Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
adverbly
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
37 ms
·
571.
▲
by
adverbly
6y ago
> ratchet effect First time I've heard of this term. Reminds me of what happens with house price inflation due to lower interest rates.
572.
▲
by
adverbly
6y ago
Some nuggets here. Totally agree with the second order effects of incentives and milestones. The author seems a bit overly enamored by SV though. It's really not all its cracked up to be. Stop trying to be SV. SV has produced some trul
573.
▲
by
adverbly
6y ago
I'm curious to get people's take on how they are approaching this migration. One thing that I am considering is keeping WhatsApp around without leaving any of my groups, but making very strong efforts to only send new messages thr
574.
▲
by
adverbly
6y ago
Give CamelUp a look! Great for kids or bigger groups! You also might want to give "Liar's Dice" a look - some of the variants are similarly great for bigger groups, and still rely heavily on strategy.
575.
▲
by
adverbly
6y ago
lol. Well at least one of us is I guess... So point taken.
576.
▲
by
adverbly
6y ago
Focusing on the individuals is WRONG. Stop doing it! People are people, and - as far as capitalist societies go - money is success, and "skimming" is unfortunately the dominant and easiest strategy to get it. Blaming the individua
577.
▲
by
adverbly
6y ago
(difficulty to execute hardest to easiest) - Starting the right business - Picking the right businesses to invest in - Investing in Crypto(used to be easier than it is now) - Convincing shmucks to invest in the wrong business - Investing in
578.
▲
by
adverbly
6y ago
I don't know why more products dont follow the software pattern of cycling the alphabet. Pick a unique new name each time, and just run it down the alphabet for the first letter of the name. So easy.
579.
▲
by
adverbly
6y ago
I found it annoying until I added a git hook and got used to auto correcting. Its low friction for me now, and well worth it imo. If you run into a very large number of linting errors then you might naturally write code in an unidiomatic st
580.
▲
by
adverbly
6y ago
I agree with you that I find it more readable. However, I've spent a long time living with this exact RoboCop setting so I am probably very used to looking at this style. The author mentioned scanning by indentation which is likely som
581.
▲
by
adverbly
6y ago
> Green lights I believe are also illegal. for R/G colourblind I guess?
582.
▲
by
adverbly
6y ago
What you are describing is tax fraud and is already possible in the current system.
583.
▲
by
adverbly
6y ago
lol ya tracking everyone would be really dumb but you don't need to do it. Many countries already have self reported high value assets as part of their tax filing process. Pretty easy to add a self-reporting step "Do you have over
584.
▲
by
adverbly
6y ago
In many countries, reporting high value assets is already part of their tax reporting and is not an added step. For countries that don't, the added step could be as simple as a single checkbox: Do you have over 25M in assets? I'd
585.
▲
by
adverbly
6y ago
Different issue. What you are describing is tax fraud and is already possible in the current system.
586.
▲
by
adverbly
6y ago
I stopped reading when I hit: > wealth tax is a colossal privacy and administrative burden on _every single taxpayer_ I have not seen a single wealth tax proposal that doesn't have a gigantic cutoff where it would do nothing for 99%
587.
▲
by
adverbly
6y ago
I'm not sure that the author's assessment that "36% accuracy rate" ... "there is no reason to doubt his date of 2029" follows. I'm no statistician, but my intuition tells me that if you drop 1/3 accur
588.
▲
by
adverbly
6y ago
Is it not possible to design a for-profit, open source, transparent architecture that could allow for a trusted review mechanism? If its clear that there are no code paths or questionable "algorithms" which would permit sale-to-a-
589.
▲
by
adverbly
6y ago
> To me, it’s like driving a van in very windy conditions. Love that analogy!
590.
▲
by
adverbly
6y ago
> The best people at playing the etiquette game are the ones you shouldn't trust. I can totally see that this is true in some circles(e.g. sales people). But in other circles (e.g. education/some types of customer service), eti
591.
▲
by
adverbly
6y ago
It actually does take that much thought though! Social dynamics are extremely complex to model, and they are way more important than we think. There is a lot of complex game theory and metagaming going on. Most of it goes over my head, but
592.
▲
by
adverbly
6y ago
As someone who finds many "social etiquette" practices painful, cringe-worthy, and the cause of many communication breakdowns, I would actually like to learn more about the benefits of social etiquette, and how I can improve on it
593.
▲
by
adverbly
6y ago
3 - I'd agree that it is not really a skill so much as an effect with many possible actionable causes, some of which can be trained. I think you and OP are actually coming at this from a similar angle if you indeed don't really ca
594.
▲
by
adverbly
6y ago
This
595.
▲
by
adverbly
6y ago
Clearly whoever wrote this android integration didn't hire enough high-quality ML PhDs to reach the necessary benchmarks for full-screen defaults.
596.
▲
by
adverbly
6y ago
Paul is surely trolling, no? This isn't even close to realistic modeling. Most startups will spend many, many years worth very little. Well under the wealth cap floor. This means that if you don't own a unicorn then you'll li
597.
▲
by
adverbly
6y ago
It physically hurts me to think about how dead accurate this is. You tell someone exactly why something won't work? Get rewarded with a door to the face. You save precious time because you care about actually building something of valu
598.
▲
by
adverbly
6y ago
I can't help but feel pity for the author here. They've correctly been able to see through all of these tactics and psychological hacks employed by the marketing industry, but then they've gone on and tried to better themselv
599.
▲
by
adverbly
6y ago
Another one who fell for the fake news right here. Part of me wishes there was a better way to model retractions, corrections, and certainty of claims within social media platforms. I almost feel like by default a post or tweet should be ma
600.
▲
by
adverbly
6y ago
I actually really like this idea. The model sort of reminds me of a cooperative. There is a cooperative for Outdoor Clothing in Canada and it had a reasonable amount of success.
More ›