Mills Baker’s answer to Why should designers work at Quora?

Product Design Manager at Quora (2016-present)

If you’re working on an unsolved problem that depends heavily on design, you’ll structure your organization and roles to empower design. This is as true at Quora today as it’s been historically at Apple, with some key distinctions relating to the types of products we make. For us, empowerment means a few things specifically:

• Design reports to the CEO, not to a non-design product executive
• Design is deeply involved in both company and product / team strategy
• Design at Quora is optimized for speed and autonomy, so we can learn extremely rapidly and can’t be blocked by the necessity for endless consensus building; thus designers at Quora code
• Functional groups at Quora take one another seriously: designers aren’t handed “product ideas” and told to draw UIs for them
• We work to make design ever-faster and easier, building new abstractions that simplify work in the product (on both native and mobile)

Designers at Quora can conceive of ideas, build them in the product, test them in our Analysis Framework on millions of users live, and can ship to production at their speed.

Designers Will Code by David Cole on Emesis

Director of Design at Quora

The fact that these components all stay current with live data automates a lot the engineering work. If a designer is making a form that submits an answer to a question, the existing list of answers will show the submission as soon as it’s posted, without anyone writing new code telling it to update. This means that changes on the interaction or visual level usually require no engineering support whatsoever. And because we run constant deployment, and designers can push to production, a Quora designer can make changes and see them live on the site in minutes.

The Wikimedia Language committee clamped down on “dead” languages and artificial languages quite ferociously, after an initial laissez-faire period. Because initially you could set up a Wikipedia in any language you liked, Latin, Old English, Gothic, and Old Church Slavonic got in. Because the Wikimedia Language Committee clamped down, Ancient Greek got rejected even though the proposal for it was far advanced.

Requests for new languages/Wikipedia Ancient Greek

This discussion was created before the implementation of the Language proposal policy, and it is incompatible with the policy. Please open a new proposal in the format this page has been converted to (see the instructions). Do not copy discussion wholesale, although you are free to link to it or summarise it (feel free to copy your own comments over). — {admin} Pathoschild 20:12, 3 December 2006 (UTC)

Same goes for Ido and Volapük and Lojban, which got in before (Esperanto would be hard to argue against), and Toki Pona or Klingon, which didn’t make it. Klingon being a pop culture conlang, it attracted disproporionate negative attention, including Wales personally wielding the axe against it:

History of the Klingon Wikipedia

In August 2005, Jimbo Wales made a decision to lock the Klingon Wiki permanently. While Jimbo has never publicly stated his exact rationale for closing the wiki, the maturation of Wikipedia and its sister projects as a whole into a vital worldwide resource meant that there was little incentive to keep a niche language that was not intended to be seriously used. Other constructed languages such as the Toki Pona language were closed at about the same time (although the Toki Pona Wikipedia, like the Klingon Wikipedia, was ultimately hosted at Wikia due to the presence of a strong community).

In fact, the Klingon letter r was removed from the Wikipedia logo in 2010, replaced by a Ge’ez character.

Old logo; Klingon <r> top right.

New logo

Just happened to me as well today.

It’s happened periodically before.

Assume bug.

For now.

Khateeb, I have no idea, but I can surmise based on:

• The Death of the Urdu Script – ali eteraz – Medium
• Nastaʿlīq script – Wikipedia

If your technology is handwriting, it doesn’t particularly matter whether your writing is vertical or horizontal, or a mix of both.

If you’re writing online in 2017, and you want to use a vertical script like Mongolian… well, read Nick Nicholas’ answer to Why doesn’t Mongolia use the Uighur script again and leave out Cyrillic?

And Nastaliq goes both vertical and horizontal. If you read the Wikipedia page, metallic, traditional typesetting of Nastaliq has been a non-starter for that reason. Digital typesetting in theory should be easier, but of course in practice it is a hassle, especially if you can just use Naskh as an alternative.

Wikipedia says that the InPage custom Desktop Publishing software, which exists to do Nastaliq, is extensively used now for Urdu. For publishing, maybe; the Medium blog above shows how little penetration Nastaliq has had among laypeople online, and how grateful they were that Microsoft started supporting Nastaliq at all in 2011.

Khateeb, you’ll have to tell me how widely Nastaliq is seen in Urdu. You asked though about Persian. If I interpret Wikipedia correctly, while Nastaliq originated in Persia, its use in Persian is limited to poetry; Pashto uses both Naskh and Nastaliq, and Kashmiri, Punjabi and Urdu—and Ottoman Turkish—used Nastaliq. For whatever reason, it seems that Nastaliq flourished as an everyday script, rather than a calligraphy-only script, only east of Persia. Possibly because Persia neighbours Naskh territory (Arabic), and Urdu neighbours Persian, not Arabic.

So, if I had to guess why Persian sticks with Naskh: combination of technical difficulties, and not a strong enough identification with Nastaliq to bother surmounting the technical difficulties (unlike Urdu).

Originally Answered:

Quora now removes the number of upvotes before one clicks it. Why don’t they also remove the number of views and who upvoted it?

Seeing many Quorans that I am following, it’s not hard to see the upvotes must be over 1000, which I believe contradicts to the original intent of not letting users see the number of upvotes before you click it. Also, I can see the numbers of views and comments; it’s not hard to guess a viral answer.

Well then, I guess they will get rid of those now too. Thanks a bunch, OP.

The removal of social media functionality over the last couple of months has been incremental. I wouldn’t be surprised if the Views preview went away before you expand out an answer—in fact, the whole line would need to go in the preview mode. Hiding it in all circumstances, OTOH, would be… counterproductive: if you do want to find out how many people viewed the answer or upvoted it, do you really want to do three clicks, or view it only if you’ve upvoted the answer yourself?

Killing the comments count, I would *hope*, they would be more reluctant to do: that goes to usability.

Stop laughing.

Well, let’s think this through.

I count three programming paradigms from when I was studying computer science 25 years ago: functional, logical, and procedural. They correspond to three types of semantics: denotational, axiomatic, and operational. The first two are pristine and beautiful articulations of mathematics and logic, respectively. The last involves modelling the internal state of a von Neumann machine, and is so ugly, it’s embarrassing to express it in mathematical notation at all.

By default, I would say denotational semantics comes closest to how formal linguists think of linguistic semantics. It uses statements about the world to formulate a model of the world. And that’s what humans do, in making assertive speech acts.

The catch is, the reason we interact with computers through programming languages is not assertive: we are not (yet) having a debate with the computer about moral philosophy or fiscal policy, for example. It is directive: we are trying to get the computer to do things for us.

The most straightforward way of directing an action is by issuing directive speech acts. We couch our commands to other people as questions and statements out of politeness, but we don’t really do that with a computer. And even when we are doing functional or logical programming, we are not truly authoring mathematical or logical proofs. We are using the mechanisms of those proofs, to get the computer to do something. What we choose to express in those paradigms is still driven by that goal.

(I did a semester project writing a parser in Prolog. My most important learning? Logic programming is still programming.)

So while functional programming is probably closest to how we think of natural language in general, procedural programming is closest to how we think of language in the context of human-computer interaction.

Complimentary, but not deep.

The interwebs widely quote Chomsky saying in Kolkata, in a 10-minute speech in 2001, “The first generative grammar in the modern sense was Panini’s grammar”: An event in Kolkata. Chomsky in fact already said that in the preface of Aspects in 1965: “a generative grammar, in essentially the contemporary sense of this term” . And the final line of The Sound Pattern of English is an allusion to the final sutra of Panini, “ā → ā”.

But as Kiparsky and Staal noted in 1969, Panini’s model of generativity has very different constructs between the two levels: they are derivations, not rewritings. An anecdote on Chomsky’s linguistic theory suggests that Chomsky ended up going back to an approach more like Panini’s with Minimalism in 1991—and might have saved himself 26 years if he’d read Panini more closely when he was citing him in Aspects.

But of course, Chomsky wasn’t learning from Panini, or honing his craft against the Indian master. Chomsky came up with transformations on his own, and merely found it convenient occasionally to allude to Panini as an antecedent. Panini and Chomsky were not undertaking the same research programme, after all.

Anvil Chorus (Verdi: Il Trovatore, chorus from act 2, scene 1). 1:02 in the attached video.

https://youtube.com/watch?v=WEMMVHAINFM

May John Clarke (satirist) rest in peace.

A Q&A site like Quora, with more “facts”, and even fewer “non-facts” (e.g. less sociability)?

Stack Exchange. Quite Savage in its culture of Just The Facts, but overall very reputable, aggressively moderated by its users, and its programming forum Stack Overflow is the canonical place for programmers to go for advice.

“So if you hate Quora so much, why don’t you go there?”

Because it’s Quite Savage in its culture of Just The Facts. I stay here because, the best efforts of Quora’s UX team notwithstanding, it is still possible to socialise here.

Ptolemy, at least, expressed them somewhat clumsily, by adding reciprocals. There were dedicated symbols for half: [math]unicode{x10175}, unicode{x10176}[/math], two thirds: [math]unicode{x10177}[/math], and three quarters: [math]unicode{x10178}.[/math] Outside of those, fractions were expressed by using double prime for reciprocals, ″.

So Ptolemy used ιβ″ = 1/12 a lot for geographical coordinates; and he would also use expressions like γ″ιβ″ = 1/3 + 1/12 = 5/12 or [math]unicode{x10175}[/math]ιβ″ = 1/2+1/12 = 7/12.

---

EDIT: For those without font support for Ancient Greek Numerals: