About 20% beyond core case of government-critical speech in formally public spaces and media I'd say.
There are only 3 real challenges to free speech at the tech level a) the need for CDNs at scale b) distortions of net neutrality c) nation-state firewalls.
All other claims are about social governance on private sites, where applying free speech doctrine is a stretch at best
About 20% beyond core case of government-critical speech in formally public spaces and media I'd say.
That's a genuine de facto increase in practical, informal free speech/expression.
This is reasonable. You should not expect no-strings-attached access to the attention of potentially billions of people under the same terms as ranting from a soapbox in a 19th century town square.
More from Venkatesh Rao
Heh, one thing the nyt piece managed was to do a Cunningham's law nerdsnipe-wmd at newspaper scale... now a bunch of people are energetically trying to post the right answer.
IMO trying to correct whatever the NYT writer thought he knew/understood is futile. "Willing to be misunderstood by the NYT" should be the default stance unless you want to waste a lot of time correcting an obsolete 2013 map for people who don't care.
The thing is, the NYT still has enough normative cultural power, even as it has fallen from newspaper-of-record, that it takes a particular sort of heretical self-confidence to sort of ignore whatever they happen to be wrong about on any given week, whether or not it concerns you
A subtle shift has occurred in the workings of the Gell-Mann amnesia effect. It used to be an individual private amnesia re: media ("I'll believe myself when I am certain they got it wrong because I'm an expert, but still believe them when I am not"). Now it's a collective effect
A sort of common-knowledge threshold has been crossed lately. "Everybody knows that everybody knows the NYT is wrong on X across largish subcultures." It's no longer mutual beliefs being validated occasionally 1:1.
One other thing I should really clarify and that the @nytimes piece got *severely* wrong: while I believe there are very strong sociological and even causal links between rationalism and NRx (especially in the Silicon Valley homes bases) their ideological and methodological
— (((E. Glen Weyl))) (@glenweyl) February 14, 2021
IMO trying to correct whatever the NYT writer thought he knew/understood is futile. "Willing to be misunderstood by the NYT" should be the default stance unless you want to waste a lot of time correcting an obsolete 2013 map for people who don't care.
The thing is, the NYT still has enough normative cultural power, even as it has fallen from newspaper-of-record, that it takes a particular sort of heretical self-confidence to sort of ignore whatever they happen to be wrong about on any given week, whether or not it concerns you
A subtle shift has occurred in the workings of the Gell-Mann amnesia effect. It used to be an individual private amnesia re: media ("I'll believe myself when I am certain they got it wrong because I'm an expert, but still believe them when I am not"). Now it's a collective effect
A sort of common-knowledge threshold has been crossed lately. "Everybody knows that everybody knows the NYT is wrong on X across largish subcultures." It's no longer mutual beliefs being validated occasionally 1:1.
I’m guessing these responses really reflect people’s weighted averages (age*current average effort fraction) though I kept it simple and asked for just averages.
I suspect a healthy weighted average should be ~ (age-20)/2. So a 30 year old should be at 5, a 40 year old at 10, a 50 year old at 15 etc.
Standard deviation should be ~average/3 maybe, so distribution spreads as you age and accumulate projects and get better at them.
Other things being equal, people get good at starting in their 20s, at follow through in 30s, at finishing in 40s.
No point learning food follow through until you’ve found a few good starts to bet on. No point getting good at finishing until a few projects have aged gracefully.
I’m in the 7+ range myself. Probably 8-9. Slightly less than healthy for my age.
I suspect most self-judgments on being good starters/follow-through-ers/finishers are really flawed because of the non-ergodicity of project management skill learning. You can’t learn good practices for the 3 phases in an arbitrary order. On,y one order actually works.
Poll: where is the temporal center of gravity of all your live projects based on average age of start-dates?
— Venkatesh Rao (@vgr) January 17, 2021
I suspect a healthy weighted average should be ~ (age-20)/2. So a 30 year old should be at 5, a 40 year old at 10, a 50 year old at 15 etc.
Standard deviation should be ~average/3 maybe, so distribution spreads as you age and accumulate projects and get better at them.
Other things being equal, people get good at starting in their 20s, at follow through in 30s, at finishing in 40s.
No point learning food follow through until you’ve found a few good starts to bet on. No point getting good at finishing until a few projects have aged gracefully.
I’m in the 7+ range myself. Probably 8-9. Slightly less than healthy for my age.
I suspect most self-judgments on being good starters/follow-through-ers/finishers are really flawed because of the non-ergodicity of project management skill learning. You can’t learn good practices for the 3 phases in an arbitrary order. On,y one order actually works.
Both this thread and the outraged response threads are... something.
This is why I never wanted kids. Way too much responsibility for another human’s development. Depending on the child, this might either be the day they discovered who they were or the day that traumatized them into a lifelong fuckup. Either way I don’t want to direct the show.
As far as the can opener goes, it wouldn’t even occur to me to try and turn it into a teachable moment. That sounds vaguely quixotic. I’d just show them how immediately. I think my default is to try and instruct clearly but not demonstrate unless the person is truly disoriented.
I think there’s basically a right answer here: show the kid. If the kid has the aptitude they’ll enjoy the mechanism so much they’ll develop the figure-it-out skill with other devices. If not, it’s a training data point that will build remedial levels of intuition more slowly.
I think perseverance is both misframed and over-rated as a virtue. Misframed as in: everybody has potential for it in some areas and lacks it in others. Aptitude is those areas where perseverance comes easily to you. Meta-skill of knowing where/why you persist is more important.
So, yesterday my daughter (9) was hungry and I was doing a jigsaw puzzle so I said over my shoulder \u201cmake some baked beans.\u201d She said, \u201cHow?\u201d like all kids do when they want YOU to do it, so I said, \u201cOpen a can and put it in pot.\u201d She brought me the can and said \u201cOpen it how?\u201d
— john roderick (@johnroderick) January 2, 2021
This is why I never wanted kids. Way too much responsibility for another human’s development. Depending on the child, this might either be the day they discovered who they were or the day that traumatized them into a lifelong fuckup. Either way I don’t want to direct the show.
As far as the can opener goes, it wouldn’t even occur to me to try and turn it into a teachable moment. That sounds vaguely quixotic. I’d just show them how immediately. I think my default is to try and instruct clearly but not demonstrate unless the person is truly disoriented.
I think there’s basically a right answer here: show the kid. If the kid has the aptitude they’ll enjoy the mechanism so much they’ll develop the figure-it-out skill with other devices. If not, it’s a training data point that will build remedial levels of intuition more slowly.
I think perseverance is both misframed and over-rated as a virtue. Misframed as in: everybody has potential for it in some areas and lacks it in others. Aptitude is those areas where perseverance comes easily to you. Meta-skill of knowing where/why you persist is more important.
More from Tech
On press call, Zuckerberg says FB users "naturally engage more with sensational content" that comes close to violating its rules. Compares it to cable TV and tabloids, and says, "This seems to be true regardless of where we set our policy lines."
Zuckerberg says FB is in the process of setting up a "new independent body" that users will be able to appeal content takedowns to. Sort of like the "Facebook Supreme Court" idea he previewed earlier this year.
Zuckerberg: "One of my biggest lessons from this year is that when you connect more than 2 billion people, you’re going to see the good and bad of humanity."
This is how Facebook says it's trying to change the engagement pattern on its services. https://t.co/3p0PGc912o
.@RebeccaJarvis asks Zuckerberg if anyone is going to lose their job over the revelations in the NYT story. He dodges, says that personnel issues aren't a public matter, and that employee performance is evaluated all the time.
Zuckerberg says FB is in the process of setting up a "new independent body" that users will be able to appeal content takedowns to. Sort of like the "Facebook Supreme Court" idea he previewed earlier this year.
Zuckerberg: "One of my biggest lessons from this year is that when you connect more than 2 billion people, you’re going to see the good and bad of humanity."
This is how Facebook says it's trying to change the engagement pattern on its services. https://t.co/3p0PGc912o
.@RebeccaJarvis asks Zuckerberg if anyone is going to lose their job over the revelations in the NYT story. He dodges, says that personnel issues aren't a public matter, and that employee performance is evaluated all the time.
Recently, the @CNIL issued a decision regarding the GDPR compliance of an unknown French adtech company named "Vectaury". It may seem like small fry, but the decision has potential wide-ranging impacts for Google, the IAB framework, and today's adtech. It's thread time! 👇
It's all in French, but if you're up for it you can read:
• Their blog post (lacks the most interesting details): https://t.co/PHkDcOT1hy
• Their high-level legal decision: https://t.co/hwpiEvjodt
• The full notification: https://t.co/QQB7rfynha
I've read it so you needn't!
Vectaury was collecting geolocation data in order to create profiles (eg. people who often go to this or that type of shop) so as to power ad targeting. They operate through embedded SDKs and ad bidding, making them invisible to users.
The @CNIL notes that profiling based off of geolocation presents particular risks since it reveals people's movements and habits. As risky, the processing requires consent — this will be the heart of their assessment.
Interesting point: they justify the decision in part because of how many people COULD be targeted in this way (rather than how many have — though they note that too). Because it's on a phone, and many have phones, it is considered large-scale processing no matter what.
It's all in French, but if you're up for it you can read:
• Their blog post (lacks the most interesting details): https://t.co/PHkDcOT1hy
• Their high-level legal decision: https://t.co/hwpiEvjodt
• The full notification: https://t.co/QQB7rfynha
I've read it so you needn't!
Vectaury was collecting geolocation data in order to create profiles (eg. people who often go to this or that type of shop) so as to power ad targeting. They operate through embedded SDKs and ad bidding, making them invisible to users.
The @CNIL notes that profiling based off of geolocation presents particular risks since it reveals people's movements and habits. As risky, the processing requires consent — this will be the heart of their assessment.
Interesting point: they justify the decision in part because of how many people COULD be targeted in this way (rather than how many have — though they note that too). Because it's on a phone, and many have phones, it is considered large-scale processing no matter what.
THREAD: How is it possible to train a well-performing, advanced Computer Vision model 𝗼𝗻 𝘁𝗵𝗲 𝗖𝗣𝗨? 🤔
At the heart of this lies the most important technique in modern deep learning - transfer learning.
Let's analyze how it
2/ For starters, let's look at what a neural network (NN for short) does.
An NN is like a stack of pancakes, with computation flowing up when we make predictions.
How does it all work?
3/ We show an image to our model.
An image is a collection of pixels. Each pixel is just a bunch of numbers describing its color.
Here is what it might look like for a black and white image
4/ The picture goes into the layer at the bottom.
Each layer performs computation on the image, transforming it and passing it upwards.
5/ By the time the image reaches the uppermost layer, it has been transformed to the point that it now consists of two numbers only.
The outputs of a layer are called activations, and the outputs of the last layer have a special meaning... they are the predictions!
At the heart of this lies the most important technique in modern deep learning - transfer learning.
Let's analyze how it
THREAD: Can you start learning cutting-edge deep learning without specialized hardware? \U0001f916
— Radek Osmulski (@radekosmulski) February 11, 2021
In this thread, we will train an advanced Computer Vision model on a challenging dataset. \U0001f415\U0001f408 Training completes in 25 minutes on my 3yrs old Ryzen 5 CPU.
Let me show you how...
2/ For starters, let's look at what a neural network (NN for short) does.
An NN is like a stack of pancakes, with computation flowing up when we make predictions.
How does it all work?
3/ We show an image to our model.
An image is a collection of pixels. Each pixel is just a bunch of numbers describing its color.
Here is what it might look like for a black and white image
4/ The picture goes into the layer at the bottom.
Each layer performs computation on the image, transforming it and passing it upwards.
5/ By the time the image reaches the uppermost layer, it has been transformed to the point that it now consists of two numbers only.
The outputs of a layer are called activations, and the outputs of the last layer have a special meaning... they are the predictions!
