Talk:Heimlich (disambiguation)

On 18 August 2024, it was proposed that this article be moved from Heimlich to Heimlich (disambiguation). The result of the discussion was moved.

Requested move 18 August 2024

The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this discussion.

The result of the move request was: moved. (closed by non-admin page mover) Bensci54 (talk) 16:51, 27 August 2024 (UTC)[reply]

Heimlich → Heimlich (disambiguation) – per WikiNav, 94% of visitors to this page are looking for Abdominal thrusts, thus making it the primary topic for this term. This page should be moved so that Heimlich can redirect to Abdominal thrusts. theleekycauldron (talk • she/her) 07:40, 18 August 2024 (UTC)[reply]

Per WP:DPT, let's not jump to conclusions about outgoing traffic alone. WikiNav for July '24 says there were 238 views of "Heimlich" that month, and we could identify 109 outgoing clickstreams towards the maneuver. That's actually ~45.8%. We could also identify 11 clicks to Henry, the eponym. At that point we obviously get close to the anonymization threshold of 10, so any long tail is hidden.

Fundamentally, I don't think we have a navigation issue that needs fixing here.

The readers looking for the maneuver and the person it was named after are served first, with the top link, and the top link in the people section. It's very straightforward.

I doubt anyone is astonished to see that a foreign term that is hard to spell and pronounce is also ambiguous.

If we short-circuit everyone else to the maneuver, we don't know if that would accomplish much - I don't know that readers who can't be bothered to click the first link in a short list would be likely to read a longer article. Maybe a few? If so, how would we measure success there, how would we know that they did?

Note that there's a lot of variety in our disambiguation statistics with regard to various percentages, I've posted more info at WT:D. --Joy (talk) 10:02, 19 August 2024 (UTC)[reply]

Support per nom. Definite primary redirect. -- Necrothesp (talk) 10:40, 19 August 2024 (UTC)[reply]
Definite based on more gut feeling? :)

I tried a forward-looking argument above. Let me try a backward-looking argument while we're at it, because our stances are often informed by experience:

A look into the page history shows this was a redirect to the person between creation in 2005 and disambiguation in 2012. If it takes us 19 years to notice that something is wrong at a page that is seen by 300 people a month, it just can't be that wrong.

--Joy (talk) 14:35, 19 August 2024 (UTC)[reply]
Support per overwhelming percentage of outgoing views. This is clearly a case where it is pure speculation to attribute too much significance to the incoming views that do not have a corresponding outgoing view. We simply do not know why this happens. older ≠ wiser 15:11, 19 August 2024 (UTC)[reply]
We can feign pure ignorance, but we could also look at some basic data :) Some recent topics where we didn't choose a primary topic and the same statistic shows something like this 45% (based on my poor man's knowledge base at WT:D) include:
- Angles
- ATB
- Dave Hill
Can someone draw some useful parallels or contrasts between these? --Joy (talk) 17:30, 19 August 2024 (UTC)[reply]
And do you think it is a good thing that we are inconveniencing so many persons coming to such disambiguation pages? older ≠ wiser 18:04, 19 August 2024 (UTC)[reply]
It's been a disambiguation for about 150 months, and the viewership pattern seems steady since '15, so if we extrapolate it, it's been like this for about 45,000 views. Who knows how many views the redirect to Henry got in the preceding 85-odd months, too.

If after all that we have this single inquiry, it's not clear that a significant contingent of readers were being really inconvenienced. I appreciate that it takes a non-trivial effort to make a complaint about this, but it still seems like this hasn't been much of an issue for readers.

Also, maybe the length of time it took for someone to notice it's an ambiguous term being shorter than the time it took for someone to notice that maneuver could be a primary topic - is also indicative. --Joy (talk) 20:24, 19 August 2024 (UTC)[reply]
I don't understand what you're saying. It almost seems like nothing matters unless there is some sort of firestorm. I don't think that is how things work. older ≠ wiser 23:44, 19 August 2024 (UTC)[reply]
There needs to be a middle ground between overreacting to a single complaint and underreacting by requiring a firestorm to react. WP:DISCUSSCONSENSUS says that middle ground is discussion, using reasons based in policy, sources, and common sense. Have I not provided the latter? :) --Joy (talk) 09:02, 20 August 2024 (UTC)[reply]
Sure, but then in response to others in discussion you seem to imply no change is needed because no one noticed for some years. That's not much of an argument. older ≠ wiser 10:11, 20 August 2024 (UTC)[reply]
Okay, but that was in response to the specific claim of inconveniencing so many people. I first posted far more general arguments. Can you review them, do they form a coherent whole? --Joy (talk) 17:54, 20 August 2024 (UTC)[reply]
Sorry, I don't see that you actually addressed the question of why it is a good thing to inconvenience so many people coming to the page looking for exactly one thing. older ≠ wiser 21:30, 20 August 2024 (UTC)[reply]
Like I said before, these many people are less than half of all the people coming to the page. Even if they are an apparent plurality in that month's reading of statistics, we have no reason to start assuming that they're a supermajority that needs to be catered to extra. --Joy (talk) 07:49, 21 August 2024 (UTC)[reply]
And as I said, we have no idea whatsoever WHY these views come to the page and do not go on to another page. We do not even know whether these are actual readers or some sort of unidentified web crawling agents. That such views do not go on to view some other page says nothing about what readers are actually looking for. I've said this before, in most cases, the incoming views is mostly irrelevant for considering whether there is a primary topic. older ≠ wiser 10:06, 21 August 2024 (UTC)[reply]
I thought we've been over this already... Yes, this is supposed to be user traffic, as meta:Research:Wikipedia clickstream says:
We attempt to exclude spider traffic by classifying user agents with the ua-parser library and a few additional Wikipedia specific filters.

Finally, any `(referrer, resource)` pair with 10 or fewer observations was removed from the dataset.

The fact that the mapping of incoming traffic to outgoing traffic varies a lot is examined at length at WT:D#on what statistics should look like for hatnotes, primary redirects, primary topics. We have pages where a previously-thought-to-be-primary topic gets single-digit percentage of clicks compared to incoming, and we have places where such a topic gets up to 85% of incoming, and everything in between.

That some incoming views from readers go on to click, and some do not, cannot be said to be statistically insignificant. I do not see the rationale for your generalizing claim that in most cases, the incoming views is mostly irrelevant for considering whether there is a primary topic. It is perfectly conceivable that some readers do not click through for a multitude of reasons, just like it is perfectly conceivable that some click through once, or indeed that some click through multiple times (and we also have a documented example of the latter). The fact that we can't exactly identify which of these reasons apply in which case does little to invalidate these readers, that would not be logical. --Joy (talk) 15:02, 21 August 2024 (UTC)[reply]
We attempt to exclude spider traffic ... emphasis added -- such attempts rely on well-behaved spiders and responsible operators. There is nothing to prevent agents from operating web crawlers that are difficult or impossible to identify as such and there is no shortage of bad actors on the web.

Honestly, I do not see how WT:D#on what statistics should look like for hatnotes, primary redirects, primary topics provides any conclusive evidence one way or another with regards to the discrepancy between incoming and outgoing views.

Perhaps it is true that such discrepancies between incoming and outgoing views cannot be said to be statistically insignificant -- but I have seen nothing to establish any sort of clear guidance about how to interpret such discrepancies. At present it is the sort of useless statistic that anyone can interpret to confirm whatever biases they might have.

In this particular case, the ONLY other item that provides any competition at all for primary topic to Heimlich maneuver is the namesake Henry Heimlich, although by a fairly distant margin. Given that the originator is prominently mentioned early on in the article (and if this move succeeds, could also be added to the hatnote), I see little reason not to move the article. older ≠ wiser 15:37, 21 August 2024 (UTC)[reply]
There's a big difference between a lack of clarity and being useless.

Someone could add bias to the interpretation - they could. Just like they could add bias if something was even less clear, slightly more clear, even more clear, or even if it was perfectly clear! :)

Indeed, that's what we see happening here as well, ironically enough. You're taking the parts of the statistical data that seem clear to you, and making resolute statements like the above. You're reading certainty from a source that you just described as otherwise uselessly uncertain. The fact that nothing in the definition of WikiNav, Clickstreams or Pageviews supports or even allows for such resolute statements - doesn't stop this from happening.

You mentioned a possibility of misidentified bot traffic in the data. Let's think about that for a second then - if that can exist in the 'dark' portion of the data, can it not also exist in the 'light' portion? In other words, how do we know that some of the clicks on the first link are not also bots?

With a data sample of just 109, even just one crawl a day by a single bot could have added 31 clickstreams to that - which would be almost a third of it. Of course, the obvious retort is how likely is it that someone does that.

Just like the obvious retort to that in turn is - how likely is it that all of the remaining 129 views went without clickthroughs? How likely is it that the larger portion of views are safe for us to disregard out of hand?

--Joy (talk) 07:43, 23 August 2024 (UTC)[reply]
No, I'm sorry, any interpretation of the discrepancy between incoming and outgoing views in speculative. We know nothing about why it happens or what reasons readers who do not go on to an outgoing view might have for doing so. Sure, it is possible that some outgoing traffic might also be non-human agents. However, the bot portion of either is minor part of the point. The point once again is that the raw incoming view numbers tell us NOTHING about what readers might have been looking for, other than possibly that A) what they were looking for wasn't on the page or B) that what they found on the page was sufficient; and we can't even tell the difference based on this data. older ≠ wiser 11:20, 23 August 2024 (UTC)[reply]

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.