antipaul

GitHub too is useful for solo work: you can code review your own code.

I’ve found it useful because it’s in a separate context/UI (GitHub web view) as opposed to your code editor or even git diff on command line.

So far, they haven’t given any Fs - subscribers keep growing:

https://www.nytimes.com/2021/02/04/business/media/new-york-t...

Maybe HN and other crowd activism will make them change?

Great story! All the best!

Indeed, this was a creative, even literary, entry but it makes a memorable case for overcoming writers block, I mean, entrepreneur’s block.

But then that tech department would become government itself, and would that be better?

Maybe, maybe not.

Though rare, there _are_ private success stories, such as healthcare.gov 2.0 [1] and Palantir, though the latter gets in hot water for some tangential reasons. But maybe figure out how these happened, and scale that.

[1] https://www.theatlantic.com/technology/archive/2015/07/the-s...

The small problem with your kind of approach is that it would make the army of administrators and contractors like “Deloitte.” quite redundant, and then they’d actually have to do non-Bullshit Jobs!

What’s the big deal about type checking?

I haven’t worked on large code bases, only run small ML analyses on tabular data. Can someone be concrete with the advantages and use cases?

The article is so true, my latest mantra at work is “engineering is more important than data science”.

Everyone is buzzing about the latter, and few even realize what is the former.

I have a dream - and it looks like this!

Sounds very true to me.

I'd summarize the problem as lack of real engineering leaders. It's as if programming/engineering is looked down on or not taken seriously enough.

- Biotech (in Silicon Valley to boot!)

Dear author,

Please follow-up on your last paragraph.

Sincerely,

A less courageous soul

I also _want_ to like Apple Maps. But I always need to double check business hours using Google.

Thanks for the reference. Here it is:

“We show, however, that deep networks learned by the standard gradient descent algorithm are in fact mathematically approximately equivalent to kernel machines, a learning method that simply memorizes the data and uses it directly for prediction via a similarity function (the kernel)“

https://arxiv.org/abs/2012.00152

It’s always fascinating to apply data/results like from this paper to help evaluate the hypothesis that machine learning/AI is mostly just a rough “lookup table” or memorization.

Genuinely curious why not combine syntax? Do we need 2 different pipes in R? When to use which? Thanks for your thoughts!

Thankfully they didn’t use just any “machine learning” nor even “AI” but “advanced machine learning AI”. Phew!

Anyone know how many of these “features” are _already_ provided in Google GSuite/Workspace?

As usual with ML, I now wonder how “similar” the test set is to the training set, compared to the examples that are neither in the training set, nor the test set:

TODO: 200 million - 170,000 training - 100 test ~= 199.8 million proteins

What size was the test set?

From what I gather, training was done on 170,000 Amino Acids (features) and the resultant protein structure (labels). This is out of 200 million possible proteins.

How many examples were in the test set?

EDIT: Looks like N=100 for test set: “Entrants get amino acid sequences for about 100 proteins whose structures are not known” https://www.sciencemag.org/news/2020/11/game-has-changed-ai-...

Again, M1 chip shows impressive performance gains. First graph shows, what, 3-4x improvement over Intel?

Fascinating thought. Will Apple make a standalone screen/dock to plug in your iPhone/iPad...?

This MBP refresh replaces the lower-end MBP which also only had 2 ports.

This is a small detail that is easy to miss.

The "higher-end" MBPs with 4 ports are only availabe with the intel CPUs for now

Not OP, but I’ve had same difficulty switching among windows.

I remap “other window” shortcut to something more convenient. See https://emacs.stackexchange.com/a/3471

Not to mention the condescension in calling them, right off the bat, “a grumbling shuffle of ingrained habit and stubborn resistance to change”.

Habit and sticking with something that works are, like, good things most of the time!

Or they had no choice but to follow the AI strongman, I mean, CTO, because of bad culture.

Culture eats strategy, after all.

I like the point you make.

I think there are the AI leaders, and then there is everyone else. What is the difference?

The leaders are mostly big tech, who have driven the step-change advances you describe. This, IMHO, was due to their pre-existing mastery of data. They already had a ton of well-organized data, because they were engineering cultures, and data was the lifeblood of their business (ads, search, shopping). Once ML/AI came into the picture, it was full steam ahead.

Most others are (blind) followers, and cannot tease apart the engineering bit from the (data) science hype. They get fixated on the latter (data science and ML/AI) and forget the engineering, or "scale" part.

The first question should not be about AI/ML, but on the other hand, do you have solid (data) engineering where your data is easily accessible to any data scientist? By now it should be apparent that "data is the new oil" and will be useful even if you don't plan to do deep learning.

If you don't have solid (data) engineering and "data at scale" for anyone, anywhere, then your ML/AI efforts are doomed.

Data first, only then ML/AI. See "Data Science Hierarchy of Needs": https://hackernoon.com/the-ai-hierarchy-of-needs-18f111fcc00...

Perhaps here it would be more advisable to just describe the data, and leave inference to the reader.

Prediction is hard, especially about the future ;)

Isn’t “Back” usually just a swipe from the left edge of the screen??

(Admittedly not 100% always and probably not on non-native apps)

AI as “ideology” may be a stretch, but it tends there with all the mythos behind a word like “artificial intelligence”.

Naming things matters.

While the head of Apple’s ML/AI strategy likes “machine intelligence” [1], I prefer “machine learning”.

“Intelligence” is a loaded word. I think (hope) that “learning” can be understood as narrowly and crudely as the field actually requires.

A more sober term would help the entire world manage this better on average, IMO, than if we use the unnecessarily scary and confusing term “artificial intelligence”.

[1] “For this reason and others, many AI experts (Giannandrea included) have suggested alternative terms like "machine intelligence" that don't draw parallels to human intelligence.”

https://arstechnica.com/gadgets/2020/08/apple-explains-how-i...

This was a good outline of the leadership failures common in large orgs trying to get good value out of data science.

Edit: In summary:

“Any organization’s adoption of AI and machine learning requires three technical tools: usable digital data ..., computational capabilities ..., and a development environment that engineers use to code”