Mollyycolllinss Categories For later read

7 days 30 days All time Recent Popular
It is difficult to change a 10-year trend.

Long-term expectations do not change as frequently as daily market fluctuations would make it seem.

A quick update on Treasury rates through the lens of the DKW model

*As of Dec. 31*

1/

In previous threads, I made the distinction between long-term secular trends in growth and inflation and shorter-term (2-6 quarters) trends in nGDP


Right now, the long-term trends are unaltered because long-term trends just don't change that fast but we have a very strong cyclical upturn in the economy, centered primarily on the shift to goods consumption bolstering the manufacturing sector and industrial commodities.

3/


As long as the industrial sector continues to roar, TSY rates will have an upward bias as rates generally follow the trend in nGDP growth

A 10yr TSY has longterm expectations embedded in the rate so several qrters, while important, won't necessarily change the longterm trend

4/


This is confirmed by the Dec update to the DKW model which breaks down *actual* inflation expectations, the expected real short-term rate (real growth), term premium, liquidity premium etc.

The DKW model is one of many models that is useful but has many limitations.

5/
Merry Christmas everyone with a new #WednesdayWagner instalment. Today: TRISTAN CHORD IN 'DIE WALKÜRE'! What happens next will surprise you
#Christmas #Wagner #Tristan #Walkure #TwitterCultural


Yesterday (all my troubles seemed...). No, seriously. Yesterday, I made a discovery that blew my mind and I am so happy to share it with you. Let's start by listening to Act 2, Scene 5 of 'Die Walküre' in a superb performance by Böhm in

The scene begins with the 'Fate' leitmotif, which was already presented in the thread below ⬇️Note that harmony goes from a minor chord to a dominant chord just 1 semitone


This sequence of chords are used in different keys, with different configurations, even as the closing part of the 'Annunciation of Death' motive (this can be listened to at 0.33 in the link above ⬆️)


The last bars of this 'Annunciation of Death' contains the 'Fate' motive with a slight variation of harmony. Notice that we can almost hear the Tristan Chord there (marked in red in the figure below).
This thread attempts to cast light on how the mgmt of #UBCLite worked in lockstep with the EC to mislead, misdirect, deceive and captivate viewers.
FYI, #UBCLite was coined by Tom Ddumba after NBS consistently became a state mouthpiece. The "adjective" fits like a glove 0/13


1/13 Take note that prior and during the elections the EC repeatedly announced that Absolutely NO ONE was allowed to have a tally center, not even the candidates. But somehow NBS was able to ignore those directives and implement what appeared to be a live tally center. Why? How?

2/13
Now the station reportedly have Monday staff meetings. On the Monday of the election week, staff were notified about an election tally center going to be hosted at their station by a selected few (the special ones)- you know these people, they stick out like a sore thumb

3/13 On 15/1/21, when Canary took us through the infamous tour of the "command center", showing us their agents, and how their tally system was using phone calls to aggregate results, it was meant to build confidence in what they wanted to announce. They primed us for fake news

4/13 Interestingly, NBS actually had real staff deployed across the country to collect & relay actual tallies from districts. They collected results & called them in using phones, so on the 15th the teams down stairs were busy working out totals for different districts
This is what happens when you train neural networks largely on tone and its stylistic relics. They pick up formal features of arguments (not so much fallacies as tics) that have almost nothing to do with semantic content (focus on connotation over implication).


This is a secular problem in the discipline. It's got nothing to do with the Analytic/Continental split in the anglophone world. They've both got the same ramifying signal/noise problem, it's just that the styles (tics and connotations) are different in each pedagogical context.

And this is before we start talking about tone policing and topic policing, which are both rife and essentially make the peer review journal system completely unfit for purpose, populated as it is by a random sampling of pedants selecting for syntactic noise over semantic signal.

We've allowed a system of self-reinforcing and ratcheting filters to evolve that effectively *fuzzes* our contribution to the growth of human knowledge (https://t.co/VmW15pGt7J), because it selects for properties only loosely related to those we claim to want. Let that sink in.

This is literally the opposite of what a filter is supposed to do: extract signal from noise, syntactic compression that preserves semantic content. Instead we are awash in syntactic artifacts optimised for minimal criticisable content and maximal pedantic posturing.
Amazigh (Berber) languages are quite close to each other and in most places nearby varieties are mutually intelligible. They function like a discontinuous dialect continuum.

A loooong thread with maps (and no memes☹️).

So can we do sub-classification?

Many people say “no”, like André Basset’s famous quote: “cette langue s’éparpille directement ou à peu près en une poussière de parlers de 4 à 5 mille peut être” (1952:1) and Alfred Willms (1980). Others are a bit more nuanced.

For example, the fantastic studies by Lafkioui (
https://t.co/hHKMgEPjK2) give a synchronic classification of Tarifiyt dialects. To cut a very long story irresponsibly short: all variables are counted the same.

However, we should ask ourselves: Does this continuum hide a more discontinuous past? Has there never been major disruption, or has much of it been smoothed out by later convergence?

In order to study this, one has to classify variables and their isoglosses. Some variables are continuous and can be assigned to the latest convergence period. Others are clustered in a group unrelated to the continuum. Still others are scattered.
Pompeo is a perfect example of this thing that just baffles me. He owes America *everything.* His grandparents immigrated from dirt-poor regions of Italy at the turn of the C19th. He graduated *first in his class* at West Point.


He has a law degree from Harvard. He was the editor of the Harvard Law Review. Went to DC and joined a blue-chip firm. Also made a fortune in private enterprise. America gave the grandson of dirt-poor immigrants the opportunity to do all of that.

He knows *damned well* what the Constitution says.

He's also seen enough of the world to know *damned well* how lucky he was to grow up in a country at peace, governed by that Constitution--and to know *damned well* what happens to a country in a civil war

--which is exactly the brink to which Trump has taken the United States, only recently the inarguable leader of the free world. We now have a GOP that thinks, "storming the capitol in an attempt to assassinate our elected leaders" is somewhere between "great" and "no biggie."

He knows *damned well* what the world thinks of us now, and how endangered we are because of it--and not just us, our generation, physically, but the idea of government of the people, by the people, for the people. It could, truly, perish from this earth-
Have you ever worked with pipelines in Laravel?!

Probably you have heard it while having an interview?

It has lots of use cases and one of its usage is middleware implementation in its kernel.

I'm going to write about pipelines and its design pattern.

thread 🧵

1/8

First of all Pipelines in Laravel are implemented based on chain of responsibility design pattern.

In chain of responsibility which is a behavioral design pattern we pass data to receivers and this receivers implement a specific interface that has a handler method.

2/8

Handler methods may deal with the passed argument and pass it to the next handler.

read more about the design pattern here :

https://t.co/XrJKXecyTv

Let's go for the main part :

3/8

Imagine we want to implement a twitter bot that gets some tweets around the trends and filter the words that we don't wanna show finally send as a new tweet from our bot.

About the filtering part, if it was me, I would implement it with pipelines.

Why?!

4/8

Currently I know exactly what words should be filtered but what about the future?! What if I want to add more filters, I will have to modify the source code and add more switch, if else statements, etc.

I want my code to be robust and maintainable keeping SOLID principles.
An extremely important point that seems to be completely missed in the discussion of disinformation.

We are exclusively focused on the *supply chain*. We neglect the *demand* for this content.

The truth is that people searched for an excuse, or opening, to be radicalized.


I find the comparison to drugs and addiction helpful here, and it is one that I do carefully.

100 people to to the doctor and get an opioid after a procedure. About 95 will never use again.

A lot of us are exposed to extremist content. Most of us don’t get radicalized.

So while the supply and exposure played a role in both addiction and radicalization, it is not magic. It doesn’t just take over people. It taps into demand.

So maybe the question is less about how they got the supply and more why they find it so appealing.

That’s tougher.

It is much easier to imagine an immediate policy solution to Trump’s twitter account or YouTube’s auto play than to the coercive impact of 401 years of America apartheid and racist myth making.

But again drugs suggest that those quick supply cuts don’t work.

A century of drug busts made the problem worse. This observation led to the iron law of prohibition: the harder the enforcement, the harder the drugs.

When you cut supply, people who want drugs *will* find drugs — and often more dangerous drugs.

This is where my insight ends.