Talk:Steam devil/GA1

GA Review

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

I'll take this one on. Reviewer: Chiswick Chap (talk · contribs) 09:35, 10 December 2016 (UTC)

Thanks Chiswick Chap. I'll try to address as much as I can over the weekend, but don't be surprised if my response time becomes slow if it goes over into next week, I've got quite a lot on. SpinningSpark 14:30, 10 December 2016 (UTC)

I'm happy to pass this interesting and informative article now. I note that nom has been working on it on and off since 2009, which may account for its high polish! Chiswick Chap (talk) 18:29, 10 December 2016 (UTC)

Thanks for the kind words, but I have not really been working on it that long. I'd forgotten about it for years and was only prompted into thinking this should really be a GA when it popped up on my watchlist recently. Anyway, again, thanks very much for taking on the review. SpinningSpark 18:45, 10 December 2016 (UTC)

General comments

• A few phrases (e.g. "These phenomena") could be avoided with a little copy-editing.
  • What is wrong with the phrase?

Ah, ok, since you ask, I find it a needless vague phrase, as it's both plural and using a Greek-derived technical word, when "Steam devil" (for instance) would do.

Fixed the specific example. The remaining uses are in the singular and seem appropriate to me, but I have reworded the occurrence in the lead anyway. Not sure how else to generalise out from the specific case.

• I'm not totally convinced by the referencing style, with e.g. 5 citations grouped into ref 2 at the end of a longish paragraph (260 words). Would it not be clearer to have 5 refs here? In fact, since it seems doubtful that all 5 cover every sentence in that Appearance paragraph, it would certainly be clearer to place the 5 separate refs with the statements to which they relate.
  • I would rather have the discussion of referencing style outside the GA review, it is not supposed to be a GA requirement. But in short, I favour this style because it is significantly less restrictive in constructing the prose.

It's less restrictive, sure, but by the same token substantially less verifiable, which is within the GA requirements as you know: every statement must be checked against all 5 citations to determine if it is in fact correct, and few reviewers can be expected to have the time, knowledge and determination to carry out any such procedure. In my understanding of policy, any referencing style is fine as long as it's clear and facilitates verification. So it's here. I trust you as a good and careful editor of many years' experience, but in other hands this would be a most dangerous implement.

It is not less verifiable. If you are claiming that cite bundling is against the GA criteria then we have a real problem. If you don't, then as I said above, this would best be discussed outside the GA review. If you think that the article is not verifiable, then quick-fail it now, because I'm not changing it. I am also stopping responding to this review until that point is answered. The GA requirements do not require the reviewer to check all five citations. If it did require you to 100% check everything then you would have to read all the sources no matter how they were presented. To verify a specific fact you do not need to read them all, just carry on reading till you find one that has the fact. On average, you would have to look at 2.5 sources per check if every citation was bundled in groups of five. But they're not. The average bundle size in this article is 1.47. The expectation value for finding a particular fact is thus less than one. That is, you are most likely to find the fact you are looking for part way through reading the first one you pick. All that is moot anyway, because most of the core references are behind paywalls. I'm guessing that you have not obtained them to read yourself and are thus relying on me to give you the actual wording of anything you want to check. I'm happy to do that, but unbundling the citations wouldn't help that porcess in the least.

@Spinningspark: Gosh. You're right, my beef is with the policy, not the article. Struck the item.

• Note 1 raises quite serious concerns about Bluestein's account. I know this is only a note, so I won't flash any relevant policies about, but we're missing a review article here to critique the differing viewpoints.
  • I appreciate the concern and yes, a discussion of the different viewpoints is missing and would be useful. But the bottom line here is that I have no information on why Bluestein is so different. It is not even certain that he is describing a different phenomenon. Personally, I think it is probable that Bluestein has typically been observing smaller steam devils in his local area than, say, Lyons and Peace on the Great Lakes and that the difference in metrics is down to personal experience. I can't say that in the article of course, but the alternative to having the note is to remove Bluestein from the article altogether. While that would not result in losing a huge amount from the article, I think it serves our readers best to simply note the difference and let them make up their own minds.

Ok, thanks.

Specific comments

• Please link Lake Michigan in lead image caption.
  • Done
• Please link geyser in lead; you might want to link lake and ocean also.
  • Done geyser. Lake and ocean would be WP:OVERLINK to my mind
• I wonder if we shouldn't mention water spout or tornado in the lead.
  • Done
• Please link vortex/vortices in Appearance.
  • Done
• Please gloss "fog streamers": given the redlink, it is not obvious what the difference between those and steam devils themselves, really. Of course a linked definition would be nice too.
  • Done
• Not sure why we need a redlink to "vertex vortices". Guess someone liked the phrase "vertex vortex"... If it's real (as implied by the 2 sources immediately afterwards!) then at least a brief gloss would be useful. GA does not depend on a stub article with a quote or two and the beginnings of a definition, but that would be nice.
  • I've glossed this as best I can, but the preceding sentence already about said it all anyway. It's certainly a real phenomenon. The formation of these hexagonal cells is something that comes up in all sorts of fields besides fluid flow; from superconductors, to magnetic domains, to planetary formation. As an electrical engineer who was taught the mathematics of div, grad and curl ad nauseum in field theory lectures, it seems self-evident that vortices can/will form at the junction of the cells. I'm even told one can get cells to form in a layer of oil in a frying pan if heated slowly, but my kitchen experiment attempts to achieve this have failed so far. As for the sources, Zurn-Birkhimer et al uses the phrase directly, while Lyons and Peace state descriptively that the vortices form at the junction of the hexagons. In any event, there is certainly an article to be written here so the redlink is justified.

Thanks! I'm all for redlinks, they encourage growth.

• "horizontal steam devils". Since hot gases rise, how is that possible? A word or two of explanation seem to be necessary here.
  • The source just states it as a fact without explanation. I would guess that a light wind at the surface forms the devil, but it then passes through a boundary layer into a stronger wind which stretches is out horizontally. But that's just my guess.

Interesting.

• " being produced every hour at the most productive locations" could be better worded. I'd suggest starting a new sentence ("...anywhere. Several steam devils form every hour...").
  • Done
• The category "Tornado" is a bit doubtful here and I don't have any strong opinion on the matter, but perhaps a word of justification would be in order.
  • Removed and tightened another category. Can't really justify it, I didn't categorise the page in the first place.
• The word "will" is used 3 times in Appearance and once in Formation. Personally I'd remove all 4 as unnecessary and perhaps even colloquial, but I recognise that their presence may indicate a variety of English not my own. If the latter, that's fine; if not, I think the article would be better without them.
  • Never thought of this as an WP:ENGVAR issue, but I've removed most of them as superfluous.

I can't prove this, but suspect that Americans of a younger generation use the future with 'will' where I'd favour a definitional present.

• I think we're about done. We could consider linking a few terms in Formation: cumulus, convection, perhaps convection cell, and maybe even boundary layer if that's the right target article.
  • Done all of those except boundary layer. That should be the right article, but it does not deal with fluid-fluid boundary so would be very confusing to link. Linked thermocline instead.

Roughly what I suspected, thanks.