Chris Brew [email protected]

@40PercentGerman Nothing to do with grammar. It’s perfectly good grammar to use the “wrong” one, just rude.

One part of German I've struggled with is whether to address people with Sie or Du. Germans have internalised it to such an extent they're awful at explaining when to use them. Luckily, I found this, but how simple can grammar be when it needs a flowchart to explain?

@skdh In spoken language, German modal particles translate into variations in intonation. ‘Doch’ translates into the marked intonation contour in “you CAN do that” . You need something to put the intonation in, of course, so it is not as neat.

Instead of "Guten Tag" you can use "Schönen Tag noch". The word "noch" in that phrase acts as a temporal modal particle and is difficult to translate, it roughly means "have a pretty day afterwards". English doesn't seem to have modal particles for all I can tell.

"Guten Tag", literally "good day", is the German any-purpose greeting. You probably knew that. Less well known is that if you use it in case it's obviously not a greeting -- because you've already greeted the person -- it means either "fuck you" or "piss off", depending on whether you are leaving, or want the other person to leave. Example 1: You have an argument with a co-worker in their office and it's not going anywhere. You say "Guten Tag." and slam the door behind you. Example 2: Your co-worker came to your office to shout at you. You say "Guten Tag" and stare them down. It's their cue to shut up and leave. I've never used this in my life until recently when a journalist rang on my door unannounced to ask for an interview.

@davidfrum Have you heard of Joschka Fischer? He was a violent radical, then Foreign Minister of Germany. I guess you must think that is not a leadership position.

3) The "leaders of tomorrow" will be found among those going to work or class today, not among those who went to jail for their violent disrespect for law and the rights of others.

1) In 1968, Tom Hayden was perhaps the most famous student protest leader in the country, In 1968, Sam Hayakawa of San Francisco State was the college president most conspicuous for confronting campus disorder. ...

@TheSpinDr7 @GadSaad I don’t think my Western compatriots are serious about their culture or identity, but they are very serious about the privilege and power that has come with them for the last 300 years.

In 2013, I visited Oslo to deliver a couple of invited lectures. When I arrived at the airport, it was beautiful to see that it was akin to having just landed in Mogadishu. No native Norwegian in sight. I asked my Norwegian host about some of the very real problems that Norway was having with some of its Noble Immigrants. He answered by saying that this was the fault of the Norwegians for not doing enough to help them better assimilate. I did not challenge him in order to be respectfully diplomatic. Suicidal empathy. Get ready for my next book.

British Army Humour An innocent typo was made and all parties involved have gotten as much enjoyment as they could out of it! Lt. Colonel Robert Maclaren retired from the British Army in 2001 after a long fulfilling career. On the day that he retired he received a letter from the Personnel Department of the Ministry of Defence setting out details of his pension and, in particular, the tax-free ‘lump sum’ award, (based upon completed years of service), that he would receive in addition to his monthly pension. The letter read, "Dear Lt. Colonel Maclaren, We write to confirm that you retired from the Royal Scots Dragoon Guards on 1st March 2001 at the rank of Lt Colonel, having been commissioned into the British Army at Edinburgh Castle as a 2nd Lieutenant on 1st February 1366. Accordingly your lump sum payment, based on years served, has been calculated as £68,500. You will receive a cheque for this amount in due course. Yours sincerely Army Paymaster” Colonel Maclaren replied; “Dear Paymaster, Thank you for your recent letter confirming that I served as an officer in the Royal Scots Dragoon Guards between 1st February 1366 and 1st March 2001 – a total period of 635 years and 1 month. I note however that you have calculated my lump sum to be £68, 500, which seems to be considerably less than it should be bearing in mind my length of service since I received my commission from King Edward III. By my calculation, allowing for interest payments and currency fluctuations, my lump sum should actually be £6, 427, 586, 619. 47p. I look forward to receiving a cheque for this amount in due course. Yours sincerely, Robert Maclaren (Lt Col Retd)” A month passed by and then in early April, a stout manilla envelope from the Ministry of Defence in Edinburgh dropped through Col. Maclaren’s letter box, it read: “Dear Lt Colonel Maclaren, We have reviewed the circumstances of your case as outlined in your recent letter to us dated 8th March inst. We do indeed confirm that you were commissioned into the Royal Scots Dragoon Guards by King Edward III at Edinburgh Castle on 1st February 1366, and that you served continuously for the following 635 years and 1 month. We have re-calculated your pension and have pleasure in confirming that the lump sum payment due to you is indeed £6, 427, 586, 619. 47p. However, we also note that according to our records you are the only surviving officer who had command responsibility during the following campaigns and battles; The Wars of the Roses 1455 -1485 (Including the battles of Bosworth Field, Barnet and Towton) The Civil War 1642 -1651 (Including the battles Edge Hill, Naseby and the conquest of Ireland) The Napoleonic War 1803 – 1815 (including the battle of Waterloo and the Peninsular War) The Crimean War (1853 – 1856) (including the battle of Sevastopol and the Charge of the Light Brigade) The Boer War (1899 -1902) World War One (1914-1918). We would therefore wish to know what happened to the following, which do not appear to have been returned to Stores by you on completion of operations: 9765 Cannons 26,785 Swords 12,889 Pikes 127,345 Rifles (with bayonets) 28,987 horses (fully kitted Plus three complete marching bands with instruments and banners. We have calculated the total cost of these items and they amount to £6,427,518.119.47p. We have therefore subtracted this sum from your lump sum, leaving a residual amount of £68,500, for which you will receive a cheque in due course. Yours sincerely . . . .” 😁😁😁

@JujuAadi @AravSrinivas Odd view of simplicity. With the right database your complex one is easy SQL. Your simple one requires law school or equivalent to do well.

@AravSrinivas I absolutely love using Perplexity for queries as complex as "best rajasthani restaurants in houston within a 3 mile radius of 77007 zip code" and as simple as "summarize NYT's lawsuit against chatgpt"

A lot of people ask about Perplexity.ai: why do we exist? Google would just eat you and do the same. How dare you think you have better engineers than Google? Well, we’ve never ever said we have better engineering. I’m a fan of every OG Googler and the founders. The reason we exist is because actually improving search and making information access a lot more efficient with fewer keyword queries, less sifting, link clicking, page viewing; and directly getting to the bottom of things and making a decision with a succinct clean personalized answer, is at odds with the financial and business goals of Google, and that’s the opportunity for someone else to start from a blank slate and rethink the product and business from scratch. The world wants that. Because there’s an overload of information and there’s limited time to ingest them. That’s what Perplexity strives to do. Excepts from Google’s discussion on why making search worse and having users ask more queries is better for them, making the search and ads team misaligned.

@alexandersclark This view pretty definitely shaped by early exposure to dynamic logics as arbitrarily general semantic formalisms. You can do anything, but it doesn’t mean you should.

@alexandersclark I don’t think one can ever say that a natural language is non-compositional, because there is always the alternate possibility that it is compositional but you haven’t yet found the right decomposition into parts.

I was thinking about this a bit: what is your favourite example of a *non-compositional* language? (Resist the temptation to just say "English")

@ProfMaxNew If you are talking of NL parsing the issue is that parsing generates a representation that most obvious use cases don’t need. They can jump straight to downstream products like (heavy quotes) ‘meaning’

When did we as a CS community decide that parsing was a "solved" problem? It's a very common sentiment so I wonder if it originates from a source we can actually identify

@AlwaysUhhJustin @random_walker I’ve also talked to the Microsoft people who support company deployments of 3.5. They haven’t been told much about what OpenAI’s plans are, apparently. The two companies are not (yet??) sharing all the detail with each other. Maybe a result of OpenAI’s charter.

I don't think much of the speculation is accurate. I met with Microsoft to implement GPT-3.5 at my company 2 weeks ago. They transparently didn't know GPT-4 release date (but said I could piece together a multi-modal option). And they said they didn't yet have access to true ChatGPT but had GPT-3.5 that we could use for now (DaVinci003 iirc)

Considering that OpenAI did a decent job of filtering ChatGPT’s toxic outputs, it’s mystifying that Bing seemingly decided to remove those guardrails. I don't think they did it just for s***s and giggles. Here are four reasons why Microsoft may have rushed to release the chatbot.

@matbury @adelegoldberg1 It would certainly fail formal language theory.

@adelegoldberg1 It would get good grades on linguistics exams!

@DCCyclone @mixedlinguist @nytimes I think I agree that he deserved to be mocked. What he did was to unintentionally reveal how he was thinking in trying and failing to get it right.

@cbrew @mixedlinguist @nytimes Sorry but that guy was a moron and deserves every bit of public flogging.

@stephlepp Readers, deprived of the easy signals that used to indicate that a piece is not worth their time, may not have the patience to read deeper and notice the difference between “good” and ‘looks good”. Expect lots of fake David Foster Wallace.

@cbrew So then what do you mean that 'good work will go undetected'?

There seems to be an implication that ChatGPT lowers the bar for creation. If anyone can make almost anything, then what's the meaning of making? I think it raises the bar. If anyone can make almost anything...then what are ✨you✨ gonna make?

@stephlepp No. But I am assuming that input by humans who have something to say is needed to produce good work. I’m sure we’ll see excellent stuff done by people using AI models to augment what they do.

@cbrew Are you assuming that "good work" can only be done by humans with no help from technology?

@mixedlinguist @nytimes Yep. And not everyone accepts the standard linguist view that some of this is just arbitrary. When we are deep in it emotionally, that view doesn’t feel right, even if it is correct, which I think it is.

@sitKonrad @emilymbender @Sonja_Drimmer Yes. Some of my colleagues have worked out ways of getting it to provide checkable citations. Sometimes. But ad hoc, messy and untransparent

@emilymbender @Sonja_Drimmer I tried. It told me since it’s crow sourced it couldn’t give proper attribution.

We're seeing multiple folks in #NLProc who *should know better* bragging about using #ChatGPT to help them write papers. So, I guess we need a thread of why this a bad idea: >>

@betsysneller The chicken is ready to eat. Ambiguous, though might not be syntax, could be semantics or even semantax . Carl Pollard’s word for the between thing of syntax and semantics.

hey syntax friends, what are your favorite crowdpleasers for explaining some things about syntax? Think: you're at a party and talking to a nonlinguist who is genuinely interested but doesn't have any background

@compositiomal Plausible claim. Some syntacticians don’t believe in movement. Do they do the gesture?