gnat

Yes, that's how major outages happen. By this stage of maturity any single failure generally doesn't break things dramatically. When things go this wrong, it's ALWAYS a combination of failures: failure of recovery system, omission in detection systems, gap in automated review, oversight in ...

The vacuous gotcha language is indicative of the low quality of the whole article. As Metalnem says in comments here, see the official incident report for a better writeup and more insight. https://status.cloud.google.com/incidents/ow5i3PPK96RduMcb1S...

I was aware as I was writing it that consistent mediocrity is indeed a profitable target. Perhaps we're quibbling over "mediocrity"? Staying within the spec seems enough of the challenge for a grid, I'm not sure that I'd define excellence as narrower and narrower variation around the spec there.

I was making a moral judgement in "unworthy". I like cars to come off the factory floor consistently good. I own a Tesla, this statement has high salience for me. That seems like a challenge. I wouldn't respect the Lada factory for consistently turning out cars that break down or fall apart, just as I don't respect McDonalds for consistently delivering underwhelming food. I acknowledge the consistency, I acknowledge it's profitable, but they're not achieving consistent greatness.

The US Military have admissions criteria. They bounce people out in Basic Training and in every other part of training. Not everyone gets to be an ATC, a pilot, or a nuclear specialist. Air Traffic Control turns out not to be a process you can put an unskilled person into. It requires training and careful integration, and once you have a skilled person with many hours invested in them, perhaps you can let them direct traffic.

Training isn't a magic process that takes unskilled people who don't care and delivers great results every time. The equivalent in our world would be "I'll hire cheap people who don't know how to program, put them through a Bootcamp process, and then I'll have great programmers." That didn't work.

I am trying to say that: every version of programming where there's been The Process You Just Have To Follow (from Jackson Structured Design) has failed significantly and hasn't been a substitute for hiring smart people who care. If someone came to me with a business plan that was "hire mediocre people who don't care, and we'll achieve great results because of My Process", I'd be veeeeery skeptical indeed.

Another variation: the magic process that lets you get great results from people who don’t care.

Excellence doesn’t come solely from process. The benchmarks of great process with unskilled workers who don’t care is fast food, and this product is consistently mediocre. Consistency is good but consistent mediocrity is an unworthy ambition in many fields.

NASA is massively process driven but they get people to space and back, and the people in their process are highly skilled and care deeply.

Larry Wall, inventor of Perl, used to say (and probably still does) that complexity has to go somewhere. If you’re solving a complex problem then either it’s a simple program with complex tools, or a complex program with simple tools.

That’s really stuck with me. The art of library, framework, language, API design is to provide a set of tools — if you make simple tools then complex problems become complex programs. And if you offer complex tools, complex problems can have simple solutions. And people will gripe at you for all the punctuation for decades. :)

But the complexity exists and can’t be wished away. Which is kinda your point too. Thanks for writing!

Good luck with the wild ride of parenthood!

Each god has a bunch of different associations/responsibilities in traditional Māori culture. Applying those to book classification is the novelty, which they claim as novel and are proud of having created.

I have a Real Job now but still collect interesting links. Here’s my 2024 list, from before I got interested in AI. Perhaps some of that is interesting to you, and New To Me is almost as good as New To The World right?

https://docs.google.com/document/d/1S9pK6EYiJvRlBDdlIoNQCOkV...

I admit that I’m no longer trawling with a wide net. Interesting to me professionally and interesting to me based on what I already know are weighting that list heavily. They’re pretty “adjacent possible” is what I’m saying.

Beyond those, the trends I’d be watching are:

* Apple took chip design in house and the results are incredible. I’d be watching for other companies following suit, which industries/players does it make most sense for, are there any low hanging fruits in this world, what does it do for hiring, and then downstream for education?

* Quantum Computing is still in the physics phase of research. There are bullshit artists, claims and counter claims. I’d be trying to take a position on whether national security is the biggest driver or whether there’s commercial investment logic yet. Does NatSec tech even matter in an age of Doge? The technical advances themselves are still coming slowly. There was a Queensland U podcast that I followed for a while, and I’d dive in by looking for its spiritual successor.

* correctly implementing distributed systems seems like a rich company’s achievement because of the teams and systems required. I’m watching for systems that let teams of 8 build an “enterprise” product that can survive network and software outages without a massive ops team and platform engineers and and and

* AR/VR fizzled as virtual escapism and The Next Big Things. I’m interested in moving beyond physical monitors. Is it possible? Is it healthy? The VR advocates sneer at “just looking at monitors fixed in space”. That seems like low hanging fruit given I walk around offices full of monitors every day. I want to understand why we still have physical monitors and look for ways to replace them.

* privacy. Capitalism drove investment and development of ever more tracking and data exfiltration. I’d be looking for intermediaries like PiHole as ways civilians can limit tracking and exfiltration. Bit screwed for phones when every app is phoning home with fingerprints and metadata galore. Advances in this area excite me, as tech bandaids before the regulation that is the only way the big players will get reined in.

I don’t know enough about fab techniques to pay attention to nanometres news, but I do know the political topsy turvy will be interesting there in terms of who develops, protects, and houses advances in chip manufacturing.

Space tech doesn’t interest me. Med tech does but it’s very hard to separate press release hype from substance and then to understand the future based on that substance. No medical degree and too old to pack a PhD from ChatGPT U into my noggin.

I like my job but I miss being in a place where people would pitch me to get me excited by their tech and businesses and vision of the future. Tim O’Reilly’s “create more value than you capture” was my first filter, and it was good to have a source of excitement and optimism from what passed that. I reckon we all could do with a little more of that.

The Center of Gravity shifted in the late 90s/early 2000s, I reckon. Real world needed distributed systems, and neural networks, and crypto math, and hardware design, and …

There’s still a place for academic research but it’s a much different place in the Gilded CPU Age. It’s the god of the non-commercial gaps: for a while there if it seemed like it’d make money, someone .com giant with gigabucks of free cash had someone on it. Why work for a university if gigabucks.com is hiring?

Not saying USENIX didn’t have factions. I just question the word “capture” about academics, who invented all this stuff before there was money in it. It feels like retiring the Annual Technical Conference is the final step in Usenix’s capture by industry.

> Our prompt is currently very simple: "OCR this PDF into this format as specified by this json schema" and didn't require some fancy "prompt engineering" to contort out a result.

Many people make different choices on those drugs. Eg once they’ve lost some weight and it’s easier to move, they move more and the sedentary lifestyle changes. Similarly, I’ve seen antidepressants used to get enough functioning and control back that the chain of causes and effects can start to be addressed and intervened in.

Jeff wrote "Mastering Regular Expressions", which grew from that talk. You probably want a copy even though it was first released in 1997. For the mindset of RE, you can't beat it.

Learning REs is a roll through:

   * how matching happens (advancing, matching, backtracking)
   * using * ? and {} to match repetitions
   * greediness and stinginess within the RE
   * character classes, both [manual] and escapes like \s \W etc
   * anchors and "what a line is"
   * grouping and backreferences
   * accessing groups outside the RE
   * substitution and access backrefs in substitutions
   * find ALL the matches
   * complex parsing (just don't, it's rare not to regret it)

> The researchers analyzed the health records to estimate the percentage of Covid patients who had long Covid one year after being infected. The lowest rate of long Covid in the study, 3.5 percent, was among vaccinated people who were infected during the latest period in the study, between mid-December 2021 and January 2022.

> That compares with a rate of 7.8 percent for unvaccinated patients in the study who were infected during the same period.

> “We found that much of the decline is attributable to vaccination,” said the study’s lead author, Dr. Ziyad Al-Aly, the chief of research and development at the V.A. St. Louis Health Care System and a clinical epidemiologist at Washington University in St. Louis.

...

> Among vaccinated people who had been infected, the rates of long Covid were markedly lower. Differences in variants and other aspects of the Delta and Omicron periods played a role, the authors said, but they attributed about 72 percent of the decrease to vaccines.

In this work, we propose a light-weight (~20M param.) causal voice conversion solution that can run in real-time with low latency on a commercially available mobile device. The key design elements are: (1) using a causal encoder to learn soft speech units; (2) injecting whitened f0 to improve pitch stability without leaking source speaker info.

In our later V2 version, we found that f0 rescaling followed by a NSF-style harmonic-plus-noise conditioning (as is done in RVC) results in better quality.