Jump to content

英文维基 | 中文维基 | 日文维基 | 草榴社区

Talk:1957 Sunfield tornado

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Accuracy issues

[edit]

According to the event report, the tornado did not pass through Sunfield; the intersection in question is a mile north of the town, and the path as noted is almost at right angles to the direction it would have needed to go reach the town. The report in question refers to damage at the intersection, not in the town itself, as the new article also makes clear. Mangoe (talk) 03:40, 14 April 2024 (UTC)[reply]

(Copied from Wikipedia:Articles for deletion/1957 Sunfield tornado for talk page discussion consistency) – The entire NOAA report for the tornado can be seen on Wikisource (Wikisource: NCDC Climatological Data National Summary for the 1957 Sunfield tornado) and it clearly states, "Occurred at junction of highways 51 and 154. Small crossroads settlement at Sunfield "Y" wiped out. Very heavy destruction in small area. Several survisors took cover in buildings. Man remaining in open killed. Tornado moved east-northeastward." I believe this is referring to the storm event database source. Yeah, don't use the map. The map is not a representation of the tornado track. It just draws a straight line from the start to the end of the track. The 2013 El Reno tornado is a very good example of that straight line path drawn for those maps. NOAA produced an actual map of the tornadoes track (an image in the Wikipedia article), but if you look at the Storm Event Database report for it, it just draws the straight line. NOAA also says this just above the maps: "Note: The tornado track is approximate based on the beginning (B) and ending (E) locations. The actual tornado path may differ from a straight line."
Also just a side note, you can take a look at User:WeatherWriter/LLM Experiment 1 and User talk:WeatherWriter/LLM Experiment 1#Follow up with 32k version of GPT4, where myself and another editor actually used A.I. to basicaly fact-check and check the verifiability of the article. Both of us came to the same overall conclusion of it being verifiable and accurate based on the sources. The Weather Event Writer (Talk Page) 04:04, 14 April 2024 (UTC)[reply]
Here's the problem: regardless of anything else, the statement that "The east-northeastward moving tornado struck Sunfield at the junction of U.S. Route 51 and Illinois Route 154" is incorrect because Sunfield, Illinois is not at this intersection. It is actually located a mile south of there, and none of the cited sources claims otherwise. If the AI is failing to recognize that Sunfield and "the Sunfield Y" may not be the same place, it is easily verified that the intersection isn't in Sunfield by looking at a map, and likewise, it is implausible to claim that a ENE-travelling storm went through a place that is over 90 degrees in a different direction. Mangoe (talk) 04:41, 14 April 2024 (UTC)[reply]
Sunfield was leveled and relocated. Where Sunfield is now is not where Sunfield was back in 1957. Either way, that would be a WP:SYNTH/WP:OR argument since sources say that was Sunfield at the time of the tornado, including the government and media. But it relocated. There is no accuracy issues in terms of the sources and content. The Weather Event Writer (Talk Page) 05:22, 14 April 2024 (UTC)[reply]
I'm struggling to find the direct source for the relocation, but a reddit post does mention it: [1]. Relocation was to south of that intersection, hence why you are seeing it north of the settlement. Either way, without a source disputing that the settlement, at the time of the tornado, was not at that intersection, it would technically be a SYNTH argument. Wikipedia has to be verifiable rather than original research truth. The Weather Event Writer (Talk Page) 05:25, 14 April 2024 (UTC)[reply]
A 1938 aerial shows the town exactly where it is now. And again, how many times do I have to say this, the sources do not say that Sunfield was destroyed; they say that the structures at the Sunfield Y, the intersection, were destroyed. Mangoe (talk) 17:04, 14 April 2024 (UTC)[reply]
Actually, checking again, online topos from 1926 and 1950 show Sunfield exactly where it is now. The article is wrong. Mangoe (talk) 17:06, 14 April 2024 (UTC)[reply]
If that is the case, then the article is wrong due to reliable sources. WP:VNT comes into play here. Sources, including the U.S. federal government, says the town was hit by the tornado. Whether or not that is true is irrelevant per VNT. The Weather Event Writer (Talk Page) 18:30, 14 April 2024 (UTC)[reply]
Also, just note I am not mad or anything like that. There was a weirdly similar situation that occurred on another weather article (Tornadoes of 2022), where sources stated one thing, but true numbers were different. VNT was held up in a RFC discussion (see WP:VNTIA). This may be, oddly enough, time for a request for comment discussion to see how to handle this. Either we (1) hold up to RS and disregard their factual errors or (2) go with the town not being where the tornado hit and ignore RS saying the town was hit. Just like the RFC that occurred with WP:VNTIA, it may honestly be time to start an RFC to see if WP:VNT or WP:TNV should be upheld in this circumstance. Thoughts about that RFC idea? The Weather Event Writer (Talk Page) 18:35, 14 April 2024 (UTC)[reply]
Personally, I'm going to stick with the principle that "a source has to be accurate to be reliable." But that doesn't matter here, because it seems to me that you have simply misunderstood the sources. They all say that the damage was at the Sunfield Y, which they identify as that particular intersection which is and always has been north of the town. None of them says that the town itself was struck, or that the intersection was in the town. BTW, in searching I discovered that the town had been struck by an F1 in 2020. Mangoe (talk) 22:22, 15 April 2024 (UTC)[reply]