Wikipedia:Administrator elections/October 2024/Candidates/Rsjaffe

Nomination

Rsjaffe (talk · contribs · he/him) – I'm Rory Jaffe. I have been on Wikipedia probably since soon after its inception, though I didn't register a username until 2007. I have never edited Wikipedia for pay.

I roam around doing many different things. When I edit, I generally do wikignoming. I use AWB for mass work (e.g., typo fixing, or, in one major instance, using regular expressions to fix messed up references in a mass-created series of articles). I occasionally create an article, typically when I look up something and get surprised that there isn't an article for it yet. Examples include Merrell Williams Jr., Van Norman Dams, Verbena lilacina and Fannie Quigley.

I also prowl for serious issues: e.g., undeclared paid editing, vandalism, and sockpuppetry. I try to be very clear and concise when posting about these issues. For example: Wikipedia:Sockpuppet_investigations/Umama99/Archive, and this recent ANI thread (contributions are Special:Diff/1250036606/1250036802 and Special:Diff/1250038072).

I have a doppelgänger account Truth69420 to make a point about usernames. It's been incorrectly blocked once, by ScottishFinnishRadish who then came back with User talk:Truth69420#I've been hoodwinked! to apologize. That account has made eleven edits in 2 years to ANI: all very brief. While I enjoy humor, I try to avoid derailing things, and often hold myself back when I'm tempted to say something that I find incredibly funny. — rsjaffe 🗣️ 20:10, 8 October 2024 (UTC)

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 want to help address the backlog in certain areas and improve responsiveness to issues. There currently are an insufficient number of active admins, so those who are active struggle to keep up. While I'm willing to pitch in where needed, my initial idea is to help with sockpuppet investigations, vandals, and undeclared paid editors.

2. What are your best contributions to Wikipedia, and why?

A: A few come to mind:

1. Identifying the emerging LLM-generated text issue when it first appeared and bringing it to the attention of other Wikipedians. I was asked to write an article for Signpost, which contains much of the story: Machine-written articles: a new challenge for Wikipedia. Why? Because I identified and provoked action on a serious threat to Wikipedia's mission.
2. Cleaning up Sir Gawain and the Green Knight. A featured article, it had accumulated issues over the years, including disjointed text, messed up references, and a few factual errors. I spent a few months (May to December) in 2021 cleaning it up, to the point where 21% of the current text was written by me. Why? Because this was a very difficult and detail-oriented task that fixed up an important article.
3. Using auto wiki browser with regular expressions to fix orphaned references in well over 1200 autocreated plant articles. (I lost count, but it's probably a lot more than 1200.) Why? Because I identified a serious systemic issue with a large number of articles and found a practical way of fixing them.

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: Yes, I've been in conflicts. I generally try to avoid perpetuating a conflict by simply not responding to personal attacks, just sticking to factual responses. I know who I am and I don't feel the need to start up a fight with someone who is already convinced of my wrongdoing. This is very successful in avoiding derailing a conversation. As an example, see ANI regarding CutePeach, where both I and the contents of my report were attacked, but I avoided responding and getting off topic. However, this isn't very satisfying, frankly, and also may not leave my name unsullied in the view of others. For example, I have been accused of lying about doing a WP:BEFORE search on some AfD nominations and decided not to get into a tussle about that even though I had done a before search. My actual fault there was not doing a Japanese script search before search, and even then, it was still unclear whether some of those were notable. I backed away from non-Latin alphabet AfD subjects after that.

I do engage when there is a good reason to do so (other than defending my reputation). For example, User talk:Rsjaffe#Could you please stop removing my entry, do you have a bias?, where I tried to counsel a new user who was destined for failure if they didn't learn to assume good faith. Unfortunately, that user didn't reform and ended up blocked.

If I become an administrator, I'm going to have to reconsider my nonconfrontational stance, as it is not only my own reputation I have to think about, but rather the effect on other administrators' reputation in general. I probably will have to engage with unjust accusations more directly, though I will continue to try keep the conversation on track and avoid getting into a fruitless back-and-forth on those side issues.

---

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.

Optional question from Liz

4 In the work you say you do, how do you identify an undeclared paid editor?

A:

I won't go into a lot of detail, as this is already starting to turn into a long answer, but I'm happy to answer follow ups to any of this. Even though I'm listing it step-by-step, it's not a linear process: more like following my nose when something doesn't look right.

First step is establishing suspicion. It starts with looking at an article and seeing a lack of neutrality, or many insignificant references, or information that only the subject would know (e.g., a birthdate when there isn't public info on that), or a professional picture of the subject (often hard to get if you're not connected to the subject), etc.

Then, is the editor a single purpose editor or an editor creating articles that don't naturally go together? That is, are the articles plausibly related to some interest the wikipedian may have or do they appear to be tasks the editor has taken on from an external request? Do the articles have images that are of similar quality and nature?

Then comes the investigation. One of my favorite things is to drill down on the images. I focus on images because 1) a paid editor will have access to good images and the subject will want a nice picture, and 2) people may not think of scrubbing information from the images, such as metadata. What does Wikimedia say about the provenance? If there are multiple articles in question, does the metadata on the images indicate using the same camera? Are the images similar in the way they were taken? What does a TinEye reverse image search say about the image (e.g., original source)?

I also try to make plausible guesses as to the editor's identity and search on the web, on the company's site, and other related places for evidence. To avoid using confidential info (which can be submitted following instructions at WP:COIN), I try to use what I've learned externally to improve my inside-Wikipedia sleuthing in the hope of finding internal validation of the paid relationship.

At some point in all this (earlier or later depending upon my initial level of suspicion), I will give the editor the standard templated notice regarding UPE. Where it goes from there depends upon the editor's response (or not). This is the key step. A conversation with the editor usually provides me with the final impression of UPE or not.

I expect that the better UPEs are never detected, but this at least sweeps up the clumsier ones.

Optional question from Thryduulf

5. Why did you choose to seek adminship via election rather than via a standard RFA?

A: The first part of this answer will resonate with the older folks out there. I see the standard RfA process as being traumatic, like the bowel prep for a colonoscopy. Sure, the doc will get a clean look at your potential problems after going through all that prep, but, really, is drinking that whole gallon of prep necessary?

To get a little more serious, the shorter time interval, the attempt to keep the discussion constructive, and the confidential voting all make this a less stressful process for the candidate, though this process has stressed me a bit. The scrutiny shouldn't turn into a hazing: it should be focused on identifying appropriate candidates, and I think this process should do a good job of that.

Optional question from Ganesha811

6. Are there any areas of adminship you do not plan to participate in, due to unfamiliarity or lack of technical knowledge? If you later decided you wanted to help in these areas, what would be your plan to become an effective admin in those areas?

A: In preparation for this election, I reviewed some policies with an eye as to how an admin would function in them. I started with a simple one: vandalism and was impressed with how much judgment was required even there. Sure, there are lots of clear vandalism cases. But the section Wikipedia:Vandalism#Administrator response to vandalism uses "likely", "usually", and other such words frequently, which means that there are judgment calls even there. In short, even in areas that I am familiar with, I'm going to start with enhanced awareness as to edge cases and seek help with edge cases to understand the unwritten norms that arise from the experience admins have had in the area.

So the first answer is that I'm going to be careful even in areas I am familiar with, as there are very few "bright line" rules when deciding cases, and I'll seek out advice as needed. Secondly, I do plan to start out in areas I'm most comfortable with. Third, if entering an area that I'm less familiar with, I'll familiarize myself with the applicable policies and seek out advice from an experienced admin. How I develop competency in a new area will probably depend upon the area. For example, SPIs (which I'm familiar with but don't feel comfortable starting there yet), I'll need advice on the particular evidentiary requirements and the language used (e.g., when addressing IPs that might be engaged in sockpuppetry, as the policy is to not publicly connect IPs to accounts except in extraordinary circumstances).

Discussion

• Links for Rsjaffe: Rsjaffe (talk · contribs · deleted · count · AfD · logs · block log · lu · rfar · spi)
• Edit summary usage for Rsjaffe can be found here.

---

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

AfD record: 84.20% match rate, n of 272. 12 keep !votes to 250 delete !votes. Mildly subjective comment: most participation is from 2022 and earlier; in all 10 of the AfDs he participated in over the past year, his !vote matched the result. Some might find his !votes a bit curt but I don't see any red flags. Of the deletes in this data, 107/272 were from nominations. Most of these numbers are from 2022 and 2021. -- asilvering (talk) 01:04, 22 October 2024 (UTC)

Personal comment: I wouldn't encourage anyone to draw conclusions from these numbers other than "has had a lot of good AfD experience, which is a good thing in an admin candidate". 86.5% certainly is an eyebrow-raising delete % (it does go down to 73.7% if you look only at !votes and not nominations), but I don't think the number itself is terribly relevant given that most of the participation is from 2022 and earlier. Having read through many of the AfDs rsjaffe has been involved in, I would describe his approach as "skeptical", which in my opinion is a strength. Anyone who brings that many articles to AfD is going to get accusations of lying about having done a WP:BEFORE search - I've experienced my fair share of this, and my delete vote/nom rate is a good 30% lower than his.

tl;dr: this is a strong positive for this candidate. -- asilvering (talk) 01:04, 22 October 2024 (UTC)

Thank you for your feedback, and I want to expand a little bit about the change in participation. My overriding goal is to improve Wikipedia. I started to realize that, while there are many articles that do not meet notability, the harm to Wikipedia of having those articles ranges from very little to a lot, depending upon the nature of the article. So my focus turned away from most AfDs, based on the notion that many non-notable articles don't create major problems.

There are also different types of harms. A short list off the top of my head of potential harms from non-notable articles (and again, some articles have the potential for some of these and not others);

1. Fragmentation of information. Encyclopedias should provide context, not just definitions or statistics. Fragmenting into little bits makes it harder to see the big picture when reading a specific article. This is true for some notable subjects as well, and you'll see a number of Merge votes by me: redirect and merge. See WP:PAGEDECIDE for more on this.
2. Potential for contradictory information in the same topic. Fragmentation means multiple pages have to be maintained, and if they contain similar content, when new info arises, they might not be updated simultaneously.
3. Verifiability. Notability (excluding certain special subject-specific notability policies) is about significant coverage in independant reliable sources. If the sources aren't meeting notability, verifiability suffers. When verifiability is weak, you can end up with bad information in the encyclopedia.
4. Opportunity for bad articles. The less traffic an article has, it seems the less likely it is to have had reasonable scrutiny. NPP helps with this, at least at the article's inception.
5. Using Wikipedia as an advertisement. This to me is most serious. I see it mainly with companies, schools and biographies of living persons.

In general, I found other areas where my time was better spent if I wanted to improve Wikipedia. I do, however pay attention to that last category, above, and to a lesser extent, to good merge opportunities. — rsjaffe 🗣️ 02:00, 22 October 2024 (UTC)