Talk:Negative resistance/GA1

GA Review

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

Reviewer: Lingzhi (talk · contribs) 13:18, 8 July 2015 (UTC)

• "Timely" is a relative word; this review may take weeks. But I hope I can help.
  • Many assertions are backed by multiple cites. No need for that unless the statements are controversial, or if different sources offer some meaningfully different explanation(s) or insight(s). Thinning these would be particularly helpful. In general, keep the most reliable source (of course)... One option that covers cases where multiple cites are needed (see "Most authors[3][20][21][22][23][45][52][53][54][58][73][87][88]") would be to use your Notes section for a single note that lists the sources [In that particular case, even my untrained eye suspects that some of your sources are not among the most cited on the planet, e.g., "Physics Quick Study Guide for Smartphones and Mobile Devices." Trimming some of these would reduce eyestrain/cognitve load.]

    The reason for the mass of sources seems to be to justify the "most authors" claim. This is not a legitimate way of verifying such a claim and amounts to OR, the proper way of doing so would be to present a source saying most authors do it this way. I have avoided the problem by rewording and trimming down to two of the more authoritative sources. Chua, in particular, has great standing in this field and him alone would be enough for most verification purposes. SpinningSpark 15:27, 8 July 2015 (UTC)

@SpinningSpark: Very good catch on labeling that OR, SS. I slipped up a little; I should have realized that myself. Thank you very much for pointing that out. I will always bear that in mind. • Lingzhi♦(talk) 00:46, 9 July 2015 (UTC)

Edit looks good, SS. You're right about the sources; I was hoping to find a source saying how "most authors" treat negative absolute resistance. I never found one so I just left the list of example sources I had accumulated in the article. Should have taken it out. --Chetvorno^TALK 23:12, 8 July 2015 (UTC)

• Fix (Hull 1920)

Were you objecting to the cryptic phrasing? Rephrased and cited it.--Chetvorno^TALK 01:52, 9 July 2015 (UTC)

Harvard referencing like this is a legitimate form of citation and permitted on Wikipedia, but it should not be mixed in the same article with another style. I am guessing that is the objection here. SpinningSpark 06:32, 9 July 2015 (UTC)

Yes, the prob was mixing, not use of Harvard. But if you do use Harvard, be prepared for marching minions to try to "fix" your "mistake": case in point

Actually, the original text wasn't meant to be Harvard referencing. "...magnetron (Hull, 1920)" was just a shorthand phrase for "...magnetron, invented by Hull in 1920." When I wrote that I didn't realize how much it resembled a Harvard citation. --Chetvorno^TALK 09:53, 9 July 2015 (UTC)

• Many probs with text wrapping when punctuation follows math formulae. I'vebeen putting nowrap templaes around these, but now I wonder if including the punctuation within the <math></math> would be cleaner and similarly effective...?

  Including within the math tags is a popular method as it keeps the typeface and point size consistent. SpinningSpark 15:30, 8 July 2015 (UTC)

  @SpinningSpark: OK, I tried the method we discussed. Results were not wholly as desired. If I use nowrap templates, everything looks as should be. If I place the punctuation inside math tags, e.g. Thus power sources formally have negative static resistance (<math>\scriptstyle R_\text{static}\;<\;0).</math>[20][21][22][45][52][53][54][55], the punctuation no longer wraps, but the multiple cite numbers do. Thinning those cites down to one or at most two would vastly reduce the probability that the cite would fall precisely at the end of a text line and thus would wrap. However, there still would be that small chance that there would be a single stranded [20] at the beginning of a line somewhere... Your call, but I am leaning toward the nowrap templates. • Lingzhi♦(talk) 01:06, 9 July 2015 (UTC)

  Personally, I tend to avoid attaching refs to display formulae, preferring to place them at the end of the text sentence that led into the formula. However, this would not solve the wrapping issue, a line of refs will wrap wherever they are placed. My preference in articles I write is to use cite bundling, although that can have its own problems, particularly with text-source integrity. By the way, you do know that citation style is not a GA criteria don't you? GA does not even require a consistent style. In any case, what certainly should not be done is the removal of references just to make the layout tidier. Preserving the verification is way more important and that actually is a GA criteria. SpinningSpark 14:17, 9 July 2015 (UTC)

  My qualm about the excessive refs isn't mainly about "tidiness" per se, though I may have expressed it that way. It's about standard academic practices (which is, in turn, about verification). It's about offering precise arguments and evidence. If one cite is clearly strong enough to establish a fact, then attaching 3 or 4 or 5 others simply isn't done. Attaching one or more refs from some weak reference(s) following a far stronger one implicitly suggests that the stronger one is in some way insufficient, which in turn casts doubt on the author's ability to distinguish between the references, which in turn casts doubt on everything the author writes. I'm very sorry if this creates an extra bit of work for Chetvorno. I hate to sound like a broken record, but I almost never go a whole week without linking to meta:eventualism. This is one step in a process that requires patience. The goal here is not to grab a GA and move on to the next article. GA is an intermediate step in a process that takes time. The goal is to work this article up to something near-ish academic standards. On a more personal note, there's very little chance I can bring my understanding of the topic domain up to the point where I can pass GA (or anything else) based on its correctness. I am actually hoping to find some domain expert to take a look to decide whether too much or too little has been said about any given subtopic... All I can do is spot-check various formatting things (I still see more than few things that could be improved; I'll try to list them tomorrow) and spot-check a few facts here and there. If that is unacceptable to you, then I can withdraw from reviewing. Let me know. Thanks. • Lingzhi♦(talk) 15:28, 9 July 2015 (UTC)

  I'm fine with you bringing up issues wich are not strictly part of the GA criteria. I'm not even averse to fixing some of them to just improve the article. Where I would have a problem is if you insisted on failing the article for such a shortcoming if we did not get around to fixing it or just declined to do so. It is unfair to demand in a GA review that an inordinate amount of non-required work is done (and reviewing numerous refs to investigate perceived overreferencing is an inordinate amount of work). SpinningSpark 17:48, 9 July 2015 (UTC)

  @Spinningspark: No way would I fail an article based on something not mentioned in the Criteria, but on the other hand, I also could not personally pass one with this issue. This tension could be resolved in any one of 3 ways: 1) We work as much as we can on all other issues, then in the end I put it on Hold with an explicit note that the only qualm I have is one that is not mentioned in the Criteria; 2) I step aside as a reviewer before that happens, or 3) Someone thins the refs. Chetvorno has suggested that at present he is willing to follow the third option. • Lingzhi♦(talk) 01:48, 10 July 2015 (UTC)

  I'm taking this to your talk page. SpinningSpark 09:13, 10 July 2015 (UTC)

  Although I was hoping to keep them, I don't want to give the appearance of povpushing through CITEOVERKILL, so I will cut back the citations. The reasons I used so many were:

1. I thought certain aspects needed more support. Although negative resistance is not controversial in electrical engineering, it is controversial and baffling to laypeople, as forums [1], [2] and the archives of this Talk page [3], [4], [5] show. In particular many people don't believe there is negative absolute resistance, so I thought that area needed to be widely sourced.
2. Extra refs would give the article a longer "shelf life" before new ones have to be found. In my experience, links to high-tech sources are more subject to linkrot and tend to go dead or become unviewable quickly. I'm finding in the 2 years since I wrote this article, many of the refs have already evaporated. With extra refs we could just delete the dead ones as they die.

But I take your point, Lingzhi, lots of refs don't look good. I'll start picking out the ones to keep. --Chetvorno^TALK 20:37, 9 July 2015 (UTC)

The way to protect against link rot is to archive the page with a service like WebCite. It is free and easy to do so. Here's a link to the archiving form [6]. The cite web template has parameters for dealing with the archive copy. See the archive-url=, archive-date=, and dead-url= parameters in the template documentation. SpinningSpark 00:05, 10 July 2015 (UTC)

That looks good; I'll try it. Thanks a lot! Wish I'd known about it two years ago. --Chetvorno^TALK 01:57, 10 July 2015 (UTC)

@Lingzhi, how many refs on a single fact can I get away with? I have a number of primary sources that I want to include for historical reasons, and since primary sources have to be backed up by secondary ones WP:PSTS, it has to be at least two. How about three refs? --Chetvorno^TALK 10:25, 10 July 2015 (UTC)

• "and a device can have negative resistance over only a limited portion of its voltage or current range" Some devices, or all devices? The sentence immediately following this one strongly suggests the latter.• Lingzhi♦(talk) 01:16, 9 July 2015 (UTC)

  Rephrased, see what you think. --Chetvorno^TALK 10:07, 9 July 2015 (UTC)

• "How it works" illustration: "In a positive static resistance, [formula], so v and i have the same sign." It's not clear to me, looking at the illustration, that v and i have the same sign. Could you pls explain the labeling of the illus.? • Lingzhi♦(talk) 02:59, 9 July 2015 (UTC)

  The animation shows positive charges, plusses, entering the top side of the device from the wire (although the moving charges in a circuit are actually electrons which have negative charge, electrical engineering employs the convention that the direction of current is the direction that positive charges would move). On the left, the top wire is labeled with a plus and the bottom with a minus. This indicates the polarity of the voltage v applied. When v has a positive value, the top wire is positive with respect to the bottom. So the animation shows the current entering the positive voltage terminal and exiting the negative voltage terminal. The passive sign convention, in the previous paragraph, specifies that this must be the direction of current in a passive device such as a resistor, in which v and i have the same sign. Maybe you can suggest a better way to say this? --Chetvorno^TALK 10:56, 9 July 2015 (UTC)

  I'm thinking about this. <strikethrough>What does the arrow from negative to positive, with V in the middle, indicate?</strikethrough> • Lingzhi♦(talk) 01:56, 10 July 2015 (UTC)

• "The physical nature of “negative resistance” remained mysterious, but the lack of theoretical understanding did not prevent engineers from using the effect in their daily practice." Hong p. 165. This is interesting. Did you draw out this point? • Lingzhi♦(talk) 04:15, 9 July 2015 (UTC)
• Speaking of Hong, the Hist section seems to skate somewhat close to reproducing Hong's organization. Forex, Hong steps out of straight chrono order to mention Linvill 1953, as does this article. • Lingzhi♦(talk) 04:40, 9 July 2015 (UTC)
• "A tuned circuit connected" This seems to be one long sentence fragment. It doesn't have a matrix clause (missing the main verb). • Lingzhi♦(talk) 12:04, 9 July 2015 (UTC)

  No, its alright. The main clause is "A tuned circuit...will not oscillate,..." --Chetvorno^TALK 10:03, 10 July 2015 (UTC)

• Is it standard/conventional in similar Physics articles to have "Negative resistance devices" section at the very top, before definitions, explanations, etc.? To me that seems odd, but of course if it is standard format, then it is all good too go...• Lingzhi♦(talk) 12:19, 9 July 2015 (UTC)

  No, it isn't standard. It's probably more common to put a list of types of electronic components after the definitions section. The "Negative resistance devices" section could go there. I think I put it where it is just because there is a nice segue from the statement "...and each type of device has its own negative resistance characteristics, specified by its current–voltage curve." at the end of the "devices" section, into the discussion of current-voltage curves in the "Definitions" section. --Chetvorno^TALK 15:59, 9 July 2015 (UTC)

• Dead links • Lingzhi♦(talk) 01:52, 10 July 2015 (UTC)

• I found some online sources for the dead links, since some of them are used repeatedly: forex here and here. I have no idea whether these are illegal "pirate" sources; I suspect they are. • Lingzhi♦(talk) 09:40, 10 July 2015 (UTC)

• All the Bell System journals are available at the Internet Archive. SpinningSpark 10:30, 10 July 2015 (UTC)

  Thanks, I put them into the article. --Chetvorno^TALK 16:55, 12 July 2015 (UTC)

• Are these WP:OR as well? "much erroneous pseudoscientific information[73][74][75][76][77][78] " • Lingzhi♦(talk) 05:04, 10 July 2015 (UTC)

  Yeah. Let me see if I can find an WP:RS for that. --Chetvorno^TALK 05:33, 10 July 2015 (UTC)

  Deleted statement. --Chetvorno^TALK 16:55, 12 July 2015 (UTC)

• Possibly another WP:OR. Are the multiple cites meant to establish that this is "commonly" done? " commonly classified into two types:[9][18][24][65][81][82]" • Lingzhi♦(talk) 06:27, 11 July 2015 (UTC)

": Not really OR in the sense of "likely to be challenged" but probably don't need all those cites. SpinningSpark 07:52, 11 July 2015 (UTC)

@Lingzhi: No, those are legitimate WP:RSs for the statement, as Spinningspark or another electronics guy can tell you. I want to make clear that I didn't give bogus cites in this article to try to disguise OR. I used a lot of cites on statements that were likely to seem paradoxical to laymen, but on other statements like the above I just gave a lot of citations because I happened to have them. The article was originally an experiment in "full citation", and I was going to cut them back later. There were a few cases where I cited a lot of specific examples to support a generalization, as in the "some authors" and "much erroneous pseudoscientific information" statements above. I agree that was OR, I should have removed those statements. But I think we've got most of them out of the article now. --Chetvorno^TALK 13:57, 11 July 2015 (UTC)

Oh, I never thought you were disguising OR or doing any other dishonest sort of thing. I thought you didn't notice that it was OR. That is, I thought you were using a number of cites to show that a number of cites exist, thus buttressing your assertion that something is "commonly" said or done... If you read above, I made exactly that same mistake. The reason I made that mistake is because I have done that little move before, but I did in in actual real-life OR, and I got so used to doing it that I forgot it was OR. If that makes sense. • Lingzhi♦(talk) 14:27, 11 July 2015 (UTC)

Okay, now I get your use of the term "OR". --Chetvorno^TALK 17:02, 11 July 2015 (UTC)

I just read the article. These are my impressions on the issue of multiple citations:

• The existence of multiple citations is not distractive in this article.
• The percentage of statements with more than three citations is small.
• This topic attracts fringe science; multiple citations may head off future edit wars.
• Multiple citations increases the probability that I will actually be able to find one of them.
• If it were my job to review every citation for accuracy, multiple citations would increase my work. I would have to look at each one. After verifying some of the citations, I might begin to see the extra citations as useless redundancy. I might request that the citation list be pruned. Constant314 (talk) 14:58, 11 July 2015 (UTC)

• I can't see a lead. What there is in the "first section" are three paragraphs that do not act as a lead, such to allow "the average reader on the Clapham omnibus" to read that far and then be able to give a vague but well-directed answer to, "What is an <article topic>?"

As a general rule for a good lead, bluelinks in it should be sparse and the lead should be comprehensible without the likely reader having to follow them (anyone seeking negative resistance can be assumed to have some prior concepts of voltage and current, if not resistance itself). The first para here is OK, but the second para is very dense indeed. Do we really need concepts like bistable hysteresis here? As they are applications of negative resistance in some particular non-linear devices, rather than something inherent in the physical principle.

The third para is OK for scope but it doesn't flow well. The "Therefore" at the start of the third sentence doesn't follow obviously from the second. In the whole lead, there's an overlap between negative resistance (graph slope) as a principle and the non-linear aspects of practical devices. These need to stand clearer apart if it's to be readable to new readers.

Andy Dingley (talk) 22:04, 12 July 2015 (UTC)

Thank you, Andy, for the considered critique. Yeah, the lead. I definitely geeked out in the 2nd para. One of my pet peeves on WP is excessively technical, jargon-ridden intros that are incomprehensible to ordinary readers. Now I wrote one myself. It's funny how a person can be blind to his own poor writing. Thanks for catching that. --Chetvorno^TALK 01:54, 13 July 2015 (UTC)

My feeling about hysteresis is that it is an important enough property that it should be in the intro in some form. --Chetvorno^TALK 01:54, 13 July 2015 (UTC)

GA review – see WP:WIAGA for criteria

1. Is it reasonably well written?

   A. Prose is "clear and concise", without copyvios, or spelling and grammar errors:

   this article written at what college level? Freshman? Sophomore? Junior? Read carefully to spot opportunities for down-leveling, i.e. "explain the obvious". See forex my edit to the "How it works" section, where I explained that the rectangle is an electrical component. All such "explain the obvious" details are customarily omitted in college texts beyond 101 level, so it is easy for editors versed in the topic to overlook opportunities to include them in writing.

   B. MoS compliance for lead, layout, words to watch, fiction, and lists:

2. Is it factually accurate and verifiable?

   A. Has an appropriate reference section:

   B. Cites reliable sources, where necessary:

   C. No original research:

   History section, structure and all, may be more or less completely derived from Hong.

3. Is it broad in its coverage?

   A. Major aspects:

   B. Focused (see summary style):

   The article passes the "readable prose" test, but I am not qualified to determine whether subsections go into excessive detail.

4. Is it neutral?

   Fair representation without bias:

5. Is it stable?

   No edit wars, etc:

6. Does it contain images to illustrate the topic?

   A. Images are tagged with their copyright status, and valid fair use rationales are provided for non-free content:

   B. Images are provided if possible and are relevant to the topic, and have suitable captions:

   No alt text for any images, but describing the illustrations would be a challenge. Captions are occasionally completely missing. Text such as "fig. 1–3 above, and table" is not clear, because there are figures in each section and the reader may not make the mental connection to the captions to the appropriate figures in the definition section (may be looking for figures in section currently reading), and more than one table exists (one image appears to have table-like quadrants).

7. Overall:

   Pass or Fail:

Thank you for all the work and an excellent review, Lingzhi. Sorry it didn't go smoother. You have pointed to a lot of ways the article can be improved, which is the ultimate purpose of these reviews. I will work on the refs and the other issues. Best regards. --Chetvorno^TALK 22:01, 12 July 2015 (UTC)

I would just like to add my thanks to Linghzi for taking on a review of a difficult technical article. SpinningSpark 06:30, 13 July 2015 (UTC)