Wikipedia:Main Page/Errors

To report an error in content currently or imminently on the Main Page, use the appropriate section below.

• Where is the error? An exact quotation of the text in question helps.
• Offer a correction if possible.
• References are helpful, especially when reporting an obscure factual or grammatical error.
• Time zones. The Main Page runs on Coordinated Universal Time (UTC, currently 20:16 on 28 July 2024) and is not adjusted to your local time zone.
• Can you resolve the problem yourself? If the error lies primarily in the content of an article linked from the Main Page, fix the problem there before reporting it here. Text on the Main Page generally defers to the articles with bolded links. Upcoming content on the Main Page is usually only protected from editing beginning 24 hours before its scheduled appearance. Before that period, you can be bold and fix any issues yourself.
• Do not use {{edit fully-protected}} on this page, which will not get a faster response. It is unnecessary, because this page is not protected, and causes display problems because this is not a talk page. (See the bottom of this revision for an example.)
• No chit-chat. Lengthy discussions should be moved to a suitable location elsewhere, such as the talk page of the relevant article or project.
• Respect other editors. Another user wrote the text you want changed, or reported an issue they see in something you wrote. Everyone's goal should be producing the best Main Page possible. The compressed time frame of the Main Page means sometimes action must be taken before there has been time for everyone to comment. Be civil to fellow users.
• Reports are removed when resolved. Once an error has been addressed or determined not to be an error, or the item has been rotated off the Main Page, the report will be removed from this page. Check the revision history for a record of any discussion or action taken; no archives are kept.

Administrators: Clear all reports

Errors in the summary of the featured article

Please do not remove this invisible timestamp. See WT:ERRORS and WP:SUBSCRIBE. - Dank (push to talk) 01:24, 29 September 2022 (UTC)

Errors in "Did you know ..."

Current DYK

• ... that some estimate that maintenance of existing software costs up to nine times as much as creating it in the first place?

This is another blatant weasel – who estimates this? And it's obviously not a "definite fact which is unlikely to change" as there are a range of estimates. The hook discussion started with the figure of "three times" and it appears that such estimates are unreliable because the article repeatedly says that the topic has not been well-studied. Other issues include:

1. The ceiling is obviously a lot higher than 9 and I reckon the sky's the limit. Consider the case of Wikipedia. This was developed on a shoestring using open source software. But now the WMF has a large staff and a huge budget. As I understand it, the staff headcount has gone from half a person to over 700 – that's a factor of about 1,400. And much of the code is still the same legacy stack.
2. Deciding what is development and what is maintenance is often an arbitrary accounting decision. For example, consider Vector 2022. Is that new development or maintenance of the existing system? The article indicates that it would be considered "enhancement" and so classified as maintenance but it's all still a matter of subjective definition.
3. Changes are often incremental and so there's the Ship of Theseus problem. At what point is a system a new creation?
4. The supposed hook fact does not seem to be clearly stated in the article. I've searched for "nine" and "9" and can't find it.
5. The idea that you can generalise in a definite way about such varied activity and systems is inherently suspect. See all models are wrong.

Andrew🐉(talk) 06:28, 28 July 2024 (UTC)

0. "Who estimates this?" Ulziit, Bayarbuyan; Warraich, Zeeshan Akhtar; Gencel, Cigdem; Petersen, Kai. If you look closely, you'll realise that the definite fact is the estimation, not the costs.

1. "I reckon the sky's the limit. Consider the case of Wikipedia." Well, if you happen to note these considerations of yours in a reliable source, we can take them into account. Until then, we prefer to avoid original research.
2. "it's all still a matter of subjective definition" Yes, that's why we leave it to reliable sources to do the research.
3. See above notes on reliable sources and original research.
4. I would suggest reading the article, not just using ctrl-F. You may find the hook fact in the "Software life cycle" section. WP:2+2=4 may be useful.
5. See 3). ~~ AirshipJungleman29 (talk) 11:16, 28 July 2024 (UTC)

   AirshipJungleman29 seems to be referring to this source:

   Ulziit, Bayarbuyan; Warraich, Zeeshan Akhtar; Gencel, Cigdem; Petersen, Kai (2015). "A conceptual framework of challenges and solutions for managing global software maintenance". Journal of Software: Evolution and Process. 27 (10): 763–792. doi:10.1002/smr.1720..

   On the cited page 764, this states

   The maintenance phase is the longest part of software lifecycle and, in most cases, also the most expensive. For the last several decades, the cost of software maintenance is continuously growing. In the 1970s, the costs were around 60%, while in the 1990s and 2000s, the reported costs increased to about 90% and more.

   Note that this says "90% and more" and so the hook is clearly inconsistent by stating "up to nine times". This hook should be pulled as the source does not verify it but instead contradicts it. Andrew🐉(talk) 18:40, 28 July 2024 (UTC)

   i tried to fix this hook in a hurry (the previous one didn't verify at all) and came out with this one, my bad :) how about we change the wording so that the hook matches the source? theleekycauldron (talk • she/her) 19:19, 28 July 2024 (UTC)

   I noticed that theleekycauldron correctly challenged the original hook which was "... that maintenance of existing software is estimated to cost more than three times as much as its development? ". The source for that wasn't quoted in the nomination and failed verification and we have the same problem with the version that's now on the main page. As I noted above, these generalisations are too fuzzy to be presented as definite facts and they are very subject to change as the technology and techniques move fast. Just pull it, please, and we should then start a post mortem. Andrew🐉(talk) 20:14, 28 July 2024 (UTC)

Errors in "On this day"

Today's OTD

The 2005 Provisional Irish Republican Army item looks inconsistent with the linked article, which has 1997 as the end year in the infobox and says the IRA accepted the terms of the Good Friday Agreement in 1998 as a negotiated end to the Northern Ireland conflict. Perhaps the years 1997, 1998 and 2005 should be sorted out for clarity. Brandmeister^talk 11:23, 28 July 2024 (UTC)

For background, the item is consistent with the lead and infobox of the first bold link Provisional Irish Republican Army and the lead of the second link Provisional Irish Republican Army campaign, but 2005 is not mentioned in the infobox of the second link. TSventon (talk) 12:30, 28 July 2024 (UTC)

Errors in the summary of the featured list

Friday's FL

(August 2)

Monday's FL

(July 29, tomorrow)

Errors in the summary of the featured picture

Notice to administrators: When fixing POTD errors, please update the corresponding regular version (i.e. without "protected" in the page title) in addition to the Main Page version linked below.

Any other Main Page errors

Please report any such problems or suggestions for improvement at the General discussion section of Talk:Main Page.