User:Selbsportrait

Who I am matters less than what I do, so here is a memento for editing the wiki discovered along the way, a way that is shorter than the number of years might indicate. The "you" on this page mostly refers to me. Including in the claim that follows: you do you.

Instead of reading criteria and tactics, you could consult my précis on style or my editing kit. Preferably, you would disambiguate a link, improve an entry, add citations to a barebone article, rescue an orphan or read a random page.

Criteria

Editing in a nutshell: ABC—Always Be Constructive.

In more words, the opposite of deleting something is creating something. The criteria that follow amounts to: preserve the good stuff, be gone with the long,^[1], say what you mean, and compete with grace.^{[nb 1]}

Cooperation

We form a rewriting system that composes a public encyclopedia of everything worth noting. We need one another to pull it off. Minimally, our cooperation involves making contributions that improve the encyclopedia; maximally, it solves the open exposition problem.^[2]

Cooperation involves you above all: nobody but you can carry your weight. Your edits need, on the aggregate, to improve pages. A modicum of competence is required. Displaying overconfidence won't hide it.

Cooperation involves readers too: we expect them to be able to count and know how time works. We also expect them to be able to read three-syllable words from time to time. This is not the Simple Wiki.

Cooperate to preserve appropriate content.

Conciseness

We cannot consign every human thought to the wiki, so we must summarize. Cutting words and sentences hopefully increases informativeness. Reducing irrelevant information too. Both are a boon when the number of notable topics keeps growing. Adding will always be easier than subtracting.

The shorter our contributions the better, a preference both subjective and fallible. Shortening creates space for others. Every new information can improve the sum of what we have so far.

Conciseness alone does not justify removing information from an article.

Clarity

Text comes and goes as a result of a mutual effort to make information shines through. This work entails tradeoffs. We can't please everyone.

We should not expect other editors to pay more than minimal attention to what they read. Don't take feedback too personally. Make the best out of it, and focus on leading readers to the fountain of knowledge.

It is usually preferable to make an edit that retains at least some elements of a questionable prior edit than to revert it entirely.

Concurrence

Edits occur in parallel and with a shared commitment to veracity. Two types of concurrence (co-occurence, agreement) that make us compete toward a common goal, ideally without acrimony. Trying to see eye to eye with everybody else won't reduce friction, far from it. Let us look in the same direction, rules be damned.

The five pillars are too often misunderstood. You will encounter contentions that notability applies to each disputed claim; it doesn't: it's a test to justify article creation. Or that secondary sources support notability; it rather increases reliability. Or that making an inference implies originality, as if common sense was unneeded. Reliability is about authenticity, not veracity: few encyclopedic entries are built out of statements about brute facts. Relevance, applied to pages, is reliability; the concept is relative: it is a part-whole relationship. A claim is contentious only if you wouldn't like it to be made of your mom without proper support.^{[nb 2]}

None of that matters. You're not here to change people. You're here to fix pages. We all are.

Constructiveness

Our North Pole will always be to improve the encyclopedia. Ask yourself if what you do adds to or tightens its construction. If not, let go.

Work to make gnomes and elves appear. In the end only selfless joy makes the wiki worthwhile. Call for their help: they're diligent, and kind. When fairies enter the scene, it's time to move elsewhere. Unless you're a fairy yourself, your work is done, at least for now. Whatever happens afterwards is out of your control.

Stay away from windmills and let knights be. Many mean well, most are clumsy. Expressions like "WP:" simplify assertion; they do not replace reasoning. Many essays are wrong, most guidelines are misinterpreted. Better to say what you think, and why.

The most important principle is to continually improve rather than destroy.

Tactics

The principles above can help provide an interpretation regarding the five noble truths. They don't tell you what to do. In day-to-day editing, tactics suffice.

TL;DR—Give more than you take. Leave pages better than they were when you arrived. Read before you write. Have fun.

Tact

Being considerate is always worth a try, if only for oneself. Consider that you can always be wrong. The feeling of being right may induce righteousness.

Being wrong is part of the editing game. I know better than anyone else how often I were. To use wrongness against someone reveals one's own predispositions more than it teaches anything. Humans almost exclusively learn vicariously.

Manage your personal limits properly, forgive, and then forget. The task is too big to push people away. Nobody likes cops.

One can be perfectly civil and follow every rule of etiquette and still be a jerk.

Talk

Should be done for the sake of editing pages. Most disputes over content may be resolved through minor changes rather than taking an all-or-nothing position. If you enjoy company, consider joining a project.

Editor pages

My own talk page is reserved to talking. An exchange natural enough to feel a conversation going on more than a diplomatic mission. I delete them regularly when they're done, and immediately when being patronized. To discourage hawks improves everyone's experience.

Entry pages

Talk pages may help the entries they're meant to, but only as a last resort. Its syntax is beyond hope, fogs for pettifogging. Its main use is to avoid edit wars.

I still make my voice heard from time to time, and vote very sporadicly. I would prefer not to, and so comment my edits instead. (Commenting directly in the page is also possible.) Editing will always remain the best way to resolve issues.

Pro/contra

It is possible to offer arguments for or against a claim or a decision without taking a stance. It might even be the best remedy against those who defend the Truth.

Both the inclusionist and the deletionist have a role to play. The former have the facts, the latter the logic. It's very important that inclusionists don't get too categorical with the logic they apply, for we don't live in a world closed under deduction, and we need to fix first.

Rock/Talk ratio

Check your last 500 contributions. Count the number of "talk:" and "wikipedia:". That's your Talk number. Your Rock number is 500 minus your Talk number. If that ratio is consistently less than one, then you may be a metapedian. I'm here to rock.

Stick to the point

Code words are weak. If you can't say what you mean directly, drop them. If you can say what you mean, you don't need them.

Voicing an opinion is not an argument. Consensus is based on tallying up arguments, not votes. That's the meaning behind the saying that Wikipedia is not a democracy.

Appealing to intentions is irrelevant. It's the best way to introduce personal attacks in a discussion, more often than not indirectly. That includes appealing to your own intentions.

Pinging people should be done sparingly, and for the sake of getting better arguments. Not opinions, certainly not opinions about other editors.

Online experience reveals that claims that start with "I" or "You", even when perfectly justified, usually do not improve the wiki. They personalize issues at least a little, and a lot when they are sustained for too long.

Hiding I- or You-claims with, say, requests such as "please stop WP:AGF" (besides usually failing WP:AAGF) or rhetorical if clauses may not increase your chances to look less argumentative. It only makes you look less genuine.

Text

A text is composed of marks and blanks on a page that express ideas and do things. Each editor prefers to work on specific parts of a text: finding le mot juste, pursuing typos, chasing down citation trails, making the page pixel perfect. We all have our niche, each fills a role in the ecology.

I like to read stuff, and I like to find arguments. I like conceptual consistency and when claims squarely fall within the scope of our sources. Above all, I like to express more ideas with as little characters as I can.

Structure matters as much as "content". Readers don't read full page. They want to access specific information.

Transclusions are harder to get right at first, but they're more robust. Excerpts save lots of time.

Chopping

To reduce the size of a page, I start by creating a page in my notes. There I move text and citations that have some value, but may not fit the page. Then I tighten the page structure so as to see what text needs to stay. Moving text around ought to be done after weak material is eliminated, but sometimes resurfaces only after some reshuffling.

Chronological order facilitates paragraph creation. Once a first paragraph is identified, editing becomes easier, if only because one can then alternate between chopping text and omitting useless words.

Omitting useless words

The title should be "cutting", but I am echoing William Strunk's motto.

Direct flow of simple ideas often shortens expression. Too easy to add sideswipes and speculation with secondary clauses. Starting with precise nouns and verbs helps reduce the number of adjectives and adverbs.

For specific cases, see Fluff.

Thanks

Thanking editors for their work should be more common. There are so many good reasons to thank: learning something new, recognizing complementary work that saved time, signaling that you noticed. I am sure there are many more reasons.

Thanks should reflect karmic kudos more than full endorsement. Reward good deeds whatever their correctness. If you afford open pages, editors will come. The more editors feel heard, the merrier the wiki. At least up to the point where everyone steps on each others' toes. That's when cooperation is key.

There are situations when it's best not to thank. They usually involve independence. Which means there are few good reasons not to thank those with competing attitudes.

Themes

An encyclopedia is built with concepts. These concepts can refer to objects, persons or events. They provide the best way to organize entries.

Between two themes, choose one. "A and B" titles indicate indecision.

Titles

Build sections with generic titles, and subsections with specific titles. This way you invite others to contribute while preserving some kind of conceptual order.

Adding titles can help separate the weat from the shaft. Don't get too married to them. They often subsume material that gets cut.

Apply titling style consistently throughout a page. A few dissonant titles can work, for effect.

For job titles, see MOS in Short.

Times

Time spent editing the wiki counts more than the number of edits. Managing the first indicator properly is the best way for the second indicator to increase more steadily over a long period of time. In other words, don't burn yourself out.

Timestamps are from Greenwich.

An entry reaches maturity when New York Times and other inaccessible citations dwindle away from its page.

WP:Proseline should only be seen as stylistic advice. Encyclopedias are bound to provide synoptic views. Announcements ought to be timestamped, at least at first. They can be softened later.

An editor who believes in the ten-years test should consider that thought experiments are not tests, and that they don't know what we'll be eating next month. Predictions are hard, especially when they hide justifying a feeling of irrelevance.

Our wiki lives in eternal time. Things take time, perfection is not required, yet one one must imagine Sysyphe happy to improve. Festina lente.

Topics

To keep the main thing the main thing is key. The best way to do so is to divide it properly. A single and simple criterion needs to separate sections. Each section ought to express that criterion.

Topics are not transitive: a claim about A is not about B because B is what makes A relevant.

Notability is for pages, not sections, paragraphs, or sentences. Topics are not pages. Editors may use their discretion to merge or group two or more related topics into a single article.

Tradoffs

There is no perfect text. Few optimal solutions for writing problems exist. Satisficing is key.

Pages will always get vandalized at some point. You can't preempt it. It's in the nature of the beast.

It's easy to game wiki principles to get one's way. For instance, using notability and originality as a double bind is common. (The dilemma works by requiring and refusing an inference in alternation.) Double binds don't disappear because multiple editors create the bind. Mind literalism.

We are far from being perfect, and all deal with dissatisfaction in our own way. As long as the ball keeps moving forward, all should be well. Every instance of friction provides an opportunity to make the text more understandable by more people.

Trust

The adage Trust but verify can be adapted to the wiki the following way. Trust that editors mean what they say, but always verify it it's warranted. This includes ourselves.

Trust the process too. Ignore all rules, and ignore rule enforcers. Let your magnanimity lead the way.

If you think the material is verifiable, you are encouraged to provide an inline citation yourself before removing or tagging it.

Truth

Wiki truth is verifiability, thus we should be seeking justified beliefs. The more justified our beliefs are, the more circumspect is our expression of these beliefs through our statements, the best chances we got to converge toward truth. There lies our only hope against the entropy of the universe.

We still can go wrong under ideal circumstances. Our human predicament is that things can always turn out to be different than we expect. This does not mean we should qualify everything we say; it only implies we accept that our claims, like our expectations in general, are fallible. Hence the reason we speak of "claim" in the first place.

Turns

Editing is not turn-based: editors edit when they can, and when they want. Sometimes editing is a sprint, sometimes it is a marathon. Without there being no dividing line, the rule of thumb ought to be: more friction, more time.

So play a few moves, then wait for feedback. You can still gather research, both here and elsewhere. This may even improve the quality of your edits.

Each edit needs to stand on its own feet. Think of them as one move, or one improvement to the page. Too many increments at the same time can make other editors worry. Staying away from knights starts with keeping a good editing pace between them.

Do not revert a large edit because much of it is bad, and you do not have time to rewrite the whole thing.

Notes

^ Wikilawyers should note that gamesmanship "may range" from bad faith to simply engineering "victory" in a content dispute.
^ A self-sealing position: (a) assume a claim is contentious as soon as you reject it; (b) state that consensus can't be reached over a claim as long as you find it contentious; (c) shift the burden of proof; (d) refuse to acquiesce to any claim you want.

References

^ Samuelsson 1979.
^ Show 2009 "Solving an open exposition problem means explaining a [...] subject in a way that renders it totally perspicuous. Every step should be motivated and clear; ideally, [readers] should feel that they could have arrived at the results themselves."

Works cited

Chow, Timothy Y. (2009). "A beginner's guide to forcing" (PDF). American Mathematical Society. Retrieved 2025-05-01.

Samuelson, Paul A. (1979). "Why we should not make mean log of wealth big though years to act are long" (PDF). Journal of Banking & Finance. 3 (4): 305–307. doi:10.1016/0378-4266(79)90023-2. ISSN 0378-4266. Retrieved 2025-05-09.