allocate time by chunks. Either night-scale chunks in case of visitor/classical observations, or hour-scale chunks in case of service/queue mode, adding the possibility of the PI to specify many additional constraints (moon, transparency, seeing...), and of the observatory to schedule observations to match the changing sky conditions. This approach allows for an increase in overall observing efficiency of these increasingly expensive facilities. However...: Between chunks there are small holes. For some sky conditions there are no suitable observations to execute. Cumulating the little holes and idle periods adds up to a lot of lost time, diminishing the overall efficiency.

VSOP goals
Addressing the above problems, we have formed VSOP - a scientifically broad group, but focused on key goals.
 * VSOP will obtain data from all the worlds major observatories; reduce, process, publish, and make the data available to the public - all automatically.
 * Really? Why all major observatories would do that? VSOP will reduce? We don't rely on local pipeline? Process? Well, yes, ok, this one is ok. Publish? All automatically? Hum... -- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric [[Image:kmail.png|16px]]<font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  16:48, 1 November 2006 (CLST)
 * Yeah, it needs rephrasing... Yess, we rely on local pipelines because all major observatories will have pipelines (even Gemini, comes time...) -- Tdall 04:21, 2 November 2006 (CLST)


 * VSOP will obtain first spectroscopy of all unstudied variable stars in both hemispheres, determining or revising spectral and variability types.
 * This is the first and main goal. And this is large already. And this could be extended to the northern hemisphere of course.-- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric [[Image:kmail.png|16px]]<font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  16:48, 1 November 2006 (CLST)


 * VSOP will become the fastest publishing community around by generating a continous influx of serendipitous discoveries, and by providing the framework for easy collaboration among researchers across all fields of stellar astrophysics.
 * I like this quite a lot. But you must explain "publish". -- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric [[Image:kmail.png|16px]]<font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  16:48, 1 November 2006 (CLST)
 * I think "publish" is pretty obvious. I will try to make it more clear, but maybe I don't understand completely what's unclear to you. -- Tdall 04:21, 2 November 2006 (CLST)
 * * well, for these we still need to make the in-depth descriptions... but I think it goes a long way. It almost speaks for itself ;-) Anyway, the in-depth should focus on the wiki, and the automation of tasks, and the openness of VSOP... just thinking loudly now... If necessary we can add one or two more bullets to the list, but I think we don't really need it. Better to keep it simple!

== The Vision, 3rd try -- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric <font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  17:01, 1 November 2006 (CLST) ==

My try. :-)

The problems

 * Scientific: On the order of 500 new variables stars are identified every year identified from broadband photometry. This feeds the whole stellar research, and we can roughly associate XXX refereed papers directly related to these stars. However, there are numerous examples of complete misunderstandings of the physical processes responsible for the variability. Usually because of ambiguities in the light curves. Spectroscopic observations are the only mean to solve the ambiguity.
 * Operational: Astronomical observatories allocate time by chunks. Either night-scale chunks in case of visitor/classical observations, or hour-scale chunks in case of service/queue mode, adding the possibility of the PI to specify many additional constraints (moon, transparency, seeing...), and of the observatory to schedule observations to match the changing sky conditions. However, in practice, there are many short observing time not filled in because of mismatch with any of the conditions. The cumulated observing time of these holes is large, and is hardly if not impossible to reduce with the current operational philosophy. For the 3 telescopes in La Silla Observatory, we estimate it as XXX hours for a 6-months period. VSOP wants to use these holes.
 * Yes! I like this. I'll incorporate this in the 4th iteration. And I will put the goals up front - bang! in your face - and then let these justifications be secondaries. -- Tdall 04:21, 2 November 2006 (CLST)

VSOP goals
VSOP is made of a scientifically broad and international group, mostly from La Silla Observatory. VSOP is dedicated to resolve the above problem by focusing on the following goals:
 * I would not mention LSO... we're beyond that. We're appealing to everybody here. -- Tdall 04:21, 2 November 2006 (CLST)


 * VSOP will obtain first spectroscopy of all unstudied variable stars in both hemispheres, from the world's professional observatories, determining or revising spectral and variability types.
 * VSOP will engage a long-term and broad effort toward the stellar community by providing reliable scientific results and make public high-quality reduced spectra.
 * VSOP will set-up an automatic processing sequence from observing to web-publication on a wiki-website, comprising automatic spectral and variability type classification.
 * VSOP will generate a continous influx of serendipitous discoveries, and provide the framework for easy collaboration among researchers across all fields of stellar astrophysics.
 * you turned 3 goals into 4, but they became weaker I think. The second one is completely vague and useless, sorry. The first one should go first and it's very strong - absolutely right!  The third one is tech-speak and is not very strong.  We should rather combine into two or three (max) than expand. -- Tdall 04:21, 2 November 2006 (CLST)

The Vision, 4th try -- Tdall 04:54, 2 November 2006 (CLST)
Started as an ESO internal project, VSOP is now a mature and truly international collaboration. The VSOP team has a scientifically broad base, and is focused on the following visionary goals:
 * VSOP will obtain first spectroscopy of all unstudied variable stars in both hemispheres, using the worlds leading observatories.
 * VSOP will process incoming data, determine or revise spectral and variability type, and make the data available to the public - all automatic, all fast.
 * VSOP will become the top publishing stellar community by generating a continuous influx of serendipitous discoveries, and by providing the framework for easy collaboration among researchers across all fields of stellar astrophysics.
 * While I agree quite well with the first two, I am a bit afraid of "top publishing stellar community". These words are not well chosen, and might be misunderstood I guess, or I even did it already. Publishing means papers to me. And publish papers is not fast, it's slow. Stellar community seems to mean an extremely large set of people. Are you just talking about the collaboration? And BTW, why you can say so: that we will become the top publishing stellar community?... This is not a goal, this is a dream. And this is not a scientific goal. I quite understand what you seem to want to say. It seems you want to express what, in your opinion, will be the consequences of the first two points, right? -- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric [[Image:kmail.png|16px]]<font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  12:41, 2 November 2006 (CLST)
 * a vision is a dream you dream with eyes open! But you're right, we need to separate goals from hopes, and focus on the facts. The key here is discoveries! -- Tdall 03:40, 3 November 2006 (CLST)

The motivation for VSOP can be expressed by two main problems; one of scientific nature affecting the whole of stellar research, the other of operational nature, derived from the modus operandi of todays modern observatories.
 * Too long. And you repeat the same thing below actually with other words. Or you must include this into the sentences below.


 * Stellar variability types are usually assigned on the basis of photometric lightcurve appearances. Often, subsequent spectroscopy reveals that the assigned variability type is wrong.  On the order of 500 new variables stars are identified every year from light curves of broadband photometry. This feeds the whole of stellar research, last year alone resulting in XXX refereed papers directly related to these stars. However, there are numerous examples of complete misunderstandings of the physical processes responsible for the variability. Usually because of ambiguities in the light curves, or in the underlying assumptions governing the interpretations. Spectroscopic observations are the only means to provide the missing information and resolve the ambiguity.
 * We say the same thing at least twice here, maybe even three times! -- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric [[Image:kmail.png|16px]]<font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  12:41, 2 November 2006 (CLST)


 * Large telescopes operate in service mode, which relies on scheduling assumptions. Often, no observations will fit the conditions, resulting in lower observing efficiency. Astronomical observatories allocate time by chunks. Either night-scale chunks in case of visitor/classical observations, or hour-scale chunks in case of service/queue mode, adding the possibility of the PI to specify many additional constraints (timing, moon, transparency, seeing...), and of the observatory to schedule observations to match the changing sky conditions. However, in practice, there are many short "holes" of observing time not filled in because of mismatch with any of the constraints. The cumulated observing time of these holes is large, and is hard - if not impossible - to reduce with the current operational philosophy.  By providing a dedicated backup, or "filler", programme, ready for all conditions and with full sky coverage, VSOP wants to turn these holes of lost time into productive science time.
 * Very good last sentence. Will think more. -- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric [[Image:kmail.png|16px]]<font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  12:41, 2 November 2006 (CLST)

Our goals
Started as an ESO internal project, VSOP is now a mature and truly international, open collaboration. The team has a scientifically broad base, and is always welcoming new members. The VSOP team is focused on the following visionary goals:
 * VSOP will obtain the first spectroscopy of all unstudied variable stars in both hemispheres, using the worlds leading observatories.
 * VSOP will process incoming data, determine or revise spectral and variability type, and make the data available to the public - all automatic, all fast.
 * VSOP will cause a continuous influx of serendipitous discoveries, and will provide the framework for easy collaboration among researchers across all fields of stellar astrophysics.

The problems
VSOP is motivated by scientific challenges, as well as by the modus operandi of todays modern observatories:

On the order of 500 new variables stars are identified every year from light curves of broadband photometry. This feeds the whole of stellar research, last year alone resulting in XXX refereed papers directly related to these stars. However, there are numerous examples of complete misunderstandings of the physical processes responsible for the variability. Usually because of ambiguities in the light curves, or in the underlying assumptions governing the interpretations. Spectroscopic observations are the only means we have to identify such problems, that are usually well hidden. Moreover, spectroscopy is the only way to provide the missing information and resolve the ambiguity.

Astronomical observatories allocate time by chunks. Either night-scale chunks in case of visitor/classical observations, or hour-scale chunks in case of service/queue mode, adding the possibility of the PI to specify many additional constraints (timing, moon, transparency, seeing...), and of the observatory to schedule observations to match the changing sky conditions. However, in practice, there are many short "holes" of observing time not filled in because of mismatch with any of the constraints. The cumulated observing time of these holes is large, and is hard - if not impossible - to reduce with the current operational philosophy. By providing a dedicated backup, or "filler", programme, ready for all conditions and with full sky coverage, VSOP wants to turn these holes of lost time into productive science time.

== The Vision, 6th try... -- <font style="font-size: 100%;"><font color="#006AD5">C <font color="#46A3FF">édric <font style="font-size: 80%;"><font color="#006AD5">T <font color="#46A3FF">alk  04:47, 3 November 2006 (CLST) ==

Our goals
Started as an ESO internal project, VSOP is now a mature and truly international, open collaboration. The team has a scientifically broad expertise ranging from brown dwarfs to Wolf-Rayet stars, and is continuously welcoming new members. The VSOP team is focused on the following visionary goals:
 * VSOP will obtain the first spectroscopy of all unstudied variable stars in both hemispheres, using the worlds leading observatories.
 * VSOP will process incoming data, determine or revise spectral and variability type, and make the data available to the public - all automatic, all fast.
 * VSOP will cause a continuous influx of serendipitous discoveries, and will provide the framework for easy collaboration among researchers across all fields of stellar astrophysics.

The problems
VSOP is motivated by scientific challenges, as well as by the modus operandi of todays modern observatories:


 * On the order of 500 new variables stars are identified every year from light curves of broadband photometry. This feeds the whole of stellar research, last year alone resulting in XXX refereed papers directly related to these stars. However, there are numerous examples of complete misunderstandings of the physical processes responsible for the variability, caused by ambiguities in the light curves, or in the underlying assumptions governing the interpretation. These numerous "false tracks" are resolved by one-shot spectroscopy, which provide, in a single exposure, the necessary physical parameters: gravity, temperature, binary status and so on.


 * Astronomical observatories allocate time by relatively large chunks. Either night-scale chunks in case of visitor/classical observations, or hour-scale chunks in case of service/queue mode. On top of the allocated observing time, there is the possibility for the PI to specify many additional constraints (timing, moon, transparency, seeing...), and of the observatory to schedule observations to match the changing sky conditions. However, in practice, there are many short "holes" of observing time not filled in because of mismatch with any of the constraints. The cumulated observing time of these holes is large (~100 hours for a 6-months period), and is hard - if not impossible - to reduce with the current operational philosophy. By providing a dedicated and flexible all-sky any-weather programme, VSOP wants to turn these holes of lost time into productive science time.