What makes you think sending data to the Claude API is a breach of privacy? Do you not trust them when they say they won't look at or train on your data?
I've also been following Anthropic pretty closely for the last two years and I've seen no evidence that they would break their principles here and plenty of evidence of how far they go to respect the privacy of their users: https://simonwillison.net/2024/Dec/12/clio/
But how is that a personal advantage? Who are you in competition with against yourself? Maybe I'm parsing competive advantage in a different context than you mean.
I guess I'm competing against other humans at living a fulfilling, enjoyable life?
I don't take that competitive advantage particularly seriously, which is why I invest so much effort giving away what I've learned along the way for free.
It's frustrating that the definition of "drinks" varies and even when defined is not easy to adjust for various types of alcohol. So much easier to measure by units of alcohol. Just multiply ml of drink by percentage of alcohol and divide by 1000.
Where I live, and it seems in most places in the world, a standard drink is 10g of alcohol.
> The definition of what constitutes a standard drink varies very widely between countries,[2] with what each country defines as the amount of pure alcohol in a standard drink ranging from 8 to 20 grams.
> The sample questionnaire form for the World Health Organization's Alcohol Use Disorders Identification Test (AUDIT) uses 10 g (0.35 oz),[3] and this definition has been adopted by more countries than any other amount.
One problem with the idea of a “standard drink” is that what people typically get at a bar (eg a pint of beer) is actually a fair bit more than a standard drink. It’s unrelated to a typical pour.
There is no standard definition of a shot. Serves in the US will vary wildly, as will serves across Europe.
In Australia, a shot is 30ml (close enough to 1 oz to make no difference) which is exactly one standard drink or 10g of alcohol, assuming straight spirit.
There's a bar I used to go to that has Whiskey Wednesdays. They're entire whiskey menu was half off. What they don't tell you (unless you ask) is they pour 1 oz shots instead of 1.5.
Note that a 568 ml pint of beer of 5% strength is by this definition 2 drinks.
4 pints over a week is basically nothing, and if they are saying it causes ill health, I don't believe them. When I was a student I drank 10 times that amount, so did many of my friends, without any apparent ill effects.
> Note that a 568 ml pint of beer of 5% strength is by this definition 2 drinks.
Yes it is. I can always tell the difference between drinking a can of it versus a pint, it definitely does feel like twice the effects.
> if they are saying it causes ill health, I don't believe them
I mean you can believe whatever you want but the scientific consensus is that any amount of alcohol is bad for you [0]. Just because you and your friends didn't have ill effects does not mean that it was actually good or even neutral for you all, and statistically and biologically it's been shown to not be so either.
There’s probably no safe limit of driving a car or using a step ladder.
Alcohol has been with us since the beginning and will probably be with us until the end. The point of these studies should be to make people aware of the risks and motivate them to make appropriate choices.
But part of the understanding of risk should be that there is a floor below which something else will most assuredly kill you. Is that floor 10 drinks per week? 2? Who knows.
Either way, I struggle to find a likely definition of "drinks" where 8 per week does not seem excessive.
Like even if "half a pint of beer" is the definition (unlikely), that means drinking 4 pints of beer per week, throughout a long time? I.e. not just for 2 weeks in a row or so? That does not fit my definition of "social drinking" anymore.
And mind you that that's a purposefully underestimating assumption. The actual definition of a drink is likely larger than half a pint of beer.
Always interesting to see different perspectives, visiting my aunt (almost 70yo) it was half a liter of beer with lunch, and another half with dinner as kind of just a normal thing. Maybe not every day but also some days you'll have another after dinner. Nobody thinks twice about that level of consumption, there is zero negative stigma.
Even if someone gets a little drunk most evenings it's not thought of as a problem, work meetings are had over beers after lunch. Life expectancy exceeds that of the United States, so personally to have someone say 8 beers a week is excessive is mindboggling to me.
I think a bog standard IPA is about 2 “units” of alcohol. I like the old alcoholic “I’ll only have one or two” and it’s a wine glass filled to the brim or a mason jar half full of whiskey
You probably wear suits that fit and the confidence probably shows through. Not to stereotype, but I suspect a large number of developers have one or two suits they wear for job interviews, weddings and funerals, and they bought them long enough ago that they are too loose or too snug by now, and consequently feel uncomfortable when wearing them. At least this used to be me.
Personally, I feel uncomfortable in any suit, not because of size or fit, but because I must be wary not to make it dirty at all, otherwise cleaning is a pita. Say you want to go for lunch with coworkers and someone orders spaghetti/pasta. Great ... I rather wear something simple, so there is no problem, if something unfortunate happens.
>The ML world being nearly entirely in Python, much of it untyped (and that the Python type system is pretty weak) is really scary
I think this has a ton to do with the mixed results from "vibe coding" we've seen as the codebase grows in scope and complexity. Agents seem to break down without a good type system. Same goes for JS.
I've just recently started on an Objective-C project using Cline, and it's like nirvana. I can code out an entire interface and have it implemented for me as I'm going. I see no reason it couldn't scale infinitely to massive LOC with good coding practices. The real killer feature is header files. Being able to have your entire projects headers in context at all time, along with a proper compiler for debugging, changes the game for how agents can reason on the whole codebase.
reply