Wikipedia

Wikipedia talk:Reliable sources/Perennial sources#Wikileaks


MREL subpage

[edit]

Since the MREL classification has been a recurring topic of discussion here, I've been working on a taxonomy of MREL sources, which can now be found at the subpage WP:RSP/Further classification. The groups are determined based on the reasoning presented in the RSP summaries. The goal is to provide information on the types of MREL sources, along with clarifying the differences between them, and to help with the related discussions.

The page also includes a classification of the 109 individual MREL sources. Initially, I wasn't planning to include this, as my focus was on the overall system, but I had done a brief classification as part of my review, and rereading some of the discussions convinced me that it may be useful. That said, I found that the evaluation was more subjective than I expected. Broadly speaking, I considered a source to qualify for a group when the group description is part of the justification for the MREL classification, in a way that wouldn't just apply to most or all sources. While I have included specific numbers below, they shouldn't be overly emphasized. Review would be appreciated, as the current version is just a first pass.

Description:

  • There are three primary groups: variation in reliability within the source, the source is marginally reliable, and unclear/no consensus (groups 1 through 3). Every MREL source is classified in at least one of the primary groups. There are also three additional groups: the source is used for opinion, it is only reliable for specific primary-sourced information, and it is no longer updated (groups 4 through 6). Some of the groups could potentially be divided further, e.g. based on the type of concerns associated with each source (self-published, poor editorial control, etc).
  • There is considerable overlap between groups. For example, most MREL state-controlled media fall under both groups 1 and 2 (different reliability depending on topic, marginal reliability on some topics). Likewise, there can be different levels of reliability with at least one unclear aspect (groups 1 and 3), marginal reliability with at least one unclear aspect (groups 2 and 3), etc. Even among sources that only qualify for a single group, many of them have unique additional information beyond what is captured by this taxonomy, which might qualify them as an overlap if more detailed categories were added to the classification system.
  • Out of 109 MREL sources, the number of sources that qualify for each group are 68, 85, 53, 9, 2, and 4. In terms of percentages, 78% of sources are in group 2, 62% are in group 1, and 49% are in group 3. The sources that qualify for only one group are a minority (25/109, or 23%). These numbers should be considered approximations, as per the above caveats.

I'm also aware of the recent discussions discussing a potential split. This page will be relevant whether or not that happens, and could potentially be used to help with a transition. However, it does raise the potential issue that the "no consensus" sources generally can't be cleanly distinguished from the others. As a result, the method of dividing sources would have to be determined, and in many cases it would likely require some amount of discussion. I suppose I've divided the sources as well, but allowing them to be in multiple groups greatly simplifies the issue. "No consensus" is usually not the only result, and if the groups had to be mutually exclusive, it would be possible to move some sources between their presumed groups by minor changes to their summaries, e.g. "no consensus for X...consensus for Y" compared to "consensus for Y...no consensus for X". The specific numbers (again, noting the caveats above) are: among 51/109 sources recorded as no consensus (53 in group 3, but two of them qualified under a different criterion), only 8 of them had no overlap with at least one other group, and 21 of them overlapped with two or more other groups. Even among those with no overlap, some of them include potentially useful information in their descriptions that goes beyond simply identifying a lack of consensus. Sunrise (talk) 10:07, 23 May 2025 (UTC)[reply]

I like the idea. I wonder how the subpage could be made visible to those who actually need it. Aaron Liu (talk) 14:42, 16 June 2025 (UTC)[reply]
I've added a link in the legend as a first step. Perhaps a note at RSN might be warranted, though I'd appreciate additional review before that point. Sunrise (talk) 02:24, 19 June 2025 (UTC)[reply]

Editorial voice

[edit]

The entry for The Economist currently says:

"Most editors consider The Economist generally reliable. Distinctively, its news articles appear without bylines and are written in editorial voice. Within these articles, Wikipedia editors should use their judgement to discern factual content – which can be generally relied upon – from analytical content, which should be used in accordance with the guideline on opinion in reliable sources. Its pseudonymous commentary columns and other opinion pieces should also be handled according to this guideline."

I don't know what is meant by the words editorial voice, which were introduced by @SamuelRiv here. It seems to be interpreted by some editors as meaning that everything in the magazine is an Opinion piece ("editorial"). I doubt that this was the intended meaning. (The Economist is famous for their data journalism, not for their opinion pieces.)

Compassionate727 wrote the closing summary for the 2022 RFC, but did not use these words. The only editor in the RFC who used this phrase positively was Newimpartial (another editor disagreed that this example was actually an instance of editorial voice). AFAICT at a glance, Newimpartial meant something fairly close to "they don't put bylines on their articles". Mx. Granger and SamuelRiv talked about the wording in Wikipedia talk:Reliable sources/Perennial sources/Archive 7#Contradictory criteria of reliability wrt opinion but this doesn't provide a definition of the terms.

To give you an idea of what Editorial voice means in the media industry, I offer these quotations from Wikipedia articles:

  • Copley Press#History: "In 1942, Copley bought the Journal's Democratic-oriented competitor, the Illinois State Register, promising that the Register could keep its independent editorial voice."
  • Jewish Currents#Editorial position: "The magazine emerged as a leading voice of the American Jewish left. Its editorial voice, led by contributors such as Peter Beinart, is strongly critical of Israel and advocates positions such as the Palestinian right of return, and boycotting Israeli businesses in the occupied West Bank."
  • Southern California News Group#History: "The newspapers also share editorials and speak as one voice on regional issues. Some have said that the former Freedom Communications' distinctive conservative-libertarian editorial voice may be lost in the process."

Editorial voice is, in short, the "voice" (as in advocacy; as in the Voice of the people) of the publication's Editorial board as expressed in Editorials (i.e., not "news articles").

Within the non-news publishing industry, Editorial voice is also to indicate something stylistic, i.e., the book editor's equivalent of Writer's voice.

Neither of those meanings actually make any sense in this context. What is meant by these words? WhatamIdoing (talk) 20:09, 8 June 2025 (UTC)[reply]

Misleadingly-titled, the (haphazard-as-considered-by-many-participants) RfC was only on trans topics, thus it didn't relitigate its general style. I would suggest looking into the four discussions the entry had at the time of this wording's addition in 2018. Aaron Liu (talk) 21:26, 8 June 2025 (UTC)[reply]
I would interpret this entry as "The Economist frequently mixes opinion with factual coverage in its articles" and that editors should be able to tell which phrases are opinion if they scrutinize the passages. Aaron Liu (talk) 21:49, 8 June 2025 (UTC)[reply]
Since nobody knows what it means to write a news article in "editorial voice", I've removed that (and added a brief description from the lead of the Wikipedia article). Maybe it will help. WhatamIdoing (talk) 19:47, 15 June 2025 (UTC)[reply]
Good change. Firefangledfeathers (talk / contribs) 22:42, 15 June 2025 (UTC)[reply]

NewsNation

[edit]

The NewsNation entry links to NewsNation (UFOs, UAPs). but the link doesn't go there. Bubba73 You talkin' to me? 14:36, 16 June 2025 (UTC)[reply]

Should be fixed. Aaron Liu (talk) 14:41, 16 June 2025 (UTC)[reply]
Done[1]. -- LCU ActivelyDisinterested «@» °∆t° 14:59, 16 June 2025 (UTC)[reply]

Tasnim

[edit]

Why does WP:TASNIMNEWSAGENCY not set some automatic trigger when invoked similar to other deprecated tabloids (Mail, Sun, etc)? It is quite difficult having to manually scrape them off tons of affected articles over the years. Borgenland (talk) 16:00, 16 June 2025 (UTC)[reply]

It appears it was never added to 869 (hist · log), the edit filter that triggers the deprecated message. It can be requested at WP:EFN -- LCU ActivelyDisinterested «@» °∆t° 16:47, 16 June 2025 (UTC)[reply]