Talk:Global Positioning System/Archive 7

Archive 1

←

Archive 5

Archive 6

Archive 7

Archive 8

Archive 9

Multidimensional Newton-Raphson for GPS - Can't make heads or tails of it

Do you think we could get an animated graphical demonstration of this process? --  Denelson83  02:05, 23 October 2010 (UTC)

Let us attempt to isolate where some of the confusion lies. This will provide a first step toward providing any needed clarification. Do you find it clear down through equation (1)? If not, can you state clearly what in your opinion is not clear: RHB100 (talk) 21:52, 26 October 2010 (UTC)

Yes, the part ending with equation (1) is clear. --  Denelson83  05:37, 27 October 2010 (UTC)

I would rather see proper sourcing for this subsection. As it is now, it does not need clarification, but sources. Otherwise I'm sure we can safely remove the subsection as classic case of wp:original research. I have tagged the section. DVdm (talk) 22:00, 26 October 2010 (UTC)

References have been added to show the published equations on which this section is based. In view of the statement that it does not need clarification, the templates are no longer needed. RHB100 (talk) 00:43, 27 October 2010 (UTC)

That's a start, but I don't think this is sufficient. When I said that "it does not need clarification, but sources," that means that the entire derivation (i.e. every single equation) needs to be sourced. We don't just need a source with "published equations on which this section is based", together with a source that explains Newton-Raphson. Using both sources together to produce this exposé, is still a classic case of wp:SYNTH, and thus of wp:OR, as clearly these equations do not qualify as trivial calculations (see wp:CALC). So, unless we have a single source that solidly backs the subsection, I think it will have to go.

Now, even if/when it can be properly sourced, i.m.o. this entire subsection does not belong in this article. It should be largely sufficient to write a single sentence saying that N-R can be used to calculate some equations and then point to the (single) source that does so, or that says that it can be done. DVdm (talk) 11:35, 27 October 2010 (UTC)

I suggest you thoroughly study the references provided before making any judgement as to whether they are adequate or not. I see no evidence from your comments above that you have read the references provided. Also I suggest that you take into consideration that your lack of interest in the useful mathematics provided does not imply that all other people have this lack of interest in useful mathematics. Leaving out material that interests a small minority does grave damage. Leaving in material that may not interest a large majority does essentially no damage. RHB100 (talk) 18:49, 27 October 2010 (UTC)

DVdm, you have completely disrupted a discussion I was having with Denelson83. Denelson83 had responded to my attempt to take an organized approach to see how the section could be improved. Then you jump in and say it does not need clarification and now you are saying it does need clarification. RHB100 (talk) 19:03, 27 October 2010 (UTC)

Here is the question: Are the equations from the article literally present in the sources? If they are not, it is very simple: they must go. Whether I am interested in useful mathematics or not, and whether I have disrupted a discussion you were having with Denelson83, is irrelevant. Either the listed equations are in a source, or they are not. I they are not, they should go per elementary policy as spelled out in wp:SYNTH, wp:OR and WP:V. So, please answer the question: are the equations from the article literally present in the sources?

What do other contributors think about this? DVdm (talk) 20:00, 27 October 2010 (UTC)

The fact that you ask the question in the paragraph above indicates that you have not read the source material. This indicates that you are not qualified to make any judgement on the section. Whether you like it or not, you have got to educate yourself so as to overcome your ignorance before you are qualified to do productive work. Your comment that it is very simple is a result of your ignorance. RHB100 (talk) 00:28, 28 October 2010 (UTC)

Indeed, I have no access to the journal, but that is not relevant. That is why I ask you, who does have access: are the equations from the article literally present in the sources? This is a very simple question that you can answer, so please do. DVdm (talk) 06:46, 28 October 2010 (UTC)

This section discusses the NR solution method, which is obviously a suitable method for this application. However another unsourced method "trilateration and one dimensional numerical root finding" is mentioned, of which I seriously doubt if it is ever used in a real GPS device, as it is not extensible to more than four satellites. Perhaps that should go first. −Woodstone (talk) 14:44, 28 October 2010 (UTC)

Hi Wood, sure, if it cannot be sourced, it should be removed, but the fact that it is not or hardly used in a real GPS device, whereas NR is indeed extensively used, is not really relevant to the question whether the NR-exposé should remain here. The policies are clear on this: if the pack of equations does not appear in a reliable source, it is wp:OR and has no place here - Wikipedia is not a textbook (wp:NOT#TEXT). Feel free to tag or even remove the trilateration bit already. DVdm (talk) 17:56, 28 October 2010 (UTC)

I added a reference to Bancroft's algorithm a few months ago. Prior to that time we did not have any readily viewable souce to Bancroft's algorithm. In adding this reference I stated that "Bancroft's method is perhaps the most important method of solving the navigation equations since it involves an algebraic as opposed to numerical method". No repetition of Bancroft method was provided in the GPS Wikipedia article, since the reference provided was an excelleb source, clearly explained the Bancroft method, and I did not think I could provide a better explanation. RHB100 (talk) 21:56, 28 October 2010 (UTC)

Now I couldn't care less if the numerical methods are removed now that we have access to an excelent external document with the Bancroft method. However, these two numerical methods have at least one and possibly two advantages. One is they are somewhat easier to understand. And two they may have applications to surveying although I am not certain about this since I have limited knowledge of surveying applications of GPS. RHB100 (talk) 21:56, 28 October 2010 (UTC)

I disagree with the view of DVdm stated above that anything that is not a literal copy is original research. I think this view is based on a lack of mathematical maturity. RHB100 (talk) 21:56, 28 October 2010 (UTC)

So it is clear that the equations in the section do not appear in the cited sources ("These equations are based on those found in..."), and are therefore a clear cut violation of wp:original research and wp:synth. I will therefore remove them per elementary policy.

RHB100, I have left a note about your personal comments on your talk page.

DVdm (talk) 19:17, 29 October 2010 (UTC)