User talk:BD2412

From Wikisource
Jump to navigation Jump to search

Status: Active. bd2412 T

I am a Wikipedian. You can talk to me there as well. BD2412 T 00:18, 26 December 2005 (UTC)Reply[reply]

Older discussions: User:BD2412/Archive 001

Deleting Index: page[edit]

I’m sorry to bother you, but could you please delete this Index: page? I moved it here as part of a general renaming scheme, but am unable to edit the former, and by such means, add an {{sdelete}}. Thank you. TE(æ)A,ea. (talk) 15:48, 21 March 2020 (UTC).Reply[reply]

Hoyt's New Cyclopedia Of Practical Quotations (1922)[edit]

I could use some pointers on how to use the {{Hoyt quote}} template. Looking at page 35, quote 24, I propose the "note" parameter "Phædrus. Bk. IV. Prol. 5." should be split into two parameters "work" and "place". Do you agree? - DutchTreat (talk) 13:25, 10 November 2020 (UTC)Reply[reply]

That specific case is an outlier. Virtually all of the quotes in the book contain an en-dash between the author and the work, so the template automatically generates one for the name on the "author" parameter. However, here the book only lists the author (Phædrus), and the place (Bk. IV. Prol. 5.), without listing a work. The "note" parameter is used to achieve the same outcome manually. BD2412 T 15:20, 10 November 2020 (UTC)Reply[reply]
Thanks for the explanation. Not sure if the template has changed. When I use it, the en-dash is not added. There is an issue with a missing period "." after Author. Shall we get this fixed, so the manual workaround will not be needed. This will enhance the metadata associated with each quote.

Non semper ea sunt, quae videntur; decipit
Frons prima multos: rara mens intelligit
Quod interiore condidit cura angulo.

Phædrus. Bk. IV. Prol. 5.

- DutchTreat (talk) 12:45, 11 November 2020 (UTC)Reply[reply]

  • Interesting. Perhaps the en-dash is only generated where there is a combination of content in the author and work fields. The period can be added manually. BD2412 T 15:43, 11 November 2020 (UTC)Reply[reply]
  • I agree. This use of author+place is a good way forward in these rare edge cases. - DutchTreat (talk) 11:10, 13 November 2020 (UTC)Reply[reply]
  • I have a fix in the template for the author+place case. Added an else condition when "work" parameter is not present: User:DutchTreat/projects/template. This version also adds a new parameter "year". Any concerns about these changes? The year can be added without impacting existing usage. However, I am concerned that the change for author+place will impact existing usage. I don't know how to find and fix the existing manual workarounds. Do you? - DutchTreat (talk) 12:27, 29 November 2020 (UTC)Reply[reply]
    • I didn't create the template, but have fiddled with it a bit. I'm not sure there are enough cases of a year being used to justify adding such a parameter. BD2412 T 20:17, 29 November 2020 (UTC)Reply[reply]
      • I don't see any disadvantage of adding an optional "year" parameter. I'm a big fan of metadata. It might be helpful for someone to filter (the few entries using it) by the values entered. If the parameter is available, we'll use it. Bit more concerned about adding the missing "work" option since it would add the extra "." after author for each item that is using the manual workaround. - DutchTreat (talk) 15:04, 30 November 2020 (UTC)Reply[reply]
        • It seems that if we were to do that now, every field that has already had the period added manually would then have two periods. Is there a way to fix that automatically, without having to go through the pages by hand? BD2412 T 17:00, 30 November 2020 (UTC)Reply[reply]

ascii is a subset of utf-8[edit]

I have been on a personal campaign. It is perhaps silly. When I am writing html for myself or whatever (not through the wiki interface) I like to make the html document itself in ascii, but set the encoding to utf-8 because the characters will be needed to display the ascii-compliant document.

This is done via escape characters, like æ which produces æ which looks like æ.

It is difficult for me to watch my documents be transformed from lovely, simple ascii into complicated and needy (of fonts) utf-8. It is just as difficult for me to call "html" "coding" but there are some good practices that are shared between authoring a software and authoring a document of browser instructions.

So, that's about 2 cents....--RaboKarbakian (talk) 17:44, 28 December 2020 (UTC)Reply[reply]


cf. WS:S#Bureaucrat,

Are you still up for it? And if so, would you prefer to self-nominate or that I do so? --Xover (talk) 16:36, 9 January 2021 (UTC)Reply[reply]

I am certainly willing to take on the job if there is not another admin who is interested. I had not intended to self-nominate. BD2412 T 17:19, 9 January 2021 (UTC)Reply[reply]


Hi! Would you mind changing your signature from <font> to <span>? HTML Font tags set off Special:LintErrors/obsolete-tag because they're deprecated by the elder gods of the Internet. :-D Inductiveloadtalk/contribs 14:23, 12 January 2021 (UTC)Reply[reply]

Index:Gentlemen Prefer Blondes (Anita Loos).pdf[edit]

I noticed you did this ([1]). Can you help me with some more? I am slightly discouraged because I didn't see anyone proofreading/validating. AnotherEditor144 t - c 19:23, 9 March 2021 (UTC)Reply[reply]

I created most of the pages, and therefore unfortunately can not validate them. BD2412 T 19:41, 9 March 2021 (UTC)Reply[reply]


Please do not insert ligatures like this [2]. We've agreed not to do this as a community. --EncycloPetey (talk) 15:26, 3 April 2021 (UTC)Reply[reply]

Thanks for letting me know. I was not aware of that convention, but it makes sense given the difficulty of finding them in the character set. Cheers! BD2412 T 15:29, 3 April 2021 (UTC)Reply[reply]
Not only that, but many of those "extra" ligatures do not exist in standard fonts, so when a user selects a font that does not contain that ligature, it stands out because it is is a different font from the rest of the text. The agrees upon ligatures in English texts are a-e (æ) and o-e (œ), including Æ and Œ. If there are others, I cannot recall them at the moment. Of course snippets of text in languages other than English may include additional ligatures. --EncycloPetey (talk) 15:33, 3 April 2021 (UTC)Reply[reply]
I assume, then, that "ff" should be transcribed as "ff" rather than "f", yes? BD2412 T 17:27, 3 April 2021 (UTC)Reply[reply]
Yes, we split the ligatures into their component letters. --EncycloPetey (talk) 17:44, 3 April 2021 (UTC)Reply[reply]

What has wikisource done with w:en:WP:LTA/GRP[edit]

Greetings here,

I'm SHB2000 from Wikivoyage here and I'm just here to ask you for what wikisource has done with him. Is there any edit filter that you guys have in place and what does it do. Could you please comment on voy:Wikivoyage talk:User ban nominations. If that's page protected, then just comment on voy:User talk:SHB2000.

A little bit of background history about his targets on Wikivoyage:

Thanks, and looking forward to your response at wikivoyage.

Regards, SHB2000 (talk) 07:52, 6 May 2021 (UTC)Reply[reply]

I am not aware that this person has ever edited on this project. BD2412 T 20:20, 7 May 2021 (UTC)Reply[reply]
@SHB2000: Yes of course he has been here, and he hits my pages among others, though that comes from both my steward days and my currentxwiki protective actions against him. The most important thing that we do is not feed the troll. This person feeds on reaction and being in control, not so much what they are posting, so don't give them that satisfaction. As Antandrus and DannyS712 chase this person all over the wikis trying to not have a skerrick of the garbage to be anywhere, as such they feed the troll. So we utilise protection, we utilise filters including blocking filters (immediate, 2 hr), we utilise tradition and partial blocks. Don't let them play the revert game, so revdel those intervening edits. Know that they are going to vandalise, so push it to the quiet non-mainspace spaces. Be relaxed about it, don't fall into the trap of being troll food. Noting that I have written some global filters, but this person is adaptive, and will just do anything to be annoying and try to control the game. Most effective I have found is two hour blocks where they operate as that shuts things down and gives times for admins to manage, noting I reserve that for known target spaces. — billinghurst sDrewth 00:39, 8 May 2021 (UTC)Reply[reply]
Thanks! SHB2000 (talk) 01:10, 8 May 2021 (UTC)Reply[reply]

Request to restore edits under blue link[edit]

en:Draft:Untitled fourth Fantastic Beasts film?

How we will see unregistered users[edit]


You get this message because you are an admin on a Wikimedia wiki.

When someone edits a Wikimedia wiki without being logged in today, we show their IP address. As you may already know, we will not be able to do this in the future. This is a decision by the Wikimedia Foundation Legal department, because norms and regulations for privacy online have changed.

Instead of the IP we will show a masked identity. You as an admin will still be able to access the IP. There will also be a new user right for those who need to see the full IPs of unregistered users to fight vandalism, harassment and spam without being admins. Patrollers will also see part of the IP even without this user right. We are also working on better tools to help.

If you have not seen it before, you can read more on Meta. If you want to make sure you don’t miss technical changes on the Wikimedia wikis, you can subscribe to the weekly technical newsletter.

We have two suggested ways this identity could work. We would appreciate your feedback on which way you think would work best for you and your wiki, now and in the future. You can let us know on the talk page. You can write in your language. The suggestions were posted in October and we will decide after 17 January.

Thank you. /Johan (WMF)

18:14, 4 January 2022 (UTC)

Please set the bot/bot-edit flag when using a program, otherwise it clogs up recent changes. TE(æ)A,ea. (talk) 02:34, 25 February 2023 (UTC)Reply[reply]

I am not using a bot, and I have minor changes checked. However, since it concerns you, I'll set the flag. BD2412 T 02:37, 25 February 2023 (UTC)Reply[reply]

I hope you don't mind, but as preparatory to closing the undeletion discussion I replaced the text you had manually added to Campbell v. Acuff-Rose Music with the previously proofread (but now properly redacted) and scan-backed text. It's turned into a bit of a minefield, so I'm trying to navigate a bit carefully, but I'm going by the principle of trying to preserve as much of the original contributions and revision history as possible (even if we had to redact parts of three of the thirty-two pages). I have no particular investment in this text beyond trying to somehow navigate the copyvio issue, so once we've finally gotten it out of that mode I will have no objection to whatever content decisions interested contributors make (i.e. which version of the text to use etc.). I hope that's ok from your point of view? Xover (talk) 10:41, 4 March 2023 (UTC)Reply[reply]

  • @Xover: I have no dog in this hunt except wanting the text to be available for any reader who seeks that content here. Cheers! BD2412 T 16:14, 4 March 2023 (UTC)Reply[reply]

Use of AI tools to do OCR corrections..[edit]

Based on the increasing availability of LLM's and the existence of a few 'free' models. Would you be willing to consider looking into the possibility of using an LLM style tool to scan pages for possible scan OCR errors, though use a robustly engineered prompt to one of the free engines available? (Assuming you can get a consensus to run a tool like that on toolforge.) ShakespeareFan00 (talk) 21:40, 15 April 2023 (UTC)Reply[reply]

I would support having such a thing, but I do not have the technical prowess to implement one. BD2412 T 21:41, 15 April 2023 (UTC)Reply[reply]
Do you know anyone that might? ShakespeareFan00 (talk) 21:49, 15 April 2023 (UTC)Reply[reply]
I can certainly ask around on Wikipedia. BD2412 T 22:03, 15 April 2023 (UTC)Reply[reply]
Please note that there is debate about the use of LLM's at Wikipedia right now. ShakespeareFan00 (talk) 22:11, 15 April 2023 (UTC)Reply[reply]
I am well aware of it — and have participated in it previously. It is sprawling! BD2412 T 22:26, 15 April 2023 (UTC)Reply[reply]
@ShakespeareFan00: The answers that I got were:

Do you know about LangChain? It's by far the most serious platform for building apps from LLMs in an open non-proprietary way. Although the guy behind it is on Twitter, he and others are far more responsive on their Discord server.

I believe that today's commercial OCR software does include language models for error correction, but while they are not "large" as in LLMs, I believe they are substantially larger than typical autocorrect systems. A very good correction system involving LangChain and Pywikibot should be possible to make from open ~7B size models (e.g. Dolly, see Ars Technica's summary) which run fairly fast on typically four ordinary server CPU cores. It should be possible for project communities to thoroughly test such at a sufficiently large scale to find any issues which might cause serious problems. I suspect that corrections can be automatically classified into those which should require human review, and those which most probably don't need it.

Cheers! BD2412 T 18:14, 16 April 2023 (UTC)Reply[reply]
I see you are already in that conversation. Excellent. BD2412 T 18:20, 16 April 2023 (UTC)Reply[reply]
Withdrawing my suggestion owing to the EU plans to regulate AI, which would make it much harder for individual implementers to train and use their own custom models. ShakespeareFan00 (talk) 23:02, 20 April 2023 (UTC)Reply[reply]
Quite a lot of us are in the U.S., though. BD2412 T 23:10, 20 April 2023 (UTC)Reply[reply]

New admin[edit]

Just realised that I hadn't dealt with this and started only to find you're in process. Will you complete all the steps per User:Hesperian/Notes/Admin closure, or do you want me to do some of them? Beeswaxcandle (talk) 06:03, 3 July 2023 (UTC)Reply[reply]

I'll finish up. Thanks! BD2412 T 06:14, 3 July 2023 (UTC)Reply[reply]
We seem to have run out of room at Wikisource:Administrators/Archives#Timeline. If so, I am thinking that we can just spit 'crats/CUs into a separate table. BD2412 T 06:34, 3 July 2023 (UTC)Reply[reply]
Hmm. I had to do some tweaking on the height when I added Billinghurst as Interface Admin. However, I think it's more about the width as this happened last month when I added the extra month, then it settled down after a couple of days and behaved again. Beeswaxcandle (talk) 06:40, 3 July 2023 (UTC)Reply[reply]
Well, I split it, and that did it. I had gotten a warning about the length of the timeline, so this probably buys us a few more years of admins. BD2412 T 18:19, 3 July 2023 (UTC)Reply[reply]