Wikipedia:Administrator elections/December 2025/Candidates/Epicgenius: Difference between revisions

By /

Line 57: Line 57:

====Optional question from [[User:ZKang123|ZKang123]]====

====Optional question from [[User:ZKang123|ZKang123]]====

:”’9.”’ With regards to the [[Wikipedia:Featured article candidates/Sengkang LRT line/archive1|Sengkang LRT line FAC]] and [[Wikipedia:Good article reassessment/Singapore Rail Test Centre/1|the GAR for Singapore Rail Test Centre]], do you think Wikipedia’s sourcing requirements may reinforce structural biases? Do you think these might prevent or limit participation – for example in Asia / the global south?

:”’9.”’ With regards to the [[Wikipedia:Featured article candidates/Sengkang LRT line/archive1|Sengkang LRT line FAC]] and [[Wikipedia:Good article reassessment/Singapore Rail Test Centre/1|the GAR for Singapore Rail Test Centre]], do you think Wikipedia’s sourcing requirements may reinforce structural biases? Do you think these might prevent or limit participation – for example in Asia / the global south?

::”’A:”’ I’ll answer this by addressing underrepresented regions more generally, and Singapore more specifically. In the more general case, Wikipedia requires reliable, independent sources, which are harder to come by in countries that may have no or little freedom of the press. This would necessarily mean that, for these countries, there is less reliable sourcing from independent sources such as newspapers and magazines, compared to countries where such sourcing may be more prevalent. In such cases, one may have to use other types of sources (such as books), which may be published elsewhere. Alternatively, without sufficient reliable sourcing, the topic would not meet [[WP:N]], which can certainly contribute to systemic bias. However, the need to reduce the impact of systemic bias should be counterbalanced with the need to use higher-quality sources where possible, particularly if it would affect [[WP:NPOV]].

::”’A:”’ I’ll answer this by addressing underrepresented regions more generally, and Singapore more specifically. In the more general case, Wikipedia requires reliable, independent sources, which are harder to come by in countries that may have no or little freedom of the press. This would necessarily mean that, for these countries, there is less reliable sourcing from independent sources such as newspapers and magazines, compared to countries where such sourcing may be more prevalent. , , , , the [[WP:]], , to the of to , .

::In the specific case of the two Singapore-related articles, there was a disagreement on whether [[the Straits Times]] (ST) or other Singaporean sources should be considered a high-quality source for FAC or GAN. At the time, the [[WP:RSP]] table classified the ST as marginally reliable in its “Status” column, while noting in its “Summary” column that the ST was generally reliable other than its coverage of political topics, as Singapore does not have full freedom of the press. These two articles generally relied on Singaporean sources’ non-political coverage, which were the most detailed sources available. I advocated for treating ST as generally reliable other than its political coverage, while a few other editors took the position that the ST should be treated as marginally reliable; eventually, there was a [[Special:PermanentLink/1306955280#Reliability of The Straits Times must be rediscussed|discussion at WP:RSN]] over whether to change the content of the Status column. The outcome of this discussion was not ideal, as it resulted in a FAC contributor walking away from the process completely. However, I also would not say it had much of an impact on systemic bias, since it did not impact whether the ST or other Singaporean sources contributed to notability – what essentially happened was that a source was reclassified from “marginally reliable for one topic” to “generally reliable for all except that topic”.

::In the specific case of the two Singapore-related articles, there was a disagreement on whether [[the Straits Times]] (ST) or other Singaporean sources should be considered a high-quality source for FAC or GAN. At the time, the [[WP:RSP]] table classified the ST as marginally reliable in its “Status” column, while noting in its “Summary” column that the ST was generally reliable other than its coverage of political topics, as Singapore does not have full freedom of the press. These two articles generally relied on Singaporean sources’ non-political coverage, which were the most detailed sources available. I advocated for treating ST as generally reliable other than its political coverage, while a few other editors took the position that the ST should be treated as marginally reliable; eventually, there was a [[Special:PermanentLink/1306955280#Reliability of The Straits Times must be rediscussed|discussion at WP:RSN]] over whether to change the content of the Status column. The outcome of this discussion was not ideal, as it resulted in a FAC contributor walking away from the process completely. However, I also would not say it had much of an impact on systemic bias, since it did not impact whether the ST or other Singaporean sources contributed to notability – what essentially happened was that a source was reclassified from “marginally reliable for one topic” to “generally reliable for all except that topic”.

Nomination

Epicgenius (talk · contribs) – I’m delighted to nominate Epicgenius for adminship. Epicgenius knows what it is to work hard and get things done. With their 500+ good articles, they are the fourth-most prolific GA writer of all time. Compared to writing content, working in AIV, RFPP and REVDEL backlogs is easy, and their content experience makes them a good fit to work on DYK. Their ability to speak some Cantonese and Mandarin will also be welcome in the admin corps. I’m glad they are willing to turn some of their time working on content into helping with administrative tasks and I have no doubt they will do well with the tools. —Femke 🐦 (talk) 17:02, 1 December 2025 (UTC)[reply]

Co-nomination

Anyone who’s seen Epicgenius’s work across the project over the years knows that this nomination is long, long overdue, and I’m absolutely thrilled to co-nominate him. Epicgenius is, of course, best known for being an endlessly productive and talented content writer – and that’s verifiable! I’ve handled dozens of his DYK noms over the years and can scarcely complain – in fact, sometimes I’ll look for one of his hooks to grab in a pinch because they’re pretty much guaranteed to be interesting and accurate. His content and content-adjacent work alone would make him a great fit for DYK queue moves, revision deletions, and AfD closes.

But Epicgenius is also a helpful participant at content review, content discussion, and user conduct noticeboards, with lots of experience letting admins know where trouble is and handling technical requests. The combination of those abilities gives him a knack for thoughtful analyses of user behavior, like this thoughtful pushback on an assertion that another editor was using AI. I’ve always found Epicgenius to be thoughtful, meditative, and courteous in our interactions around the project; I’m positive that with the mop, he’d be even more of an asset to the project than he already is. theleekycauldron (talk • she/her) 17:12, 1 December 2025 (UTC)[reply]

Candidate, please indicate acceptance of the nomination here: I accept this nomination. I have a few past alternative accounts and doppelgangers, which I have disclosed to ArbCom. Epicgenius (talk) 17:20, 1 December 2025 (UTC)[reply]

Please disclose whether you have ever edited Wikipedia for pay: I have never edited for pay, nor will I ever do so. I have disclosed a non-financial conflict of interest and avoid editing articles in that topic. Epicgenius (talk) 17:20, 1 December 2025 (UTC)[reply]

Questions for the candidate

Dear candidate, thank you for offering to serve Wikipedia as an administrator. Please answer these questions to provide guidance for participants:

1. Why are you interested in becoming an administrator?

A: I try to help out wherever I can, using the skillsets I’ve gained throughout my 13 years as a Wikipedia editor. These range from anti-vandal work, with which I was heavily involved in my first few years, to sockpuppetry investigations and copyright violations, which I come across periodically. Although my more recent work has been with content matters such as WP:DYK, WP:GAN, and WP:FAC, I’ve recently noticed numerous parts of Wikipedia that may benefit from additional administrator eyes. In particular, I would like to help tackle the WP:AIV, WP:RFPP, and WP:REVDEL backlogs, as well as Main Page matters including the promotion of DYK queues. Keeping these backlogs low goes hand-in-hand with content creation in improving the quality of Wikipedia.
2. What are your best contributions to Wikipedia, and why?

A: For the last few years, I have helped improve hundreds of articles, especially those related to my hometown of New York City, as well as my hobbies of architecture and transit. Many of these have been reviewed by other editors and, as a result of these collaborations, have reached Good Article or Featured Article status. Legoktm has written a Toolforge tool listing out the articles I’ve helped improve, and another list of the articles I’ve worked on is here. Although I’ve worked on many pages (500+ good articles and 37 featured articles at last count), my favorites are possibly New Amsterdam Theatre, part of a 42-article good topic promoted in collaboration with Found5dollar, and articles that I helped other editors improve to good or featured article status, such as Felix M. Warburg House and Gowanus Batcave.
In the past two years, I have been a coordinator and one of the judges for the WikiCup, a competition intended to encourage editors to contribute quality content (e.g. GAs and FAs) to Wikipedia. While not as important in the grand scheme of things, compared with my content creation work, my time as a WikiCup coordinator has also taught me how to handle delicate issues that arise. I’ve been particularly involved with reviewing the quality of submissions; in the past, there have been competitors who have submitted poor-quality articles for the sole purpose of gaining points. Understandably, this has created situations where editors have called for reform or abolition of the WikiCup, but by encouraging high-quality article improvements and reviews rather than quantity of submissions, I’ve been trying to reduce these negative impressions.
3. Have you been in any conflicts over editing in the past or have other users caused you stress? How have you dealt with it and how will you deal with it in the future?

A: The collaborative nature of Wikipedia necessarily means that not everyone will always get what they want, particularly in consensus-based processes such as AfD and FAC. Everyone will have differing opinions on how to improve articles; the right answer is only found when consensus-based discussions win out over antagonism and rancor. A relatively recent example of this is at Talk:The Queen of Versailles (musical), where, in response to a dispute over how to summarize the reception of a new Broadway musical, I created a reception section to more accurately describe each of these viewpoints in detail. Focusing on the content, rather than on the contributor, helps resolve any issues that may come up. I discuss issues when I see that something may be in dispute, and I’m willing to change my initial viewpoints in response to arguments based on policies and guidelines.
I signed up as a young teenager and did not realize these things; early on in my editing career, I made some disruptive edits and have three related entries in my block log, all from over a decade ago. I take complete responsibility for the behaviors that led to these blocks, and I’ve taken care not to repeat them. I have not had any sanctions since. Instead, I’ve come to not stress out over edits that I disagree with, and I assume good faith of other edits.

You may ask optional questions below. There is a limit of two questions per editor. Multi-part questions are disallowed, but you are allowed to ask follow-up questions related to previous questions. Make sure to use level-four section headers, not boldface.

4. How will you approach content that incorporates AI or language generated by it? To what extent should it be used, and when should it be, if at all?

A: I should make a distinction between generative AI (such as LLMs) and other types of AI (such as machine learning). Machine learning can be very useful, such as in neural networks like ClueBot, without which vandalism would be much more rampant. On the other hand, I do not use LLMs and believe that they should not be used for content on Wikipedia. There is empirical evidence that LLMs hallucinate and give out incorrect information, or, on the flip side, commit copyright violations by adhering too closely to the source text. Additionally, there is no guarantee that someone who has used an LLM has looked at the text they’re adding. If I were to see content that shows unambiguous signs of having been generated by an LLM, I would remove it. If the content looks salvageable, I’d also check the sources and add back the content in my own words.
In the limited cases where AI may be useful to regular editors, the output needs to be, at the very minimum, double-checked. For example, AI can dig up sources that may not be readily discoverable using traditional methods; of course, one still has to review the content of the source manually, as the associated LLM often gives incorrect summaries of whatever it finds. I am not discounting the possibility that the quality of AI will improve in the future, but for the moment, my opinion is that LLM is not to be used on Wikipedia.
Since some people may still be unaware of the myriad issues with LLMs, good-faith editors should be warned against using them. Repeated misuse after several warnings would be akin to a WP:CIR issue, especially as we have numerous guidelines against LLM-generated text, such as WP:AITALK and WP:NEWLLM. On the other hand, this should be counterbalanced with the need not to bite newer editors who may merely be writing in a more professional tone (or more familiar with policies and guidelines) than we usually see on Wikipedia, and, as such, are sometimes accused of using LLMs.
5. Do you believe Wikipedia has been effective in enforcing its Neutral Point Of View policy and limiting editorial bias in its content, including in contentious topics? If not, what should be done to reduce possible bias?

A: In general, I think the community has done a good job maintaining WP:NPOV in more popular topics. In contentious topic areas particularly, articles are watched by large numbers of editors, many with widely varying backgrounds and opposing viewpoints. This helps balance out whatever bias crops up, since contentious additions and statements require consensus, and moderate positions tend to be more-policy compliant and closer to the consensus of reliable sources, compared with more extreme ones.
Bias varies more significantly the more niche or the less well-known the subject is, and it can occur not only in the verbiage and breadth of articles themselves, but also in the form of systemic bias, namely in terms of which topics are covered. Wikipedia editors tend to hail from particular demographics; articles will often represent editors’ fields of interest or expertise, and, at least on this project, focus on the western world. Wikipedia is also sometimes accused of having an ideological bias. Some of the most controversial topics are also among the most viewed on Wikipedia, so details such as reference quality and word choice can play a significant role in shaping an article’s neutrality, or impressions of such.
There are initiatives to reduce systemic bias; for example, the Women in Red project seeks to create articles about notable women, while the Developing Countries WikiContest encourages participants to write about topics in developing or developed countries. As for content-related bias, this can be resolved by having editors from a variety of viewpoints and sticking to reliable sources. Though the perennial reliable sources list has itself been accused of bias, the ratings on the list are largely fair, since they are a result of consensus-based discussions involving significant discussions of evidence, rather than mere headcounts. Furthermore, even at places such as the reliable sources noticeboard, consensus can change whenever new evidence comes up.
6. If you could institute a broad change to Wikipedia’s editing culture, what would it be?

A.: To preface this, I wouldn’t implement a broad change without first gaining consensus. That being said, it would be better if we made it easier for editors and casual readers to identify and fix problems with articles (such as verifiability issues or grammar errors). More than once, I have come across instances where someone found a possible error in an article but found the process of even bringing it up on the talk page, never mind fixing the article directly, to be rather tedious. Further exacerbating the issue, when these issues do get flagged on a talk page, they tend to slip through the cracks unless the article has a particularly high number of watchers. Even a maintenance tag on the article itself, such as {{citation needed}}, may not be resolved for months or years. While I do not have a specific or easy solution to this, it would be excellent if changes in the editing culture made it easier for users to more easily raise and resolve issues.
7. Why did you choose to go through Admin elections instead of RfA?

A: I chose to undergo an admin election because I found the process to be less stressful than RfA. At a traditional RfA, candidates answer questions and see how other editors !voted in real time, and some people may !vote before the candidate has been given the chance to answer a certain question. At an admin election, despite tradeoffs such as a more rigid election schedule, these issues are not a concern. Additionally, since several candidates are running at the same time, it lessens the pressure on individual candidates.
8. Which admin action do you feel is the most sensitive and why?

A: There can be several ways to interpret the concept of a “sensitive admin action”, but in the sense that an incorrect action in this regard may cause damage, I consider changing a page’s protection settings the most sensitive action to undertake, as it can directly affect reader-facing content. There may be little downside to unprotecting a page where the risk of disruption is low, since it may encourage good faith contributions, and protection of articles in mainspace is generally not done preemptively (with some exceptions such as today’s featured article and pages under certain contentious topics). Unprotecting the wrong page, however, may result in disruptive edits being disseminated more widely. This is doubly so if one unprotects a template, since disruptive edits to a template can be seen on any article where the template is transcluded.
9. With regards to the Sengkang LRT line FAC and the GAR for Singapore Rail Test Centre, do you think Wikipedia’s sourcing requirements may reinforce structural biases? Do you think these might prevent or limit participation – for example in Asia / the global south?

A: I’ll answer this by addressing underrepresented regions more generally, and Singapore more specifically. In the more general case, Wikipedia requires reliable, independent sources, which are harder to come by in countries that may have no or little freedom of the press. This would necessarily mean that, for these countries, there is less reliable sourcing from independent sources such as newspapers and magazines, compared to countries where such sourcing may be more prevalent. Without sufficient reliable sourcing, topics would not meet WP:N. Although this can certainly contribute to systemic bias, topics should use high-quality sources where possible, particularly if it would affect WP:NPOV, the second of the five pillars. I would thus say that, while Wikipedia’s sourcing requirements may indeed reinforce structural biases, they are arguably also necessary to maintain the integrity of encyclopedic content. Unfortunately, there is no easy solution to this; Wikipedia follows the sources, not the other way around.
In the specific case of the two Singapore-related articles, there was a disagreement on whether the Straits Times (ST) or other Singaporean sources should be considered a high-quality source for FAC or GAN. At the time, the WP:RSP table classified the ST as marginally reliable in its “Status” column, while noting in its “Summary” column that the ST was generally reliable other than its coverage of political topics, as Singapore does not have full freedom of the press. These two articles generally relied on Singaporean sources’ non-political coverage, which were the most detailed sources available. I advocated for treating ST as generally reliable other than its political coverage, while a few other editors took the position that the ST should be treated as marginally reliable; eventually, there was a discussion at WP:RSN over whether to change the content of the Status column. The outcome of this discussion was not ideal, as it resulted in a FAC contributor walking away from the process completely. However, I also would not say it had much of an impact on systemic bias, since it did not impact whether the ST or other Singaporean sources contributed to notability – what essentially happened was that a source was reclassified from “marginally reliable for one topic” to “generally reliable for all except that topic”.
10. In recent years there has been a lot of discussion about the importance of assessed content (GA/FA) experience for effective administrating. As someone with a stellar record in that regard, do you think it will benefit you? If so, how?

A: Thanks for noticing, and yes, I certainly think the GA and FA experience helps. The WP:FAC and WP:GAN processes, and maintaining the quality of FAs and GAs, both require interacting with policies and guidelines at great length. To give a few examples: The GA criteria require adherence to copyright policy, and articles with significant violations of copyright may be speedily failed – something which is directly relevant to the use of WP:REVDEL. Both the GA and FA criteria require that articles be backed up by reliable sources; by extension, this means being able to spot unreliable or promotional sources, a useful skillset when dealing with promotional editing and spam. Maintaining the quality of good and featured articles also entails detecting and reverting vandalism where appropriate, which is directly related to the antivandalism skillset.
11. What is your opinion on admin recall generally?

A: In theory I like the concept of a recall process, as it generally benefits the community to hold admins accountable for serious concerns such as behavioral issues. At the same time, I think the process as currently implemented can certainly be improved. There are valid concerns that the current iteration of the recall process could be used over relatively minor issues, so I support discussions over fine-tuning the recall process.
12. Are you aware of any editor you would nominate for adminship yourself at some point after getting the mop, if you do ? You do not have to identify them

A: I don’t have anyone in mind at the moment. I know of some editors who may do well with the tools, but I’m not sure if they even want to run.
13. You have made more than 350,000 edits. Have you been using any editing tools that increase your edit count?

A: Yes. In the past, I reverted vandalism using a now-defunct tool, WP:STiki, making tens of thousands of edits that way. I’ve also used AutoWikiBrowser and a similar tool, JavaScript Wiki Browser, to make edits such as setting up WikiCup competitors’ submission pages and replacing navboxes in conformance with WP:BIDI. However, most of my edits (about 90%) are manual; I typically make a few dozen edits every day, which range from minor copyedits to article expansions. The XTools average shows that I’ve made about 77 edits per day since registering, though these statistics are skewed by the bursts of semi-automated edits that I make.

Discussion

Please keep discussion constructive and civil. If you are unfamiliar with the nominee, please thoroughly review his contributions before commenting.

Must Read

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top