Talk:Geodesics on an ellipsoid/GA1

GA Review

Article (edit | visual edit | history) · Article talk (edit | history) · Watch

Reviewer: Turnitinpro (talk · contribs) 01:48, 6 December 2013 (UTC)

Well-written:

• The Lead section is too long. Also the article is too long and appears to be somebody's essay or thesis which is being bunged in here.

Verifiable with no original research:

• Lack of inline citations, the fact that it is written by a primary author and the excessive coverage to geodesics in general, leads one to suspect this article contains original research,.

Broad in its coverage:

• Lack of focus on the topic.
• Can't tell from the article if it is a controversial topic with different points of view.

Neutral:

• Can't say. The topic is only specifically addressed in the last 2 paragraphs

Stable:

• Yes. but perhaps because it only has 1 primary contributor

Illustrated, if possible, by images:

• Lots of images (perhaps more than required). Cant tell how relevant they are to the actual topic. A formatting problem with

the image set at "Geodesic on an oblate ellipsoid (f = 1/50) with α0 = 45°."

SUMMARY:

Needs some work. Have concerns with the verifiabilty aspectsTurnitinpro (talk) 01:48, 6 December 2013 (UTC)

Initial response

Thanks for your review... Some aspects of the article (particularly the length of the lead and the inclusion of pictures of the main contributors) are in response to this earlier peer review, Perhaps it would be good for you to read this to get another perspective on this article.

I'm puzzled by the statement about lack of inline citations and the possibility that the article contains original research. Can you give specific examples? The same goes for your concerns about verifiability, "perhaps more" figures than required, the lack of focus. Thanks! cffk (talk) 05:50, 6 December 2013 (UTC)

As the other reviewer noted, who is your target audience ? A large section of our readers eyes will just glaze over while reading this. Dumb it down. (I know this contradicts what the other reviewer said). Target it at grad students wanting to base a term paper on this - not a PhD thesis. In so far as the NOR goes, the first 2/3rds of this article didn't focus on the topic directly (just building up to it I guess). You could consider shortening those sections and linking to other wikipedia articles which cover it in greater detail. Once we've done this, I'm sure most of my concerns about verifiability would melt away. Same goes for images. Lets try to trim this article by 50% and I'm sure we'll have a winner on our hands.Turnitinpro (talk) 13:14, 6 December 2013 (UTC)

Citation styles As the other reviewer noted, I also dislike speedbump notations. As an example of my concerns, lets take the section "Equations for a geodesic" .. following Bessel 1825. I can't see where you get this section from. Is it all your own work. Is it based on later references. Are these Bessel's words ? and so on. It would be nice to have a list / milestones of 4 or 5 standard texts for a "normal" reader used to APA referencing to compare with.Turnitinpro (talk) 13:27, 6 December 2013 (UTC)

The section in question closely follows the first 3 pages of Bessel's paper. You can check by visiting [1]. Bessel of course wrote in German, so the words on not his. My translation of his paper appeared in Astronomische Nachrichten and you can get a preprint of this at [2]. Both of these sources are listed in the Bessel (1825) reference. The standard graduate level text for this is probably Rapp (1993). I'll include a pointer to this.

Presumably the citation style is not a make-or-break issue. The author-date style has the advantage that the reader doesn't continally have to consult the reference section to see what "[27]" means. cffk (talk) 13:55, 6 December 2013 (UTC)

Wikipedia has a mousehover tooltip for references to deal with "[27]" situations. Seeing the excellent work on this article which has the potential to go much further I suggest we could focus on Length and Readability to address the general readers. WP:AS is a good starting point Articles that cover particularly technical subjects should, in general, be shorter than articles on less technical subjects. While expert readers of such articles may accept complexity and length provided the article is well written, the general reader requires clarity and conciseness. This article could be the image rich landing page for anyone seeking clarity on Geodesics on Ellipsoids (using standard referencing) with most of the heavy lifting and formulae on side articles where alternate citation styles are used.Turnitinpro (talk) 16:00, 6 December 2013 (UTC)

The question of the length

About the length...

I realize that the article is somewhat long (but not, I hope, excessively so). However none of the ways of shortening it are particularly appealing:

• throwing out some material;
• putting some sections into their own articles;
• somehow splitting the article into two.

A great advantage in covering the subject in a single article is that the notation can be kept consistent and the information needed, for example, for the solution of the inverse problem (which entails knowing the differential behavior of geodesics and the envelope of geodesics) is close at hand. In this way, Wikipedia avoids putting slivers of information into silos.

I will mention another couple of points in defense of the current length of the article.

• the article on geodesics is completely ununderstandable to someone who isn't steeped in the notation of differential geometry; I hope that the current article offers a more approachable resource.
• the subject of ellipsoidal geodesics provides a marvelous window into nineteenth century mathematics from Laplace to Poincaré.

I've received several favorable responses to this article. So I think I should move cautiously on any drastic rearrangement. Probably the way to deal with this question is to allow other editors to chime in via the talk page. And of course specific suggestions for changes are always the most helpful.

I'm still not sure what is the hangup with the reference style. Surely this is largely a matter of taste. I believe that the style I am using is APA referencing; so I'm not sure how to parse the remark about "a normal reader used to APA referencing". cffk (talk) 21:24, 6 December 2013 (UTC)

I'll put this as delicately as I can. Wikipedia articles are not themselves either a reliable source or a scholarly work. These articles are merely accurate summations (covering all points of view) of the body of existing scholarly work, and which cannot exclude Wikipedia's general readership. Now irrespective of the present state of other articles, it is this article under review, and we cannot expect that this article shall compensate for purported defects in other articles. In fact such an approach arouses suspicion that your views may not correspond to those of other authors on the larger subject, hence this article. I shall certainly factor all responses by you and from other reviewers, and I encourage them, into my final recommendation.Turnitinpro (talk) 04:46, 7 December 2013 (UTC)

In addition to the above. I've read Geodesic and I don't see the problem you describe (I still have a working knowledge of post-grad level calculus). In fact that article would be a shoo-in for a GA with some minor tweaks. Also [3] indicates that you showcased this article (authored essentially in isolation from 26/July/2013 onwards) as a fork from it ignoring all the alternate geometries and 2D approaches. I really can't say now how WP:NPOV or WP:PROMOTIONAL this article is and why it isn't better integrated with the geodesic article (where a lot of your initial text on early history of geodesics - eg. almost all the introductory paras of "Geodesics on an ellipsoid of revolution" could be transported to).Turnitinpro (talk) 09:37, 7 December 2013 (UTC)

Point-by-point responses

These are responses to the initial review (repeated here for clarity):

Well-written

The Lead section is too long. Also the article is too long and appears to be somebody's essay or thesis which is being bunged in here.

Thanks for the feedback. I believe that the lead follows the criteria in the manual of style. My intention that it contain the main points that someone without a mathematical background might want to know about the subject. However, I'd welcome specific suggestions on improvements. The criteria for "well-written" don't put a restriction on the length of the article. So I'll pass on that for now.

Verifiable with no original research

Lack of inline citations, the fact that it is written by a primary author and the excessive coverage to geodesics in general, leads one to suspect this article contains original research.

Please point out any places where I need to add citations. I think the coverage of general issues concerning geodesics is appropriate background for this article (especially considering that the ellipsoid was the first surface for which geodesics were studied in any depth and it is the application that lead to the name "geodesic"). No, the article does not include original research. I am one of the authors cited, but the cited publication appeared in a peer-reviewed journal. If you have any specific questions on this point, I'll be happy to answer them.

Broad in its coverage

Lack of focus on the topic

I believe that the focus is consistently on geodesics on an ellipsoid with sufficient additional material so that the main subject can be treated properly. Again, if you have specific concerns, I'll try to address them.

Can't tell from the article if it is a controversial topic with different points of view.

You should check the references. The only outright controversy was in 1826 between Bessel and Ivory (documented in a footnote). There are various approaches to solving the geodesic equations (numerical integration, series expansions, evaluation of elliptic integrals), and I think I have adequately discussed these.

Neutral

Can't say. The topic is only specifically addressed in the last 2 paragraphs

The article treats the main approaches to solving for geodesics on an ellipsoid. I believe that the treatment is balanced. I'm sorry, but I'm not clear not what you're referring to in "the last 2 paragraphs"; can you clarify?

Stable

Yes. but perhaps because it only has 1 primary contributor

Fair enough. A more charitable explanation might be that other potential contributors were happy with the job I had done.

Illustrated, if possible, by images

Lots of images (perhaps more than required). Can't tell how relevant they are to the actual topic. A formatting problem with the image set at "Geodesic on an oblate ellipsoid (f = 1/50) with α0 = 45°."

All the figures (not the images of the mathematicians involved) are directly referred to in the article and serve to illustrate some point. They are invaluable to understanding the subject. I hope I've corrected the formatting problem you saw.

SUMMARY

Needs some work. Have concerns with the verifiability aspects.

Let me know how I can help on the issue of verifiability.

Finally, on the questions you raise about the boundary between this article and the one of geodesics, Wikipedia:What the Good article criteria are not says that the good article review process should not be used to settle content disputes. This, coupled with the fact that I'm not equipped to make radical changes to the article on geodesics, suggests that you should assume that the overall scope of the present article is unlikely to change substantially in the near future. cffk (talk) 01:16, 8 December 2013 (UTC)

I'm just listing some broad concerns I have at this stage.

• This article (which you created) has a title identical to the "GeographicLib" site on sourceforge maintained by you [4].
• The "GeographicLib" site is for some algorithms you have developed used in Geodesy. Your claim is that your algorithms are vastly superior to those presently used Vincenty's formulae
• You have attempted to promote your own algorithms/site on wikipedia which met resistance for promotion and were also reverted

[5], [6], [7], [8], [9].

On reading your work "Karney 2011 (Geodesics on an ellipsoid of revolution)" I find that you favor a certain computational approach based on Bessel's methods (originally in German : 1825). The simultaneous (and probably preceding) methods (in the English language) of James Ivory in 1824 and 1826 have been essentially ignored and pushed into a footnote and reference. This is also the case for Pittman's method [10], Horst Knörrer [11], Saito etc. You curiously evade "Riemann" which the main article on Geodesics relies on extensively. As your article does not have in-line references (like "[27]" as is commonly used at Wikipedia), it is impossible to know without considerable research if the article is derived from your own papers (sub-consciously or otherwise) or from a broad set of standard references which cover all aspects of this topic. As your own works are hardly published in peer-reviewed journals, or if they are, are done under an Open Access scheme where apparently authors pay to be published online, I cannot say if your methods are well accepted or genuine. Certainly the fact that your works are not cited as references by other independent editors on Geodesy or Geodesic or Vincenty's formulae etc. arouses suspicion since your own contributions to these pages is mainly to promote your algorithms accessible from the eponymous "GeographicLib" website.

NB: I appreciate your expertise in this topic and I'm sure this article can achieve GA status without content disputes if we cooperate. Turnitinpro (talk) 03:52, 8 December 2013 (UTC)

You mention Ivory, Pittman, and Saito; however I've also omitted any mention du Séjour, Valperga di Caluso, Soldner, Thune, Puissant, Stein (and this list just gets me up to 1830 or so; see this bibliography for more papers). So, of necessity, I've had to restrict the citations on the subject to the most influential.

In the case of Ivory vs Bessel, both authors are essentially using the same framework as Legendre and Oriani. However, Ivory takes the field no further; while Bessel gives a numerical algorithm to enable the direct problem to be solved in a practical manner. Bessel's work was hugely influential, while Ivory's sank without hardly a ripple.

Pittman made quite a stir when he published his paper (see the article by Deakin and Hunter). However, the plain fact is that his Fortran program fails quite frequently. See this thread on the proj.4 mailing list. A notable feature of his method is that the terms in the series are given by a recursion relation (however he was scooped on that score by Levallois & Dupuy in 1952). But this is only really useful if you need to include many terms (e.g., because the flattening is large) and in this case you're better off evaluating the elliptic integrals directly.

As the article notes numerical integration (as used by Saito among others) cannot compete in terms of accuracy and speed with evaluating the series (for f small) or the elliptic integrals (for f large).

Knörrer and Riemann only need to be considered when ellipsoids in higher dimensions are being treated. This is outside the scope of this article (but I note this extension to higher dimensions in the last section).

In reply to

As your own works are hardly published in peer-reviewed journals, or if they are, are done under an Open Access scheme where apparently authors pay to be published online, I cannot say if your methods are well accepted or genuine.

my contributions to this subject are (for the most part) included in Algorithms for geodesics which was published in the Journal of Geodesy. The paper was subject to peer review. After a paper is accepted, the publisher, Springer, offers two choices: (1) the author pays nothing and the reader has to pay, (2) the author (or, in my case, his employer, SRI) pays $3000 and the reader (that would be you!) pays nothing. Google scholar lists this paper as having four citations, so far. None of these question the results of the paper.

I attempt to give proportional weight to my contributions in this article. Nearly all the important concepts were in place by 1900. I've merely adapted them to modern computers in a slightly different (and arguably better) way than Vincenty.

In reply to

As your article does not have in-line references (like "[27]" as is commonly used at Wikipedia)...

This totally confuses me. For example the section "Equations for a geodesic" begins "The following derivation follows that of Bessel (1825)." where "Bessel (1825)" is a link. If you click on that link you are taken to the citation. How is this substantially different from numbered references, "The following derivation follows that of Bessel^[27]" where "[27]" is the link? (The author-date style does provide additional information, so that the reader knows we're talking about an old paper even if he doesn't know when Bessel lived.) If this isn't working for you, then it's likely that others have a similar problem and I'd like to figure out how to fix it. cffk (talk) 12:51, 8 December 2013 (UTC)

Suggested changes

I'm going to suggest a few things which may clarify issues and save this page for a GA.

1. Please assume (ie. take for granted) that everything that a serious reader for this page has to know about Geodesics and Geodesy is covered in those articles (even if they are not).
2. Please delete/trim/excise all cruft about the history of geodesics/geodesy etc which are strictly unconnected with geodesics of an/the ellipsoid. Remove all those images of the 'greats' (for the moment).
3. Please limit yourself to listing the commonly accepted/used solutions (ellipsoids of revolution) for the direct problem and the inverse problems etc (even if as per you they are not used nowadays being replaced by Vincenty/Bowring). Do not use "Karney" anywhere in this article (FOR NOW).
4. Please discuss concisely the oblate and prolate cases (not dismissing them like this "however, the theory applies with minor changes to prolate ellipsoids, a < b, in which case f, e2, and e′2 are negative.."
5. Once this is done we'll see how to tackle the triaxial ellipsoids.
6. Please strive to limit this article within 30,000 bytes.
7. Please see every set of statements/propositions (say every 3 or 4 sentences) has a "reliable source(s)" where they can be directly (ie. if possible the page number/s) verified from. I presently observe huge portions of text which are unsourced for the casual reader, from this, I also suspect opinions are being passed off as facts - which proper sourcing will settle (either way).Turnitinpro (talk) 16:59, 8 December 2013 (UTC)

On point 7, could you provide some examples where additional citations are needed and, in particular, where opinions may be passed off as facts? I will note that this article (with 126 in-line citations and 74 references, most of them primary) is hugely better sourced that the one on geodesics, which you said would be a shoo-in for a GA (0 in-line citations, 9 secondary sources, 1 tertiary source). By the way, did you manage to solve the problem you were having with the in-line citations in the article? cffk (talk) 22:28, 8 December 2013 (UTC)

Dear Charles. Disgraced is an example of an article which achieved GA status a few days back. Please see how almost every 2 or 3rd sentence has a "[27]" citation. This saves everyone involved the hassle of dealing with claims of Verifiability, Original Research or synthesis. Can I suggest that you set up a Sandbox containing this article's content, and I'll do a "rough-cut" / Hack (basically retaining your content) of what the page could look like - ie. if you are disinclined to do so yourself. You can source that version, and everything added by you to it can be then additionally sourced so we can see where its coming from. A serious problem with the sourcing is, as you've pointed out, that the 74 references are mostly primary.Turnitinpro (talk) 15:52, 9 December 2013 (UTC)

OK, I'll evaluate my use of primary vs secondary sources. It take it that this has nothing to do with the (Bessel, 1825) vs [27] style of citation? Please confirm. cffk (talk) 16:32, 9 December 2013 (UTC)

I think we're missing the wood for the trees. There are several issues about Conflict of interest, POV, Promo, OR, synthesis, lack of focus etc. holding up GA status for this article. A cite like (Bessel 1825) will not suffice until a reader can see exactly where the cited source/s approximate the text/claim (no offence, but I'm not going to take your word for it). BTW, GAR is a time bound process.Turnitinpro (talk) 17:55, 9 December 2013 (UTC)

Some of these suggestions seems to go beyond the scope of a GA review. However, let me address them one by one:

1. This appears to be an implausible starting point for Wikipedia because articles change, split, and merge in a pretty chaotic fashion. If, at some point in the future, the articles on geodesics and geodesy have evolved so that there's duplication between them and this article, then it would be fine to figure out how to eliminate the duplication.
2. As presently written, the history of the subject is pretty tightly woven into the development of the theory. Furthermore, it's possible that this helps some readers maintain interest in the subject since we're talking about real people instead of dry old equations. (A good example of this may be Jacobi's letter to Bessel about solving the problem on a triaxial ellipsoid.) However, if you have specific suggestions of what might be cut, I'd be happy to consider them.
3. I'm not sure exactly what you're getting at here. As the article makes clear, the basic theory goes back to the 19th century. However, nearly all software packages use Vincenty's method and some have started using mine. And my method is needed if you need a solution of the inverse method that works for an arbitrary pair of points. Stopping history before Vincenty would be a disservice to the reader.
4. The treatment for oblate and prolate ellipsoid is so close to being the same that I've changed "with minor changes" to "with no changes".
5. Out of scope for a GA review, surely?
6. Out of scope for a GA review, surely?
7. I'm still not quite sure what the issue here is. I state that the exposition in Section 1.1 follows Bessel (1825). Is your concern: (a) that you don't think Bessel's paper contains this exposition, (b) that it might but you want me to pinpoint the exposition more precisely within that paper, (c) that Bessel's work might have been superceded by newer and better methods, so you want to see a secondary reference citing Bessel which derives the same result (I can offer Helmert 1880 and Rapp 1993), or (d) that I may have inserted some of my own synthesis in getting to equations (3) and (4)?

cffk (talk) 20:39, 9 December 2013 (UTC)

1. Your past discussion reveals that you hold contrary opinions about the Geodesics article. You have hardly edited that article, which is indisputably the parent article for this one. It is obvious that you have unilaterally developed this article to show up the others, possibly to split and merge those. The consequence of this approach has been that the conciseness of the prose employed is not met (ie. overly long sentences). Your text is often subjective at places. The Lead text makes claims which are not clear from the body text. You throughout use WP:WEASEL words/phrases.
2. Please appreciate that for a scientific (incl. mathematics) article, the NPOV, Verifiability and NOR criteria cannot be sacrificed by a reviewer. And this means Wikipedia expects proper sourcing from where the casual reader can easily and directly identify the veracity of the claim. See WP:CITE, WP:SCG. I have given you specific suggestions about what can be cut, I have also offered to assist you if you cannot do so on your own (whereas strictly speaking I ought not to involve myself with this article's content to maintain my impartiality). It is clear that you are promoting your own recent approaches, methods and algorithms through this article, instead of writing a summation of the topic. The opening paras of "solution of the inverse problem" are completely unsourced except for a token general reference to Gauss. Use of phrases like "This suggest the following strategy for solving the inverse problem. Assume that the points A and B satisfy ..." show that it is actually you who are solving the problem. There seem to be general acceptance that Rapp is an authority, lets stick to him exclusively for this solution and see what emerges. And be concise.
3. Please see for eg. Rainsford's 1955 paper. Unlike you, he gets straight to the point. Bessel, Lagrange , Euler and the rest are disposed off in a single opening para. Please appreciate also that the GA criteria No.3 affords considerable subjectivity to the reviewer when it comes to assessing the article's length and focus. Considering objectively that this is a niche scientific article whose parent article (developed collaboratively by many experts over a long period, unlike this one which has been allowed to grow in isolation like a garden gone wild) is stable at around 20,000 bytes, I see no good reason why this article should exceed that size. In fact WP:AS indicates it should be considerably less.
4. That might (or might not) be correct for the approach you have adopted. Would it hold, say, for Sodano or Mccaw's iterations also? Does it only hold for a particular range (eg. between 1/100 to 100) and breakdown beyond that. Wikipedia needs authoritative 2ndary sources for such claims.
5. Would you prefer that the treatment of triaxial ellipsoids is deleted entirely?
6. When it comes to length. GA3 criteria affords considerable subjectivity to a reviewer. You are of course free to seek a community reassessment of my final review.
7. Additionally, I observe that all the technical images used in this article are your own work, and recently created along with this article. There is no independent confirmation for their genuineness as these are not used by other editors. This is unlike, say, the main image [12] used in the parent article which has a provenance and usage.Turnitinpro (talk) 03:18, 10 December 2013 (UTC)

Some responses:

1. I realize that I sometimes lapse into the "some xxx do yyy" weaseldom. Please let me know of the worst (and also the not-so-bad) instances.
2. I've added a citation for the inverse problem. I doubt you'll be happy because it's to my paper. Rapp's treatment (which I refer to as Helmert's) doesn't allow for a solution in all cases. Wikipedia readers should be given a method that always works.
3. Rainsford's treatment is only for the in-crowd. He basically springs the series expansions on the reader without any derivation, any indication of where they come from, or any way for the reader to extend them. I'd like to do better for readers of Wikipedia.
4. Yes, these series formulations all involve even powers of e or e', so they work fine when e is pure imaginary. The series solutions assume that the flattening is small (i.e., slightly oblate or slightly prolate). The formulations in terms of elliptic integrals works for any eccentricity (oblate or prolate); however, there will be numerical problems for very eccentric ellipsoids because most of the variation of the latitude will be over a small portion of the ellipsoid.
5. No, sorry. I was just expressing a desire that the review attempt to improve this section without radically altering its scope.

cffk (talk) 20:44, 10 December 2013 (UTC)

In reply to "It is obvious that you have unilaterally developed this article to show up the others, possibly to split and merge those." No, this was not my motivation. I was able to contribute to Wikipedia; I didn't feel competent to add, in any substantial way, to the articles on geodesy or geodesics; furthermore the subject, as I thought it should be covered, did not fall neatly into the current scope of either article; as a result, I started a new article. You might not like the resulting article, but you should not question my motives.

In reply to "Additionally, I observe that all the technical images used in this article are your own work, and recently created along with this article. There is no independent confirmation for their genuineness as these are not used by other editors." There's not a lot of choice in this matter, given that the figures must be free of copyright restrictions. Creating the figures afresh meant that they are all in SVG format, use consistent (and documented) parameters, colors, etc. As far as "genuineness" goes, it might be reasonable to assume that I'm acting in good faith. (Note that I don't hide my real-world identity; any "cooked-up" images would be quickly discovered and reflect badly on me.)

cffk (talk) 19:15, 10 December 2013 (UTC)

Comments by other editors

To CFFK: Would you prefer we expanded out the GAC requirements individually and proceeded formally with all the , etc icons to facilitate comments from other editors ?Turnitinpro (talk) 16:02, 7 December 2013 (UTC)

Yes, this is probably a good idea (cffk).

Do we need to discuss differences in computational approaches like this [13] discussion set (and why you archived it) ?Turnitinpro (talk) 16:46, 7 December 2013 (UTC)

Well you're welcome to delve into this, but this was addressing the narrow (and fairly sterile) question of what is the "best" mean radius to use to approximate the earth. At the time the discussion ended, there was a consensus that the material at http://wiki.gis.com/wiki/index.php/Ellipsoidal_quadratic_mean_radius was (a) original research and (b) either wrong or meaningless. The one hold-out was the author of this gis-wiki page and I left him with the challenge to formulate some definition for "best" that resulted in "his" mean radius giving superior results compared to alternatives. He agreed to do this, but nothing materialized after a month or so and I decided to archive the rather voluminous discussion. cffk (talk) 17:29, 7 December 2013 (UTC)

See the comments by E03267 on the talk page. cffk (talk) 04:32, 11 December 2013 (UTC)

Please don't go "off-wiki".Turnitinpro (talk) 03:34, 12 December 2013 (UTC)

Interim proposal

I propose to the nominator that this GAR is put on hold for 2 weeks so that the nominator can extensively amend his contributions on the lines we have discussed. Important considerations - focus, brevity, no self promotion, no original research / synthesis etc.Turnitinpro (talk) 03:34, 12 December 2013 (UTC)

Here's a counter-proposal. I see two big sticking points which are unlikely to be resolved in the next two weeks. However the situation may be clearer on both fronts with the passage of more time. Therefore, I suggest that you fail the GA review now and that I renominate the article in a year. The two issues are:

1. Focus+brevity. You see want to see an article with a radically narrower scope. I've received feedback from several people, whose opinions I value, that the present scope (which includes some historical background, citing the original sources, discussing how this problem led to other advances, and taking the field up to the present day) is good. I am therefore reluctant to make big changes to the article based on the opinion of one reviewer. I acknowledge that the article is relatively new and that it has not involved other editors to any extent. That is why waiting a year will be a good thing.
2. No self promotion, no original research, no synthesis. These concerns are triggered because I cite my own work and, indeed, the article should receive additional scrutiny because of this. However the cited works fall under the scope of Wikipedia:SELFCITE. My main results were published in a peer reviewed journal and reasonably qualifies me as an "expert" in this area. Other background material and more extensive derivations are given in a paper on arXiv (Wikipedia views the arXiv repository reasonably favorably; see here). In addition Wikipedia:SPS says that "Self-published expert sources may be considered reliable when produced by an established expert on the subject matter, whose work in the relevant field has previously been published by reliable third-party publications." The article contains no original research; all the material appears in the references. I may missed inserting some citations to my own work. I will endeavor to remedy this. The "no synthesis" standard is needed for articles about global warming or Love Story. In a dry mathematical subjects, such as ellipsoidal geodesics, it doesn't make a lot of sense. Secondary sources aren't needed to reconcile conflicting approaches (there are different approaches but no conflicts). I believe that the article uses an appropriate mix of primary and secondary sources. Some alteration in the way a derivation is presented is hardly synthesizing a new result. Again, the passage of a year will make matters (in particular, my status as an "expert") clearer.

As a separate piece of feedback, I am concerned that your review goes against Wikipedia:AGF. You correctly noted that an editor was citing his own work. But it then appears that you assumed I was obviously a charlatan, questioning the reviewing process at the Journal of Geodesy, the "genuineness" of my figures, etc. I have a thick enough skin that I really didn't mind the tone of your view. However, Wikipedia is, unfortunately, getting a reputation of being a hostile to new editors. In my case I believe you could have raised your concerns (and I could have addressed them) in a more neutral (or even positive) way.

I agree that the reviewer seemed to be biting the expert and commend the expert for keeping it cool. Reviewer, it'd be more beneficial if attention were paid to Wikipedia:Expert editors and Wikipedia:Expert retention. Fgnievinski (talk) 23:33, 24 December 2013 (UTC)

I realize that you undertook this review as a volunteer and I appreciate the effort you put in. I am still grappling with the issue of how best to cover technical topics to a broad audience without over-simplifying them so much that the presentation stops being useful to the more mathematically inclined.

cffk (talk) 19:42, 12 December 2013 (UTC)

Your allegations of my lack of good faith / Newbie biting are misconceived. I've just passed an article for GA [14] where the nominator [15] is in school.Turnitinpro (talk) 03:22, 13 December 2013 (UTC)

I found the reviewer's mistaking open access for vanity press particularly offensive. Fgnievinski (talk) 03:03, 25 December 2013 (UTC)

I also cite "Panou (2013)" [16] where your approach / algorithms have been cited, discussed and deprecated. They have also noted that recently Karney 2013 "has attempted to solve" the near antipodal cases.Turnitinpro (talk) 04:57, 13 December 2013 (UTC)

Just to set the record straight: I asked Georgios Panou about this statement in his paper (e-mail dated 2013-03-20) and I received a reply (dated 2013-04-26) acknowledging that my approach "definitely finds an accurate solution for an arbitrary pair of points". cffk (talk) 13:00, 13 December 2013 (UTC)

Final Recommendation

FAIL at request of the nominator. My good faith assessment is that the present article fails virtually every prescribed criteria. This article can be nominated again at any time or by anybody.Turnitinpro (talk) 03:22, 13 December 2013 (UTC)