Updating starnames.dat

[richard]

Hi,

I have been testing locally some updates to starnames.dat. They involve running a script to add Gould numbers (at the end of each line), some manual name corrections, and currently I'm starting to go through Morton Wagman's "Lost Stars" to identify star designations that have been omitted.

I'm wondering how this might be included in a future Celestia version. Is this file generated, or is it currently a manually maintained file? Is the proper place for an update as an add on, or with a future update of the main package?

Many thanks,
Richard

[t00fri]

Hi,

I have been testing locally some updates to starnames.dat. They involve running a script to add Gould numbers (at the end of each line), some manual name corrections, and currently I'm starting to go through Morton Wagman's "Lost Stars" to identify star designations that have been omitted.

I'm wondering how this might be included in a future Celestia version. Is this file generated, or is it currently a manually maintained file? Is the proper place for an update as an add on, or with a future update of the main package?

Many thanks,
Richard

Hi,

starnames.dat contains largely popular names of some brighter stars, in order to complement their purely scientific designations in the HIPPARCOS or other professional star catalogs. In my view, lost stars (from Morton Wagman's book) don't seem to fulfil this criterion of popularity by definition.

In other words, once you have added names from the "Lost stars" book to starnames.dat, I am afraid very few users of Celestia will ever have heard of these additional names... So what would be the use of such an "update"? In addition Dr. Wagman is a historian, not a professional astronomer.

In my opinion, such rather elusive star name additions could only make sense, if scientific records about the usage/origin of such "lost" names are made readily available to Celestia users (e.g.by clicking). Of course, you are entirely free to make add-ons about whatever you find interesting.

Fridger

[richard]

Thanks Fridger for your reply,

The majority of my contribution would be adding Gould designations, for many stars that otherwise only have catalogue numbers. The Gould numbers that already exist in starnames.dat (e.g. 82 Eri) would be corrected to their proper format. In addition, there are some non-standard spellings (e.g. Rigel Kentaurus -> Rigil Kentaurus) that I would propose.

The "Lost stars" book, of which there would only be a handful of additions, are not for star names, but for Bayer/Lacaille and Flamsteed designations that have dropped out of common use, primarily because they were too dim to make the cut of the Bright Star Catalogue. These would be meaningful to many astronomers (perhaps even educational?), unlike obscure Chinese names for stars, for example, which I note already have made their way into starnames.dat.

A historian with a specialty in astronomy would seem to be a plus, not a minus, in my view. However, it's not a all-in or all-out question, and these few updates could be discussed by those who are interested.

Richard

[t00fri]

Thanks Fridger for your reply,

The majority of my contribution would be adding Gould designations, for many stars that otherwise only have catalogue numbers. The Gould numbers that already exist in starnames.dat (e.g. 82 Eri) would be corrected to their proper format. In addition, there are some non-standard spellings (e.g. Rigel Kentaurus -> Rigil Kentaurus) that I would propose.

The "Lost stars" book, of which there would only be a handful of additions, are not for star names, but for Bayer/Lacaille and Flamsteed designations that have dropped out of common use, primarily because they were too dim to make the cut of the Bright Star Catalogue. These would be meaningful to many astronomers (perhaps even educational?), unlike obscure Chinese names for stars, for example, which I note already have made their way into starnames.dat.

A historian with a specialty in astronomy would seem to be a plus, not a minus, in my view. However, it's not a all-in or all-out question, and these few updates could be discussed by those who are interested.

Richard

Richard,

OK, that sounds better, notably the completion of the Gould numbers. My strong recommendation would be to use the professional global astronomical database (SIMBAD/Visier),

http://simbad.u-strasbg.fr/simbad/

in batch mode for extracting all available Gould numbers in one go (without manual interference!). I would do this as follows: as input to SIMBAD submit the list of HIP star names asking for alternative names in return in a packed file. Then write a small Perl script that filters out all returned Gould numbers. It's just a few lines of scripting and takes less than ~ 30 minutes altogether.

In Celestia development, we strive to make Celestia's database entirely compatible with SIMBAD. Since Celestia satisfies scientific standards, a proper documentation of any non-standard addition to Celestia's database is required. By making everything compatible with SIMBAD we can save a lot of documentation work and rely on SIMBAD's accepted global reference role.

I have implemented perfect SIMBAD naming compatibility with the 10000+ NGC/IC galaxies, the globulars and the binary star orbits that I have incorporated into the official distribution years ago. Andrew (aka ajtribick) has analogously piped his latest HIP stars through SIMBAD, if I correctly remember. I am sure he will comment anyway on your project.

Fridger

[starguy84]

The only (slight) problem with this scheme is that SIMBAD does not differentiate between Flamsteed and Gould numbers. Your example, 82 G Eri, is in SIMBAD as 82 Eri, and searching for 82 G. Eri returns an error.
http://simbad.u-strasbg.fr/simbad/sim-id?Ident=82+Eri&NbIdent=1&Radius=2&Radius.unit=arcmin&submit=submit+id

The only way to satisfy the correct Gould name AND have Celestia match SIMBAD, would be to send your finished list to SIMBAD, with a suggestion they add the Gould names (so the star is searchable by both 82 Eri and 82 G Eri, because people will still be looking for it under the other name)
http://simbad.u-strasbg.fr/guide/ch01.htx
I've done it for a handful of stars, they seemed to make changes quickly once they know about them.

Actually, though, if you were going to fix it for SIMBAD, you might consider downloading ALL "Flamsteed" names, fixing them as appropriate (sending that list to SIMBAD) and then selecting out the HIP objects for Celestia. But I'm a completeist...

[richard]

I'm a completist too. Ideally, all the Gould numbers that exist should be added, but realistically I don't see the SIMBAD guys doing that, if only because they are busy with other things. I've sent corrections to them before, and single corrections they're quite happy to investigate and correct as required. I already have the list of Gould numbers masquerading as Flamsteed numbers (due, in part, to a cross reference file done some time ago that made this error in several places, and the error has been perpetuated).

The Gould number (contained in Uranometria Argentina) are in Vizier at

http://vizier.u-strasbg.fr/viz-bin/VizieR?-source=V/135

I have reviewed it since it was uploaded and found a small number of corrections, which have been included at http://www.uranometriaargentina.com/

I'm wondering whether uploading my larger set of changes would be more appropriate as an add-on (while I try and push a small number of changes through SIMBAD). That way, those who want to work with these designations can do so.

[t00fri]

Question: what speaks astronomically / astrophysically for a separation of Gould and Flamsteed numbers given that the world leading professional database SIMBAD didn't care either? As I wrote already, starnames.dat serves mainly as a reminder of popular names no matter whether they are Gould or Flamsteed....

Fridger

[richard]

I can't say whether they care or not (they merely copied errors from a cross-id database). Gould numbers are not Flamsteed numbers, and if they are treated the same, you will get ambiguities. It's hard enough just getting the Bayer/Flamsteed designations right (see Hoffleit's paper on trying to unravel the mess). Why not take a role in education, value precision, and clean up the mistakes rather than perpetuate them? Why not aim for a higher goal? Celestia can be whatever you decide to make it. Just where I'm coming from, and why I'm making the proposals I am. If you don't really care about all this stuff, maybe it's not worth the effort in fixing anything up?

Richard

[t00fri]

I can't say whether they care or not (they merely copied errors from a cross-id database). Gould numbers are not Flamsteed numbers, and if they are treated the same, you will get ambiguities. It's hard enough just getting the Bayer/Flamsteed designations right (see Hoffleit's paper on trying to unravel the mess). Why not take a role in education, value precision, and clean up the mistakes rather than perpetuate them? Why not aim for a higher goal? Celestia can be whatever you decide to make it. Just where I'm coming from, and why I'm making the proposals I am. If you don't really care about all this stuff, maybe it's not worth the effort in fixing anything up?

Richard

For me it is of highest priority what naming conventions astronomers / astrophysicists use in their daily professional work. If there was indeed a serious clash of Gould / Flamsteed numbers it would have long been eliminated from SIMBAD. I will certainly not support any modifications of Celestia that make its data base deliberately incompatible with SIMBAD.

Being an active amateur astronomer since childhood I am certainly familiar with 82 Eri but NOT with 82 G Eri.

I bet few people are or want to be...


Fridger

[richard]

I just checked my Millenium and Tirion 2nd ed. atlases. Both say "82 G" . . .

But in the end, we're coming from different perspectives, so I can see there's not much interest in the projects I've been proposing. Fair enough. Before heading off, I'd just indicate some typos & non-standard spellings I've noticed which I'll leave you to consider and act upon as you see fit: Rigel Kentaurus -> Rigil Kentaurus. xi Tuc -> XI Tuc. Beteigeuze. Alnasi. Tsih, Tsze, Tseen, Tsze Tseang, Kaou Pih (are Chinese names 'popular' in an English program?).

SIG Oct sometimes called "Polaris Australis". This could be added.

R

[selden]

richard,

Just because one person doesn't like the things that you want to do doesn't mean that you shouldn't do them. Rather, you should take that as a challenge to show how they are mistaken. An individual's opinions are just that -- one person's opinions. Don't let them discourage you.

[t00fri]

richard,

Just because one person doesn't like the things that you want to do doesn't mean that you shouldn't do them. Rather, you should take that as a challenge to show how they are mistaken. An individual's opinions are just that -- one person's opinion. Don't let them discourage you.

True!

Yet that individual is a longstanding Celestia developer and co-author, besides being a very experienced professional (astro-) physicist. My knowledge and judgements in these matters derive from solid training and decades of research experience (rather than from clicking web pages...)

Above, I have already emphasized to Richard that he's absolutely free to make any kind of add-ons he likes. But when it comes to the inclusion of material into the official Celestia distribution, we have (scientific) standards that I have partly initiated and continuously implemented (galaxies, globulars, binaries,...).

The only request I have is that the official 'starnames.dat' is NOT to be made incompatible with SIMBAD syntax!

Fridger

[t00fri]

I just checked my Millenium and Tirion 2nd ed. atlases. Both say "82 G" . . .

R

Richard,

I definitely think, too, you should continue with your project. Why don't you try hard to convince the people at SIMBAD that you have a point with the distinction of e.g. 82 Eri and 82 G Eri.

As soon as 82 G Eri ceases to produce an ERROR in the SIMBAD database, I am all happy.

Since I (also) was a keen reader of Sky & Telescope for 25 years, I know the story of the above two atlases pretty well. Roger W. Sinnott, the author of the Millenium atlas, was / is a S&T editor and both atlases are S&T publications. So the agreement in the star designations ("82 G") is obvious and no argument, really. Will Tirion is not an astronomer either, but a graphics person. His CV was often presented in S&T for sales reasons. S&T is often read also by astronomers (on weekends ), but it is basically a journal for amateurs.

Fridger

[starguy84]

If there was indeed a serious clash of Gould / Flamsteed numbers it would have long been eliminated from SIMBAD

Well, SIMBAD is not always 'correct' in a pedantic sense... I can cite a few specific problems with it:
1.) GJ numbers greater than 3000 are not real and were invented by CDS (and astronomers use them, despite being asked not to! http://cdsarc.u-strasbg.fr/viz-bin/Cat?V/70A) for the new stars in the Catalog of Nearby Stars, Third Edition (Preliminary). In the actual paper, they are unnumbered. There have been many schemes to deal with this, all of them more or less unsatisfactory*.

2.) NLTT numbers aren't technically real either; Luyten left them unnumbered and someone just went line by line through the list. Again, it's in SIMBAD and everyone uses them, but technically they're not real. Check the browseable version at HEASARC: http://heasarc.gsfc.nasa.gov/W3Browse/star-catalog/nltt.html
Incidentally, I just discovered the NLTT star at 00 03 39.10 -40 12 00.0 is only known as LEHPM 101 in SIMBAD. It's not even there as LP988-96 or NLTT 66. And there's no star at all in SIMBAD at the location of LP824-347 / NLTT 41, or even searchable by those names. [Edit: I'm actually not sure NLTT 41 exists]

3.) Other star catalogs are only partially in there (Liverpool Edinburgh High Proper Motion 2nd paper, and apparently NLTT)

* Method 1: Use an alternate name for the star <- Endorsed by Gliese & Jahrei?, used by everyone who doesn't use Method 2
Method 2: Number the unnumbered objects starting with GJ3001, going line by line <- What SIMBAD, VizieR, Celestia and nearly everyone uses
Method 3: Number the unnumbered objects starting with GJ3001, going system by system
Method 4: Number the unnumbered objects starting with N1 1, going by (I think it was line)
Method 5: Number all the objects starting with NN 1 (I loved seeing THIS one in a paper)
Method 6: Jahrei? & Weilen's CNS3 (revised) catalog at ARICNS (apparently also called CNS4) has FIVE new names in addition to the Gl, GJ and Wo we know: N1 (does not correspond to the unnumbered systems in CNS3p), N2, NV, NH, N3. I hesitate to call that an official solution, but Harmut Jahrei? is apparently using it.

[t00fri]

If there was indeed a serious clash of Gould / Flamsteed numbers it would have long been eliminated from SIMBAD

Well, SIMBAD is not always 'correct' in a pedantic sense...

I completely agree, of course.

Most scientific publications are not entirely correct in a pedantic sense . Yet there is no alternative to SIMBAD. Practically ALL professional astronomical software packages are linked to SIMBAD for data retrieval or identification purposes.

If some inherent incorrectness turns out to be sufficiently severe, it surely will be corrected soon or later by the global research community. And then we'll do the same in Celestia, referring again to SIMBAD for a documentation of the corrections!

Linking Celestia's data to a generally acclaimed database like SIMBAD provides besides a wealth of additional information about celestial objects, extensive documentation through original astronomical literature! Noone would deny that an indispensable aspect of scientific methodology is a concise documentation of the sources used.

A similar philosophy we have incorporated by providing an interface to the professional SPICE kernel files (NASA)
http://naif.jpl.nasa.gov/naif/
http://en.wikibooks.org/wiki/Celestia/SPICE

SPICE kernel files contain different types of data relevant to space missions. SPK files are used to calculate the position of a spacecraft, planet or other body.

As a consequence, Celestia is used also by the leading space agencies NASA and ESA, and further scientific software projects like the SpaceTrajectoryAnalysis (STA) project [ http://sta.estec.esa.int/Space_Trajecto ... /Home.html ].

I hope I have succeeded to elucidate a bit more why such links to the professional astronomy / space dynamics world are indispensable.

Fridger

[starguy84]

Practically ALL professional astronomical software packages are linked to SIMBAD for data retrieval or identification purposes.

If some inherent incorrectness turns out to be sufficiently severe, it surely will be corrected soon or later by the global research community.

I hope I have succeeded to elucidate a bit more why such links to the professional astronomy / space dynamics world are indispensable.

Oh, I know all this. My point was, inaccuracies in SIMBAD should not hold Celestia back. Rather, if someone like Richard finds genuine mistakes also present in SIMBAD, corrections should be made to both SIMBAD and Celestia. That way, the entire community wins.

[t00fri]

My point was, inaccuracies in SIMBAD should not hold Celestia back. Rather, if someone like Richard finds genuine mistakes also present in SIMBAD, corrections should be made to both SIMBAD and Celestia. That way, the entire community wins.

Of course! My saying. Whenever SIMBAD corrects their data we'll follow (using SIMBAD as reference for the corrections.)

So Richard, should really try to convince the people at SIMBAD about his findings.

Fridger

PS: I just checked with the latest version 0.10.5 of Stellarium. 82 Eri is recognized and correctly identified with e Eri. While 82 G Eri is NOT known.

[Hungry4info]

Is there any references to "82 G Eri" in any scientific literature?
(and, is there a 81 G Eri, 81 G Ori, ... etc as well?)

[richard]

On the 2nd question, almost all stars brighter than V=7.1, south of +10 degress, were assigned Gould numbers. So in Eridanus, stars are numbered up to 293. See http://www.uranometriaargentina.com/. 81 G Eri is better known as 16 Eri (Tau 4).

[t00fri]

The Uranometria Argentina catalog of bright southern stars by Gould, 1879, updated to J 2000 coordinates, is archived at the SIMBAD database

http://vizier.u-strasbg.fr/viz-bin/Vizi ... ce=V%2F135

However, unlike many other catalogs, this catalog is NOT included in the lists of star name aliases at SIMBAD. To me this indicates that the staff at SIMBAD had a reason for not including this catalog. It should be easy to find out from them. It seems this updated historical catalog is a relatively recent addition to the SIMBAD database.

Fridger