sradman

Indeed, the age group for cerebral venous sinus thrombosis (CVST) is very specific [1], mostly women < 55 (menopause?):

> The Committee’s experts looked in extreme detail at records of DIC and CVST reported from Member States, 9 of which resulted in death. Most of these occurred in people under 55 and the majority were women. Because these events are rare, and COVID-19 itself often causes blood clotting disorders in patients, it is difficult to estimate a background rate for these events in people who have not had the vaccine. However, based on pre-COVID figures it was calculated that less than 1 reported case of DIC might have been expected by 16 March among people under 50 within 14 days of receiving the vaccine, whereas 5 cases had been reported. Similarly, on average 1.35 cases of CVST might have been expected among this age group whereas by the same cut-off date there had been 12. A similar imbalance was not visible in the older population given the vaccine.

This seems to be yet another case where the news is incapable of keeping up with timely technical information. No mistakes were made, the pause could have gone either way and we should generally be more understanding of difficult but timely decisions made in the face of uncertainty.

[1] https://www.ema.europa.eu/en/news/covid-19-vaccine-astrazene...

> They focused on the people because they were either scammers

The scamming came after the failure of the core premise of the startup: microfluidic lab-on-a-chip. My point was not to excuse the scamming, it was to switch focus back to the core ideas and not the people sporting black turtlenecks.

I suspect that if the original strategy worked the story would be different. Transparently reporting technical/business failure is an admirable trait but it is a different thing than innovation. Journalists and analysts should have been reporting on the scorecard (strategy/tactics/execution). When the company did a technical pivot why was this not enough to investigate the underlying cause? I'm a potential customer that wants this solution; I feel like I've been cheated by all involved.

QUESTION: if someone else succeeds with a microfluidic lab-on-a-chip can the regulators validate its efficacy?

At the heart of progressive thought is concern over the twin evils of oppression and exploitation. O'Reilly’s piece focuses on the latter but relies on the assumption that profit==exploitation. I’m not convinced it is that simple.

I wish I could have had a chat with Milton about the Friedman doctrine [1]. I want to give him the benefit of the doubt. From my perspective, shareholder value is a relatively easy measure that acts as a proxy for the true goal of a corporation: creating sustainable customer value.

Within the context of his original article, Friedman was arguing that customers, employees, and shareholders could choose to give to charity, for instance, but it is wrong for a corporation to do it on their behalf. The concept is not intuitive but it is far from promoting selfishness.

[1] https://en.wikipedia.org/wiki/Friedman_doctrine

> for many of the tests Theranos was aiming to do there's no real way to achieve it with the tiny volumes of blood they were trying to use

This is exactly the right question to ask. Right from the start we should have had a scorecard of the common lab tests doctors request (under 50 I think). For each test you should ask “is it possible with microfluidics?” and then “has it been combined on a single strip/card yet?”

I’m not sure if any concentrations are too low for microfluidics rather than the dilution required to meet the volumes specified in the standardized tests. Are fingerpick volumes too small or are the regulatory tests outdated?

The same issue applies to self administered rapid antigen tests during the pandemic. Regulators focused on the “gold standard" PCR tests and ignored the benefit of cheap and easy unamplified tests that can be used in a daily regiment.

> as Theranos demonstrated so vividly, it is harder to sustain a hype balloon in a scientific enterprise than in many of the markets where Silicon Valley has prospered.

This narrative continues to frustrate me. Eleanor Roosevelt famously said:

> Great minds discuss ideas; average minds discuss events; small minds discuss people.

The big idea behind Theranos was that combining multiple microfluidic tests could produce a lab-on-a-chip. Theranos failed to combine more than a handful of microfluidics together and scrambled to find alternatives to bridge the gap.

Ultimately they were foiled by brain-dead regulatory tests that specified blood draw volumes of blood. This regulatory test has to change to accommodate fingerpick volumes of blood; a key feature of microfluidic tests.

Our collective small minds couldn’t focus on anything but the people involved. The core idea is still an outstanding hard problem.

Great data, mind boggling interpretation. This is exactly what I’d expect to see in Greenland’s subglacial sediment. It reinforces, not rewrites, the paleoclimate history during the 41 kyr to 100 kyr glacial cycle transition.

> Was Nikola Tesla a failure?

He was a brilliant innovator that added value to the world. He was a failed entrepreneur that did not recapture part of the value he created. That makes him a huge success in my books but I wish he was more comfortable and content in his final years.

SQLAlchemy 1.4.0 Released – featuring major rethink of the APIs in Core and ORM [1] HN thread from yesterday and Migrating to SQLAlchemy 2.0 [2] from 26 days ago.

[1] https://news.ycombinator.com/item?id=26473654

[2] https://news.ycombinator.com/item?id=26182744

The climate switched from 41 kyr glacial cycles to 100 kyr cycles about a million years ago. Greenland experienced regular interglacials during the 41 kyr cycles. None of this is surprising except that climate change obsessed scientists, journalists, and HN commenters appear to be unaware of the Marine Isotope Stages.

The sediments below the western and eastern Antarctic ice sheets should show similar signs of interglacial activity timed to their own transitions to permanent glaciations.

Written by Thomas Chatterton Williams [1], author of Self-Portrait in Black and White [2]:

> “It’s not about color for me,” my aunt said while railing against Obama. “For example, I love Thomas Sowell.”

> To that side of my extended family, I became the stereotype of a coastal liberal, writing for the New York Times and wholly out of touch with the real America. In fact, I’ve always prided and defined myself as an anti-tribal thinker, and sometime contrarian, working firmly within a left-of-center black tradition...

Why was this published in Law & Liberty, has he burned his left-of-center media bridges?

[1] https://en.wikipedia.org/wiki/Thomas_Chatterton_Williams

[2] https://en.wikipedia.org/wiki/Self-Portrait_in_Black_and_Whi...

European AstraZeneca supply is not being shipped to the USA. The FDA will not approve AZ until the ongoing local phase III trial is complete. In the meantime, the U.S. export restrictions still apply to American manufactured AZ doses:

https://www.nytimes.com/2021/03/11/us/politics/coronavirus-a...

British Columbia's Dr. Bonnie Henry discusses [1] (5min YouTube clip up until 28m37s) the decision process surrounding Adverse Events Following Immunization (AEFIs) [2] within the context of the EMA decision involving the AstraZeneca vaccine. Her explanations tend to mirror her refrain "Be Kind, Be Calm, Be Safe".

For those jurisdictions like the UK, India, and Canada that continue recommending AstraZeneca, I'd suggest that COVID-19 vaccines are like beer; the best one is the one in front of you.

[1] https://youtu.be/xbLJCh9XHl0?t=23m42s

[2] https://en.wikipedia.org/wiki/Vaccine_adverse_event

> The EMA has said that as of March 10, a total of 30 cases of blood clotting had been reported among close to 5 million people vaccinated with the AstraZeneca shot in the European Economic Area, which links 30 European countries.

"The decision today is purely precautionary..." given this level of signal. We don't have details on the age groups involved and the normal rates expected but I can hazard a Fermi Estimate that the risk is minuscule compared to COVID-19 itself. I look forward to seeing the actual data in coming days/weeks. Precaution without downside is acceptable; this is not one of those cases, IMO. YMMV.

Cornelius Vanderbilt [1] mastered steam powered shipping before tackling rail. An operator of technical machinery more than a trained engineer but relatively very technical for the 2nd Industrial Age IMO.

[1] https://en.wikipedia.org/wiki/Cornelius_Vanderbilt

Software is eating the world but you are discounting the possibility that this force is simply drawing in the best and brightest; part sorting hat and part black hole pulling in the biggest stars.

Pandas supports JOIN and GROUP BY operators so you are saying that there is a gap between Apache Arrow and other mature dataframe libraries? If there is a gap, is there no plan to fix it in the standard Arrow API?

I understand the case for a SQL-like DSL and an optimizer for distributed queries (in-memory column stores, not so much). I'm trying to understand the value add of Polars. I don't mean to come across as critical; perhaps DataFusion is a poor implementation and you are being too polite to say so.

I also think that there is a C++/Arrow vs Rust/Arrow decision that has to be made. I associate PyArrow with the C++/Arrow library. Is Polars' Eager API a superset of the PyArrow API with the addition of JOIN/GROUPBY/other operators?

My naive interpretation - The canonical Apache Arrow implementation is written in C++ with multiple language bindings like PyArrow. The Rust bindings for Apache Arrow re-implemented the Arrow specification so it can be used as a pure Rust implementation. Andy Grove [1] built two projects on top of Rust-Arrow: 1. DataFusion, a query engine for Arrow that can optimize SQL-like JOIN and GROUP BY queries, and 2. Ballista, clustered DataFusion-like queries (vs. Dask and Spark). DataFusion was integrated into the Apache Arrow Rust project.

Ritchie Vink has introduced Polars that also builds upon Rust-Arrow. It offers an Eager API that is an alternative to PyArrow and a Lazy API that is a query engine and optimizer like DataFusion. The linked benchmark is focused on JOIN and GROUP BY queries on large datasets executed on a server/workstation-class machine (125 GB memory). This seems like a specialized use case that pushes the limits of a single developer machine and overlaps with the use case for a dedicated column store (like Redshift) or a distributed batch processing system like Spark/MapReduce.

Why Polars over DataFusion? Why Python bindings to Rust-Arrow rather than canonical PyArrow/C++? Is there something wrong with PyArrow?

[1] https://andygrove.io/projects/

Divers call them thumb splitters for a reason.

Why this hybrid dataframe? [1]:

> [Arrow] adoption will take time, and most people are probably more comfortable seeing NumPy arrays. Therefore a Vaex version 4 a DataFrame can hold both NumPy arrays and Apache Arrow arrays to make the transition period easier.

There seems to be agreement that Apache Arrow is the future of dataframes across ML ecosystems. I didn't realize this transition impacted NumPy arrays in addition to Pandas dataframes in Python.

[1] https://vaex.io/blog/a-hybrid-apache-arrow-numpy-dataframe-w...

How? Subsea power cables, optical fiber, and pipelines (?). Midsize autonomous ships providing a cost effective alternative to truck and rail transport. Promotion of English as the lingua franca. Policies that attract new talent and promote the free movement of goods and people between new coastal charter cities. Partnering with people in the same boat (or sea).

Adam Smith not only promoted specialization but also extending the "reach" of trade. Politicians have the power to ruin things but they only succeed when riding the coat-tails of talented makers. I'd focus on promoting the makers rather than punishing past ruiners. Nihilism is never the answer.

The geography of southern Italy, Croatia, and Greece place all three at a disadvantage compared to continental nations connected via road, rail, and canal. The Mediterranean is a comparative advantage that can be leveraged. The natural beauty that attracts tourism can help repatriate the talented diasporas.

The question is whether brain drain, crime, and corruption are due to incurable pathologies or symptoms of transient disadvantages.

Wow, I'd like to hear more about the subsea cable [1] technology used. In terms of economics and geopolitics, this sounds like a win-win scenario. Perhaps necessity will force the eastern Mediterranean to re-emerge as an economic powerhouse. Southern Italy, Croatia, Turkey, and Egypt should pay attention.

[1] https://en.wikipedia.org/wiki/Submarine_power_cable

Good stuff, hugged to death [1]:

> In the end, code doesn’t make software – people and communities do... It’s hard to beat an incumbent, and even harder to do so without having a large target user community you can capture with a compelling use case.

We need a better name for the category of Machine Learning Workbench tools that include Julia, R, and NumPy+ (the ML ecosystem that uses Python as an Internal DSL). The compelling use case for Julia is that it is a modern language built on top of the LLVM platform (like Rust and Swift). There is no shame in not winning when you were late out of the gate; see the story of ARM. The value proposition hasn't disappeared it just didn't outpace the evolving bricolage of NumPy+ and its new complementers (Jupyter, PyTorch, TensorFlow).

[1] http://webcache.googleusercontent.com/search?q=cache:Awfh5p8...

This article by the author/podcaster Seth Godin [1] highlights the issues with Non-Fungible Tokens (NFT) [2], built on the Ethereum cryptocurrency. It extends the ideas introduced in an Everest Pipkin post [3]:

> Cryptocurrencies and NFTs are an absolute disaster for so many more reasons than the ecological.

The ecological issue is due to the energy consumption required to perform (some^) cryptocurrency computations.

[1] https://en.wikipedia.org/wiki/Seth_Godin

[2] https://en.wikipedia.org/wiki/Non-fungible_token

[3] https://everestpipkin.medium.com/but-the-environmental-issue...

^EDIT: added “(some)” cryptocurrency computations based on Tenoke’s comment

0data appears to be a curated collection of five best-in-class libraries/frameworks that enable building SPA apps with user controlled data, rather than Big Tech controlled platforms and services.

1. Autonomous Data

2. remoteStorage

3. Solid Project

4. Unhosted

5. Fission

[1] https://noeldemartin.github.io/autonomous-data/introduction....

[2] https://remotestorage.io/

[3] https://solidproject.org/

[4] https://unhosted.org/

[5] https://fission.codes/

NVDLA [1] targets IoT devices:

> NVDLA is available for product development as part of NVIDIA's Jetson Xavier NX, a small circuit board in a form factor about the size of a credit card which includes a 6-core ARMv8.2 64-bit CPU, an integrated 384-core Volta GPU with 48 Tensor Cores, and dual NVDLA "engines", as described in their own press release. NVIDIA claims the product will deliver 14 TOPS (tera operations per second) of compute under 10 W, but most of this likely comes from the GPU cores. Applications broadly include edge computing inference engines, including object recognition for autonomous driving.

> NVIDIA's involvement with open hardware includes the use of RISC-V processors as part of their GPU product line-up.

[1] https://en.wikipedia.org/wiki/NVDLA

See Ian McGilchrist’s 2009 book The Master and His Emissary:

> McGilchrist digests study after study, replacing the popular and superficial notion of the hemispheres as respectively logical and creative in nature with the idea that they pay attention in fundamentally different ways, the left being detail-oriented, the right being whole-oriented.

[1] https://en.wikipedia.org/wiki/The_Master_and_His_Emissary

That’s interesting and makes sense. SAP introduced HANA NLS (nearline storage) based on Sybase IQ to address your use case (I think). HANA HTAP is in-memory while IQ works with shared storage clusters so it’s ideal for offloading massive amounts of historical BW data that doesn’t fit in memory.

In-Memory HANA is freakishly good for running OLAP BW queries against fresh data. Column stores like HANA and IQ are both good at this, but to be honest, I don’t know how BW systems were typically configured before HANA/IQ.

SQLite is ubiquitous as an embedded database in client-side apps, especially iOS and Android. This ubiquity and simplicity make it a viable alternative to PostgreSQL during development. This is why all popular ORM/QueryBuilder frameworks support SQLite and fits with your appreciation of it.

SQLite as an embedded server-side database requires extra work and configuration to make it a viable alternative to PostgreSQL. It lacks good write concurrency and recoverability by default. It is, however, continually improving but gaps remain.

Since it does not have a wire protocol, SQLite is rarely connected to a data warehouse via ETL so it does not fit well as an alternative to TiDB.

Traditionally, many OLTP operational databases are connected via ETL to an OLAP data warehouse; they are not mutually exclusive. PingCap is the company behind TiDB, an HTAP NewSQL engine that competes with CockroachDB and YugabyteDB; the OP is content marketing for TiDB.

TiDB distinguishes itself with HTAP; transparently incorporating OLTP/ETL/OLAP in a single cluster. You have to specify the ETL layer and data warehouse in addition to PostgreSQL to make an apples to apples comparison; that is the core of HTAP positioning.

SAP HANA is the poster child for HTAP, a data warehouse with good enough OLTP performance to replace Oracle RDBMS; a single system is used for both SAP app tiers, Business Suite and Business Warehouse. The same value proposition applies to cloud apps. Independent OLTP/ETL/OLAP is still robust and is more modular while HTAP is more tightly integrated and simpler to operate.