One could argue that NVidia's advantage comes from a similar vision epiphany that led to them developing CUDA years before it was viable. The result is similar.
I played with NNs in the late 80's/early 90s, with little more than a copy of Hinton's paper, a PC and a C compiler. Obviously, I got no practical results. But I got the intuition of how they worked and what they could potentially do.
Cut to 2008-9,and I started to see smartphones, grid (then cloud) computing and social networks emerging. My MBA dissertation, finished in 2011, was about how that would change the world, because the requirements for meaningful AI were coming along - data and compute. The theory was already there, Hinton, LeCun, Schmidhuber,etc.
That got me back into the Data Science field, after years working in Data Engineering. Too bad I lived in Brazil back then and couldn't find a way to join the emerging scene in California and other top places. I'd be rich now...
I don't know. Many of these ideas sound like "give me more of the same", reinforcing my current tastes and beliefs. The thing about going out there and interacting with stuff you don't know is that it had a chance of pushing your boundaries. If these agents are "good" as defined in the article, everyone ends up in an echo chamber.
Also, it may sound great for someone transitioning from a world before these agents were created, but how should the new generations coming in be handled? What is the starting state? Who decides that? Social media was not that bad when it started, but iterations over the algorithm and the incoming new natives to it are having devastating effects a very short time after. Do we really understand the consequences of living in a world where everything is curated for you?
I don't know that I want my life to be made so easy, that I want something to remove the need for choosing, thinking, criticizing and exposing myself to stuff out of my comfort/interest zone.
> society as a whole is in agreement that minors are better off without access to pornography
Once a significant part of said society can't (or won't) differentiate sexual education and intimacy from pornography, I don't think your statement holds true anymore.
It's not a matter of life and death, I agree - to some extent. Startups have very limited resources, and ignoring inconclusive results in the long term means you're spending these resources without achieving any bottom line results. If you do that too much/too long, you'll run out of funding and the startup will die.
The author didn't go into why companies do this (ignoring or misreading test results). Putting lack of understanding aside, my anecdotal experience from the time I worked as a data scientist boils down to a few major reasons:
- Wanting to be right. Being a founder requires high self-confidence, that feeling of "I know I'm right". But feeling right doesn't make one right, and there's plenty of evidence around that people will ignore evidence against their beliefs, even rationalize the denial (and yes, the irony of that statement is not lost on me);
- Pressure to show work: doing the umpteenth UI redesign is better than just saying "it's irrelevant" in your performance evaluation. If the result is inconclusive, the harm is smaller than not having anything to show - you are stalling the conclusion that your work is irrelevant by doing whatever. So you keep on pushing them and reframing the results into some BS interpretation just to get some more time.
Another thing that is not discussed enough is what all these inconclusive results would mean if properly interpreted. A long sequence of inconclusive UI redesign experiments should trigger a hypothesis like "does the UI matter"? But again, those are existentially threatening questions for the people in the best position to come up with them. If any company out there were serious about being data-driven and scientific, they'd require tests everywhere, have external controls on quality and rigour of those and use them to make strategic decisions on where they invest and divest. At the very least, take them as a serious part of their strategy input.
I'm not saying you can do everything based on tests, nor that you should - there are bets on the future, hypothesis making on new scenarios and things that are just too costly, ethically or physically impossible to test. But consistently testing and analysing test results could save a lot of work and money.
If the law has specific clauses about this that the contract disrespects, these conditions are not worth the paper they are written on.
At least in Brazil you can't enforce something the law doesn't allow in a contract - that clause would be considered void without nullifying the contract. And Labour law in Brazil leans (or used to lean) more in favor of the employee,so yes, the law would win. Another aspect there is that unions are more common than in the US, and they will help in such cases.
I am responding to this from my living room in Berlin, sitting on a sofa that belonged to my father, after having dined on a table he inherited from my grandfather. Both were brought with us when we moved from Brazil.
So yes, people do want to inherit the old stuff. I have some IKEA stuff (the beds were just too big, and mattress sizes are different), it just can't compare.
I have an issue with calling any of the diagrams created by all the tools mentioned "ER Diagrams". Entities are not the same thing as tables, and Sr diagrams are not relational database table diagrams. A (semi) visual representation of a database schema of any size that'd require a visual representation is almost necessarily a mess, and doesn't really help discussion or design. It is at best a faster indexing into the DDL for the tables.
What I'd love to have (but never saw an affordable tool) is the ability to work at different levels of abstraction: physical (which is what all tools here actually do), logical (hiding field details, normalization and de normalization, giving better business entity names, etc) and conceptual (to show how big picture concepts relate, domain boundaries, department dependencies/relationships).
Just the physical representation does, for my purposes, little more than code highlighting.
...and you just described why this is not ready for prime time. Managing a number of physical devices tied to completely opaque secrets stored by unclear providers in places you never see, with hidden agendas promoting their locked-in solution over all others and complicating everything out of one ecosystem.
Most standard users will either mess up royally or run away scared. Damn, I've been on this field for 30 years, I've been using 4 OSs, 5 different browsers and devices from every ecosystem, and I still find this whole thing too much of a hassle.
And yes, I do have a backup passkey. Even though I had to convince my skip-level that it made sense. I just find it all too complex to adopt it broadly.
I don't necessarily agree. Blocking ads on a private site is more like going into a bookstore, reading the magazines and books and not buying anything. At some point the store clerk will approach you and suggest that you either pay or leave.
Poor people should have the option to go to the public library and read things for free. In the public transportation metaphor, poor people should get fair public transportation at an affordable price, and not need to ride without tickets.
Saying ads sustain poor people is, in my view, saying there is no expectation that the State/society will give a decent base level of service to anyone, and we're left to what private companies are willing to do and the mercy of the ones who can pay - coupled with their willingness to go along with whatever conditions the companies enforce for their "benevolence" (i.e., making you and your data the product).
So, what's wrong is that there's a gap here, and the poor people are not getting any support to bridge that gap and eventually get better off and add more value to society. Private philanthropy is not a valid welfare policy.
We need a new model like we had with libraries, one that handles the fact that most new content is now online. Unfortunately, most discussions I've seen around that were more about IP rights and protecting paywalls than enabling a 21st century lending model. And the traditional libraries are dying. Solving that would do more for the poor than subjecting yourself to intrusive ads ever will.
That's an interesting point, which leads me to my main reason for coming to these comments and leaving my 2 cents: there are way less best practices of there than one would believe by looking at all the places, people and firms offering some set of "best practices".
One thing I learned after many years working in consulting is that, more often than one would believe, best practices are just a compilation of whatever could be found (hopefully at least common practices, more often "things I could find that were minimally documented to be reusable"), with no serious analysis of their claim of superiority other than them being common.
So, first thing: learn to challenge the claim of "best". Best for whom? Under what context? What other not-so-good practices are out there, and why is this the best?
Second:if it's documented and evident enough to be treated as a best practice, it's probably fairly common knowledge already. Barring the commonality of really bad things being done out there, don't expect that you'll become much more than mediocre by adopting best practices. By the time they get to be called there, they are no longer any competitive advantage, more a basic thing you should be doing already - assuming they are indeed best practices (as per my previous point).
It's not that I'm against best practices as a concept, or compiled bodies of knowledge. Just don't expect them to do more than keep you somewhere in the middle. True leadership and innovation lies where best practices have not been established yet - together with all the dangers and mistakes you can make on uncharted waters.
Also brazilian here. Freedom of speech does not mean freedom from the consequences of illegal speech. One is allowed to go public and speak their minds, but if their speech is illegal (hate speech, conspiracy to overthrow the government, political campaigning during embargo periods), there will be consequences for those, and that does not constitute censorship.
Initially, consequences were not that bad (take down of some illegal posts), then they went to removal of recurring offender profiles.X ignored those Supreme Court Justice orders - their only legal course of action being to comply and file an appeal to the Supreme Court as a whole. That led to further escalation against their legal representation in Brazil and their executives (which is according to Brazilian law), which led to Musk shutting down the local representation rather than following the local law. Which put X in a non-compliance state and led to the order for its blocking.
If you understand the initial order to take down posts of defamation and illegal speech as censorship, you comply and appeal. Ignoring a court order is not a legal option.
That's the legal justification for blocking X now, not the root cause. Musk did more than refuse to appoint a representative in Brazil (which could be a subsidiary or any legal resident, not necessarily a native Brazilian). X had representation, and when the natural people leading that representation firm were sued (which is legal in Brazil), he shut down the representation, putting X in a non-compliance state to Brazilian law.
The whole thing started because X refused to take down posts judged as defamatory against some politicians in Brazil, as well as some profiles accused of consistently posting fake news and borderline illegal content. One can disagree with the ruling and appeal, but ignoring a Supreme Court Justice order is not a legal option, which led to the escalation.
True that no one would want to step in as a legal representative of X in Brazil right now, but that doesn't change the legal requirement - it exists so that the State has the power to enforce law over companies effectively operating in the country. The US is doing something similar (in process, motivations are quite different) by threatening to ban TikTok, for instance.
The Starlink ruling is mostly being considered an overreach by the Justice. It may take some time, but it will likely be withdrawn. Him deciding to keep the service for free, as long as it complies with the law, bears no matter and should be read most likely as a publicity stunt.
This is a Justice from the brazilian Supreme Court, they are the highest position in the court system, and allowed to make individual rulings and apply sanctions like this.
These decisions hold until ratified or rejected by the court as a whole (which they all eventually will, but it's not fast) or successfully appealed. Appeals can be made only to another Justice or to the court as a whole - no expedited process, because there's no higher authority.
Besides, who'd make the appeal? X has no representation in Brazil (that's why it's been suspended), and there seems to be consensus on that specific point, of legal representation being a requirement by law, so the general attorney or other officials will not question the main decision.
The side decisions are a different matter, that about VPN apps and app stores had been withdrawn. The fine for accessing X is more controversial: hardly enforceable (for just browsing, at least), ongoing hot debate in the country about it being within the Justice's power.
In fact, the Court will probably expedite a whole court judgement on that, and apparently there's no consensus across the Justices on that.
I think we have a problem of semantics here. Your notion of "the brain internalizes insights" is very close to what the author means as memorizing patterns. They even gäbe a few examples where they started with rote memorizations, which were not that useful at first, but eventually a pattern, an insight if you will, emerged.