The following is an archived discussion of the DYK nomination of the article below. Please do not modify this page. Subsequent comments should be made on the appropriate discussion page (such as this nomination's talk page, the article's talk page or Wikipedia talk:Did you know), unless there is consensus to re-open the discussion at this page. No further edits should be made to this page.
The result was: promoted by Yoninah (talk) 18:33, 21 September 2019 (UTC)
... that a 2008 plea bargain to settle criminal charges against financier Jeffrey Epstein has been described as a sweetheart deal? Source: A highly advantageous 2008 plea bargain to settle criminal charges against financier Jeffrey Epstein has been called a sweetheart deal by many commentators.[1][2]
Comment: Prose size expanded from 235 bytes to 2089 bytes on September 1, 2019.
5x expanded by MelanieN (talk). Self-nominated at 20:39, 1 September 2019 (UTC).
This article is expanded five-fold and is new enough and long enough. The hook facts are cited inline, the article is neutral and I detected no copyright or plagiarism issues. A QPQ has been done. Cwmhiraeth (talk) 19:53, 7 September 2019 (UTC)
I think we need clarification on what "5X expansion" means, to keep this from being challenged in Prep or Queue, or on the main page.
The article, according to DYK Check tool, is "1877 characters (294 words) "readable prose size""
Under Comment she noted "Prose size expanded from 235 bytes to 2089 bytes on September 1, 2019." MelanieN is using the byte size on the article history.
Her figures of byte size are from the June 24 version of the article, when most of the article had been deleted. DYK Check tool for that date says it was "221 characters (39 words) "readable prose size"", but close enough to the figure she used.
The size on June 23, according to DYK Check tool, was "1190 characters (204 words) "readable prose size".
So, WP:5X says, "Fivefold expansion means at least five times as much prose as the previously existing article – no matter how bad it was ..." Comments, anyone? @Cwmhiraeth:@Yoninah:@BlueMoonset:— Maile (talk) 23:01, 12 September 2019 (UTC)
It really is unclear why the IP deleted text on June 24. Yes, the first paragraph was horribly written, but what was wrong with the second? MelanieN has started her 5x expansion count from after this deletion, but Rule A4 explicitly states: Fivefold expansion is calculated from the previously existing article, no matter how bad it was (copyvios are an exception), no matter whether you kept any of it and no matter if it were up for deletion. This may be a bad surprise. IMO the expansion count should start from the June 23 version with 1187 characters], which was the highest character count the article achieved, and be expanded 5x from there to qualify. Yoninah (talk) 23:14, 12 September 2019 (UTC)
I would says that Feb 28 is the date that should be used as the starting point, as that is the last date when content had last been added to the article, except for the edits on June 23 where it grew and then shrunk on the same day. 5X is to highlight significant growth in an article, so I personally would disregard the edits of June 23 since they had no lasting affect. The larger article (664 characters) was there a lot longer than the post June 23 version (221 characters). So I would say it needs to be at 3320 characters to count for 5x (all per DYK check). MB 23:10, 12 September 2019 (UTC)
@MB: sorry, but I don't understand your reasoning at all. Content is constantly being added and subtracted from pages. In fact, nothing from that February 28 edit remains in MelanieN's updated article. Yoninah (talk) 23:17, 12 September 2019 (UTC)
Yoninah, Well, the Feb 28 version was where the article had developed over 15 years. The biggest problem was that it was entirely unsourced, so someone came along and deleted everything save a basic definition. It stood at the single-sentence definition for a couple of months until MelanieN expanded it. So I think it should be 5X over the unsourced Feb 28 version. The "growth" on June 23 was only there for 10 hours before it was deleted, so I consider that too transitory to be counted. Just my opinion. But aren't we both in agreement that the current size isn't big enough? MB 23:37, 12 September 2019 (UTC)
@MB: yes, we agree about that :) Yoninah (talk) 23:41, 12 September 2019 (UTC)
@Yoninah: Just to be clear. The current size of the article is 1877 characters, which includes expansion by MelanieN. The largest size I found before MelanieN edited, was 1190 characters on June 23. — Maile (talk) 23:30, 12 September 2019 (UTC)
@Maile66: I use JavaScript to calculate article size. I see 1187 characters in that June 23 edit, and 1878 characters as of today. Yoninah (talk) 23:33, 12 September 2019 (UTC)
Close enough. — Maile (talk) 23:36, 12 September 2019 (UTC)
I see this has become controversial. Sorry about that. I used the size it was when I encountered it as an unsourced stub. I was surprised to see that an AfD had closed as “keep,” and discovered it was kept because multiple sources had been presented at the AfD. However, none of those sources were actually added, and it remained an unsourced stub for the next two years, until I happened across it and made it into a proper article. So that’s what happened. It will have to be up to you guys, who understand the rules here, to decide what should be used as the “before” size. If it doesn't qualify, let's just drop it. -- MelanieN (talk) 00:35, 13 September 2019 (UTC)
It's clear to me that MelanieN made a good-faith nomination after working on a tiny unsourced stub to bring it up to start level. To me, the additions of June 23 were clearly problematic, and were only in the article for less than eleven hours before being removed. DYKcheck and similar prose-character-count programs can only count what was there at some point in the past, and not render judgements on how far back is a useful check; for DYK's 5x expansion, as was explained to me back in the day, what matters is what shape the article was in at the time of expansion and a ways prior to the start of said expansion. Two months, plus or minus, is in that gray area.
Before the three IP edits on June 23 and 24, the article was at 664 prose characters; it had been between 659 and 665 characters since the end of 2013. The real question here is whether over two months is enough time to "reset" the state of the article to the post-June 24 level of 221 prose characters. WP:DYKSG#A4 starts, Fivefold expansion is calculated from the previously existing article, and the key here is "previously existing article". It doesn't say "highest ever previous character count". I'm pretty sure one of my mentors back in the day, Orlady, would have said in this case that enough time had passed that 221 would be the governing number rather than 664. (The 1190 figure supplied by DYKcheck is not on my radar; it was a blip rather than a true addition, with the article being stable before and after that eleven-hour series of additions and subtraction.) If 664 is the number, then 3320 becomes the 5x number, requiring an additional 1443 prose characters over the current 1877. But if consensus is for 221, then a 5x expansion is 1105 prose characters, a number superseded by the DYK minimum of 1500. I'm inclined to the latter view in this case. BlueMoonset (talk) 02:03, 13 September 2019 (UTC)
@MelanieN, Cwmhiraeth, Yoninah, BlueMoonset, and Maile66: I have to agree with BlueMoonset and his/her reference to ""previously existing article". It doesn't say "highest ever previous character count". —A sizeable deletion was made on June 23 by an IP editor, for lack of citations apparently, and the article sat there for more than 3 months before MelanieN's first edit on Aug.31. The article prose size at that point was 221 characters. i.e.readable prose only —"Fivefold expansion is calculated from the previously existing article, no matter how bad it was " — By Sept. 1, the date of nomination, the prose size was 1434 characters - and everything was and remains sourced.
— 221 x 5 = 1105. i.e. the required 5 X expansion number, which is far less than 1434. MalanieN clearly expanded the article more than 5X and added sources. The article is good to go IMO. -- Gwillhickers (talk) 21:07, 20 September 2019 (UTC)
@MelanieN, Cwmhiraeth, Yoninah, BlueMoonset, and Gwillhickers: I think we have discussed the finer points here, and can pass this nomination while it is still topical. To debate it further would be How many angels can dance on the head of a pin? IMO we have established we can IAR when necessary, as in this case, and AGF that MelanieN did her job on this. I'm going to retick this for promotion to prep. — Maile (talk) 21:31, 20 September 2019 (UTC)
Thanks Maile, but my contention was established by the rules, not by ignoring any rules, just for the record, and MelanieN clearly did her required work -- no AGF required on that note. -- Gwillhickers (talk) 21:39, 20 September 2019 (UTC)
In my opinion, this is not a case of IAR, but just one of adhering to the rules. I approved this article initially, and I considered then and consider now that it meets the length requirements. The fact that an IP removed some content ten weeks before MalanieN expanded the article is just not relevant. Cwmhiraeth (talk) 05:04, 21 September 2019 (UTC)