This topic was discussed at some length on the old discussion lists, apparently without a consensus solution. The problem keeps coming up when a star like "1SWASP J140747.93-394542.6" is announced as a campaign. I really feel that a short, meaningful name should be assigned by somebody at the time it is entered in VSX or placed on an alert or a campaign.
Now, we have discussed names before, but again I would like to suggest a possible solution. This may not be a perfect solution, or loved by everyone, but I feel it is logical and reasonable.
Basically, assign a type identifier, followed by a harvard-like RA/Dec numerical, keeping it under 10 characters in length. For example, this "J1407" star could be named "TTS1407-39". The original long name could be kept as an alias, but this should be the primary name we use to refer to it.
TTS1407-39 seems very reasonable to me. It tells you the type of object and its approximate location at a glance. Its also easy to remember and discuss about, and to type into wherever you need to, without having to track down the long name and copy and paste it.
So, I hereby propose this as a general method of solving the ridiculously long naming issue!
Thanks considering,
Mike LMK
I fully concur with Mike regarding this situation, one that I had also posted about previously on the old Forum and had a discussion with Arne about.
For me, the naming situation is simply too inconvenient to spur me to add any of these long-named new stars to my observing program. The one and only exception I've made was a look at the latest Nova SGR one night after it's discovery...simply because I thought no one else might be looking so soon following it's discovery.
I will be the first to admit that I'm an "old fashioned" visual observer, but I feel that with so many other targets wanting for observations these days, adding an unnecessary difficulty in personally documenting my observations correctly is simply not worth it. The old Harvard Designation system, even with is A,B,C,D suffixes, was far more convenient, logical and easy to follow when recording/reporting one's observations.
If left to the currently evolving state of new nomenclature it will soon become just a totally meaningless hodge-podge of letters and numbers where no one will recognize just what object is being referred to (I'm there already!).
J.Bortle (BRJ
I am in agreement with LMK and BRJ that designations like "1SWASP J140747.93-394542.6" are actually a deterrent to observation and that some sort of simpler temporary designation would be desirable. Personally I would prefer a designation that incorporates the three-letter constellation abbreviation somehow, perhaps patterned after the system by which the 335th and "later" variable star in each constellation is named but using a different initial letter from "V", of course. Perhaps "J", since J is not used in the RR, RS,..AA,...AZ etc system, and starting with some arbitrary number higher than 1 to avoid any confusion with Flamsteed designations. Hence, "J101 CYG", "J102 CYG", etc. or something like that. A constellation name is more informative to me than RA/Dec coordinates as to whether or not I will be able to observe a new object.
I am glad others feel the same way that I do about this issue. One of the weaknesses of a sequential or arbitrary naming system is it doesn't provide much information on the star to someone who isn't familiar with it. "Meaningless name". Thats a problem with the original letter designations (eg. U AQR - what type?) and similarly for the numbered GCVS (eg. V1017 SGR ?), etc. You have to go look up the star to find out even the most basic thing about it (other than its constellation).
The naming I proposed gives you the star type as the prefix (eg. N for a nova, M for a Mira, UGZ for a Z Cam, etc.) followed by the RA/Dec in the simplest way. Such a name gives you a lot of information on the object at a glance. I suppose the constellation could be used instead of coordinates. But the problem with that is many small and obscure constellations are not easy to locate from memory (eg. where is Aps ?) versus RA/Dec everyone can figure out immediately if its visible at a given time or season. And, some constellations are so large and meander about the sky (eg. Hydra, Draco, Eridani) its quite nonspecific for location and observability. Also, you still need some way to differentiate which one in the constellation it is, so again use a sequential numbering with little meaning.
So, thats why I am in favor of the "ID RA/Dec" naming approach.
Mike LMK
Giving yet another name to something doesn't seem like a good solution to me. Campaigns are frequently, perhaps even typically, conducted in conjunction with the professional astronomical community. It seems both logical and less confusing to use the name that is used for the object in the relevent literature rather than creating a new one unique to AAVSO. More names are definitely not better when it comes to astronomical objects and adding another one that is peculiar to an AAVSO campaign or data set is not helpful in controlling the confusion.
Fore example let's look at 1SWASP J140747.93-394542.6, aka
2MASS J14074792-3945427, aka
ASAS J140748-3945.7, aka
GSC 07807-00004, aka
PPMX J140747.9-394542, aka
UCAC2 14835467, and aka
UCAC3 101-141675
How many more aliases do we need, and do we nwant another one besides AUID that is unique to AAVSO? I think it is a bad idea. For the life of me, with so many well established names I don't know why someone had to give it a new 1SWASP designation. Just because they upgraded equipment and expanded the survey we get a another new catalog with another completely new set of names on top of WASP (not to be confused with SWASP)and HAT (which already duplicates some WASP catalog entries), and TrES, and Qatar, and XO and OGLE and Kepler and CoRoT. Frankly if it weren't for the CDS "catalog of catalogs" on line, I would be completely and utterly confused, instead of my just nornally somewhat confused state. I don't think we need to add to the confusion and I don't think we want to have names for things that are peculiar to AAVSO and not in general use, or that don't follow one of the established catalogs. The AUIDs are necessary but shortcut names for convenience are not and, in my opinion, needlessly add to an already confused situation.
I vote for using the name that is given in the literature or the catalog that is the subject of our campaign.
Maybe people are misunderstanding my point. Names, first and foremost, should be a help, not a hindrance. Many bad, long names are worse than one bad, long name, thats for sure. Such as with "J1407" as mentioned. Naming a star after a catalog with a long precise coordinate is a hindrance; difficult for most people to deal with, almost impossible to remember and discuss with others, or jot down manually at the eyepiece.
The AUID is more of an internal database pointer, as far as I am concerned, it doesn't meet the qualifications of a good name at all. Good names need to be selected in a logical and consistent fashion, be descriptive and meaningful, and they need to be relatively SHORT.
My suggestion of the general naming format "TYPE:RA/DEC" (<=10 chars) seems to meet the qualifications for a good name. It is short, easy to remember and discuss with your colleagues and write down, and provides a lot of info on the star without having to look it up. The exact details of this convention can be sorted out (eg. how many chars in the TYPE, how precise to give RA/DEC ?), but I think it is a very good starting point in principle.
To resolve the naming problem completely, I feel this format (or similar one) should be made a universal convention among the entire astronomical community. I know this is very ambitious, but it just needs to be done once and for all. Even if it means doing the work to convert all the existing names to a short good name, its something that will benefit everyone in the long run. And its better to do it NOW, than wait until the problem gets worse and worse with the explosion of new objects being found.
There are precedents for this in other fields too, such as medicine. Until standard coding for medical records and DICOM and HL7 came along for interfacing between all the disparate equipment, it was one heck of a mess. Nothing could talk to anything else and there were ambiguities and confusion between different places and specialties. People were hurt as a result. It took a major and intense effort to convert everything in medicine over to standards, but it was worth it in the end. Everything works together pretty efficiently and safely now.
Some things in life are a big problem due to the haphazard way in which they grew, and its just a tough problem that needs to be faced up to and resolved. BLTN.
Mike LMK
I agree and...disagree.
Long names yes are a mess,boring,difficult to remember,not so direct like CH Cyg or AE Aur that gives you at a glance the idea of where the star is.But I think that a shorter name as suggested TTS1407-39 instead of 1SWASP J140747.93-394542.6 doesn't change so much.It is just another sequence of cold numbers.So I am for taking the long sequence that cames from the satellite or whatever discovered the star.Copy and paste 1SWASP J140747.93-394542.6 or copy and paste TTS1407-39 it is not a big deal.I am sure that nobody will remember easily TTS1407-39 or whatever.This is just my opinion.
Best regards
Stefano 'T5' Padovan.psd
my 2 cents on this question:
An example:
TCP J14250600-5845360 is a suspect object in Centaurus. Is it a nova? Or another variable type? We don't know yet.
So, at moment, let's call it "NSV Cen 2012" (New Suspect Variable in Centaurus 2012)
or
NSO Cen 2012 - New Suspect Object in Centaurus 2012
or
PNV Cen 2012 - Possible Nova in Centaurus 2012
For the first example (NSV Cen 2012), the constellation name shows the diference to previous NSV Catalogue.
NSV 20279, for example, is a suspect variable that I don't have idea where is it now.
But a "NSV Cen 2012" is the first suspect variable in Centaurus that appeared in 2012.
After more data, if it is a nova, ok, its designation can change to "N Cen 2012" or, as time goes by, Vxxxx Cen according GCVS.
Opsss, it appeared another TCP or PNV in Centaurus this year... No problem, we call it NSV Cen 2012 #2, and on...
None of the proposals I've heard so far is general enough to be considered. Mike's concept of having a variable-star type as the prefix requires knowledge of that type, often not possible when stars are first discovered (and some stars change classification as more information is available). A 4-digit RA plus 2-digit Dec is the same as the Harvard Designation. Not only does this not contain sufficient information to uniquely identify a variable (as we found out in the Orion complex, or just about any open cluster), but you would have to somehow make sure that you discriminated between your new names and the HD (which uses 1900 coordinates as well).
Basically, anyone running a survey comes up with their own unique way of identifying stars, even if they are already known by other names. It is a fact of life, and I cannot believe that any person or group has sufficient "clout" to make it change. Everyone has their own opinion, too: coordinates or constellations, variable-star types, precision, running index, etc. So even if you had enough clout, what would you change it to?
You have to look at the big picture. You need a naming convention that works for the list of stars that you observe and that will ALSO work for everyone else. We took such a look when creating the naming convention for VSX: coordinate-based, with a precision about what is needed for ground-based observations. It doesn't work for extragalactic sources. It doesn't work if the discovery is given with imprecise coordinates. Our rule is that, once VSX-named, the object name does not change if the coordinates improve, as you want a unique name to be able to identify which observations go with which object. Other groups, like SDSS, change object names in each new data release as they often completely reprocess their database; something named SDSS Jxx in one paper may not have the same name in later publications. Even the GCVS is fallible - there are a number of stars where the identification is wrong in the GCVS, and when corrected, the identifier points to a different star, or if the correct star is identified but the coordinates are wrong, and might even be moved to a different constellation in a later revision of the catalog.
Names are not perfect. IF we knew every star in our galaxy, and knew its position exactly and its future motion exactly, you might have a chance (but naming 300 billion stars will end up with a very long name). Until then, I think you just have to live with whatever convention is used for the particular object - at least when referring to it in publications, postings, etc.
That doesn't mean that internally - to yourself - you can't use something simple. I often use a truncated name when setting up an observing program. You just need a method of translating that internal name to the more commonly known name when submitting data/talking about an object/etc. VSX has the ability to enter user-specific names for objects, and we'll take under advisement the expansion of this concept to our other tools. Your spreadsheets could easily convert your simple names to the longer names with a mouse-click. Cut-and-paste works really easily if you don't want to use a spreadsheet or other software. Don't be slaves to the long formats, but don't complain too loudly about them either. They are here to stay.
Arne (on his way back from ALCON)
Well, I almost always agree with our Boss, except it seems not on this naming issue. While Arne may think its too impractical to make everyone adopt a universal method, maybe we just need to start within AAVSO, and if its a good system, eventually will be adopted by almost everyone else?
I would like to suggest solutions to each of the shortcomings that Arne mentioned, of the naming system I proposed:
1. Yes, many new vars are unknown type, but at least one can make an educated guess. So, we can assign a first character 'P' to every new such uncertain type (as we already do with PNV, etc.) Once the type is determined beyond a reasonable doubt, the 'P' is removed. So, obviously no type designation can actually begin with a 'P'. The 'P' is reserved solely as a temporary designator.
2. Yes, stars move. But, if we choose an RA/Dec far enough from the nearest integral boundary, it would be practically constant for many generations of observers! And, if proper motion and precession information is known, one can round the designator appropriately, so its furthest away from the next integral "direction" in which the star moves. For example, the new 1SWASP J140747.93-394542.6 could be rounded to either 1407-39, 1408-39, 1407-40, 1408-40 depending on its expected motion.
3. The type designator prefix could be either 1,2 or 3 characters, depending on the existing list of such identifiers, and avoiding ambiguities, etc. Most variables would be easily identified by these common prefixes. Eg. M=mira, E=eclipser, UGW= WZSge type, NA=fast nova, SR=semiregular, etc.
4. Regarding crowded fields, yes the 4+2 designator would likely be insufficient. I am sure we could agree on some solution for these special cases. Possibly: make it a 4+3 or even 4+4, add a suffix letter or letters, etc. Generally though, most variables are spread out fairly uniformly in space that this is not a fatal problem with the method.
I dont think its a major problem to have to change names now and then. People do, streets do, websites do, etc. By choosing the original name intelligently, we can avoid this in 99.99% of the cases.
Also, I don't believe that the policy of allowing everyone to name a star any way they please for entry in all the AAVSO tools is such a great idea. This will result in even more massive confusion. A major reason for one good name is to share that unambiguously with the community.
The naming convention I have proposed (TYPE:RA/Dec,<10 chars if possible) seems pretty good. If anyone can think of a better one that's as short, please post it so we can all evaluate the candidates, and decide on the best one.
Thanks!
Mike LMK
You forget that precession changes the coordinates substantially, so that every 50 years or so astronomers have to change the coordinate system to refer to another equinox. Using the equinox of 1900 (the one used by the Harvard designation), 1SWASP J140747.93-394542.6 would have the designation 1401-39, so it's more than a matter of rounding. You would only have about two generations of observers for wich the designation is of practical use.
Patrick
[quote=wlp]
You forget that precession changes the coordinates substantially, so that every 50 years or so astronomers have to change the coordinate system to refer to another equinox. Using the equinox of 1900 (the one used by the Harvard designation), 1SWASP J140747.93-394542.6 would have the designation 1401-39, so it's more than a matter of rounding. You would only have about two generations of observers for wich the designation is of practical use.
[/quote]
Well, this problem affects the long "exact" coordinate based names too! And stars could "move" across constellation boundaries as well. Any coordinate or position based naming system faces these long term problems due to motion.
Yet, a short position as part of the name has a lot of positive things going for it. One can easily locate it without having to go look it up, and depending on how precise you name it, it can be essentially a unique identifier.
Nothing I have read in this thread has convinced me that the short naming system which I proposed "TYPE:RA/Dec,<10 chars" isn't a good one worthy of adoption. And, nobody has suggested a better one.
Therefore, I feel the best way is for me to go ahead and adopt this system for naming on my own. Hopefully, others who have expressed positive thoughts about it, maybe would like to join me. I can start entering these as aliases in VSX, and hopefully this will allow entering my observations this way via WebObs to happen as well. For me, it will solve the long naming problem.
The first alias will be 1SWASP J140747.93-394542.6 = TTS1408-40
Mike LMK
Hi, Mike,
There have been a lot of replies to your post with different arguments but I have to agree with Brad that there doesn't seem to be consensus about taking up a new different naming system. You can't make up a system on your own and expect that it is used universally without that consensus.
VSX is meant to be used by the wide variable star community, it is not only an in-house tool and the "Other names" field in each Star's Detail Sheet in VSX is meant to show cross-identifiers taken from catalogues and published papers.
Adding more names to stars that already have several doesn't seem to be a good idea.
As Arne said, VSX is now allowing users to add their personal aliases to a given star. At the moment no other application takes advantage of this feature but the ground is settled and the idea is that eventually this will happen and you'll be able to submit observations through WebObs using any personal alias you want.
So be patient and keep copying and pasting for now.
Cheers,
Sebastian
This will be my last post on the subject because I think establishing another independent naming system is a bad idea and I don't want to prolong the discussion.
It has been my experience that names of stars entered into VSX are reviewed and approved and changed if necessary. I don't think we have the situation that anyone can simply assign whatever name he pleases. I am not familiar with all of the VSX procedures but I know Sebastian Otero has reviewed suggestions I have made for VSX listings and has rejected some of them. So there is at least some kind of review and approval procedure. Since VSP uses the VSX data base, the review applies to VSP as well.
Mike you last post seems to bear the prseumption that we are going to adopt some kind of new unique naming system and I don't agree that there is such a presumption.