Talk:Main Page

From Wikisource
Jump to: navigation, search

/Archive0: discussion of the old main page (discussion continued here)
/Archive1: 2005
/Archive2: 2006
/Archive3: 2007
/Archive4: 2008
/Archive5: 2009
/Archive6: 2010–2015

Gujarati Wikisource 10000+[edit]

Kindly update statastics of Gujarati wisi source. [1] --Sushant savla (talk) 02:39, 11 June 2016 (UTC)

Done. Great news! Ankry (talk) 11:29, 11 June 2016 (UTC)

Bengali wikisource 10,640[edit]

Please update statastics of Bengali wikisource. --Aftabuzzaman (talk) 20:43, 22 June 2016 (UTC)

How to make bookmarks in my account[edit]

is there any way to bookmark pages into my account so that i can read them latter or on different devices in which my account in longed in if thee is please tell me or if not then pleas inform me so that i can think of a way to make it happen beaus it would be a nice feature to have in all the sister projects of Wikisource

Updating Tamil Wikisource count[edit]

Tamil Wikisource has reached 409,974 pages per https://stats.wikimedia.org/wikisource/EN/Sitemap.htm. Please update the main page to reflect that. Thanks. -- Sundar (talk) 14:36, 19 March 2017 (UTC)

The Hebrew spelling is wrong, should be ספרייה (not ספריה).[edit]

The word סִפְרִיָּה has a Hiriq at the end, so the word should be spelled with an extra י in Ktiv hasar niqqud. Please change this, thanks! --Itsused (talk) 11:29, 10 June 2017 (UTC)

Done. --Ooswesthoesbes (talk) 13:56, 10 June 2017 (UTC)

Greek wikisource[edit]

Greek wikisource has now more than 10000 pages. --Ntetos (talk) 19:59, 25 August 2017 (UTC)

@Ntetos: According to these stats not yet. And it seems there is no concensus which stats should be used on the maim page. @Zyephyrus, Yann, Ooswesthoesbes, -jkb-: or maybe I missed something? Ankry (talk) 07:46, 31 August 2017 (UTC)
That's remarkable indeed. Also the meta stats clearly indicate it has crossed the line. It often occurs that the local Special:Stats aren't up to date, but mostly in negative way (counting too few pages), so I don't really see what's wrong here. --Ooswesthoesbes (talk) 09:49, 31 August 2017 (UTC)
In the revert by Zyephyrus that I pointed above, he reverted a change that was based exactly on the meta stats, suggesting that there is no consensus to use them. Similarily, Thurs opposed here to use the meta stats. Also note, that (regardless which stats are used), zh.ws (outside the circle) has much more pages than pt.ws (in the circle) - and a fix of that was in the reverted change. I realize that the meta stats (basing on local wiki counters) use variuos calculation methods for various wikisources, so the numbers are not really comparable. But we should decide to use one set of numbers here, if any.
There are other inconsistencies here, eg. Serbian are far over 10000 regardless which stats are used.
However, the above revert, suggests that any change here requires a consensus. And I strongly oppose basing here on inconsistent stats (the source of the values - if any - as well as for the thresholds should be clearly defined to avoid ping-pong and futher inconsistencies and to allow regular updates). Ankry (talk) 10:31, 31 August 2017 (UTC)
MetaWiki uses the usual count, which basically means at least one internal link and that's it. Therefore, it is a bad representation of Wikisource, as some Wikisources are filled with stub-sized pages with links, while others have many large pages that lack links. Using Proofread as a measurement will also be difficult, as it's not available to all Wikisources. --Ooswesthoesbes (talk) 15:00, 31 August 2017 (UTC)
@Ooswesthoesbes: Not exactly. At least 3 ws (gu.ws, pl.ws, sr.ws) have $wgArticleCountMethod = 'any', so the counter is really based on content namespace definition and total number of pages there regardless whether they contain links or not; de.ws add links to all pages using a footer (so the whole Page namespace is effectively counted there, unlike eg. fr.ws or en.ws). Likely only pages with links are counted for all others; but content namespace definitions may vary across Wikisources. Phe's stats concerning main ns. seem to me to be more reliable. However, eg. in ru.ws most pages are doubled because of two orthography variants used; in en.ws Translation ns is not counted; Author ns is not counted for Wikisources having it, while author pages are counted for others as they are in main ns. there; es.ws create a separate page in main ns for each book page (while most other ws do it in Page ns.) also because of labs database drift some pages are lost from phe's stats. Ankry (talk) 16:11, 31 August 2017 (UTC)
Mmm.. I think it will be very difficult to get a cross-wiki definition, especially because there is no simple way to automate counts like "Articles minus Pages" or "Articles minus Authors". --Ooswesthoesbes (talk) 18:14, 31 August 2017 (UTC)
You will find the vote (2009) about mul.ws main page presentation here. --Zyephyrus (talk) 16:26, 31 August 2017 (UTC)
Yes. The discussion above is mainly about the 100,000 +, 10,000 + etc listings :) --Ooswesthoesbes (talk) 18:14, 31 August 2017 (UTC)
I made two changes in the circle that I hope to be non-controversial:
  • removed outdated numbers (maybe until some consensus concerning them is reached)
  • replaced pt.ws by zh.ws; regardles of used metrics pt seems to be out-of-top-ten now, and zh seems to be in.
Ankry (talk) 20:32, 31 August 2017 (UTC)
Well done. The internal stats have always been broken, so far as I can remember. The only reliable ones are those on phetools. Anyway, since this is not a competition, I think we shouldn't show any number at all, and we should not classify wikis in ranges based on the number of pages. An alphabetical list would be much more fair. Candalua (talk) 09:13, 1 September 2017 (UTC)