Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
seadan83
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
79 ms
·
871.
▲
by
seadan83
3y ago
For the data we have, year on year increase of 30x, from 3 to 90, is not enough orders of magnitude for you to object saying "sky rocket" - that you call it a scare campaign? It is part of the point the city is missing data, these
872.
▲
by
seadan83
3y ago
I like that litmus test. Another thing I notice is organizations where PMs are assigned to projects as they come up. Meaning, "please cut these corners so I can launch sooner, get all the credit, then move on." In this case obviou
873.
▲
by
seadan83
3y ago
I don't believe contracting of any type us needed for a feature factory. It is more when product and project managers hijack the roadmap (or even create a roadmap at all) and then March along adding things for the sake of delivering st
874.
▲
by
seadan83
3y ago
I hear this, comments generally should not draw attention to themselves. For this, short & terse win. I routinely look to cut any unnecessary words from comments. It was the most painful code review where I asked someone to remove a jok
875.
▲
by
seadan83
3y ago
Focusing on the technical issue at hand - can anyone tell the exact issue at play? Is this an issue of client retry logic that on a 429 without backoff? Hence, the self-ddos?
876.
▲
by
seadan83
3y ago
The real problem with the columbia and salmon is that the columbia is becoming an unviable spawning ground. That entails a full collapse of the species in the region and any other species depending on salmon for food. Doesn't matter wh
877.
▲
by
seadan83
3y ago
The columbia is a spawning ground, but it is spawning less than 10% what it used to. During the heat dome many salmon were getting burns on their body. The water level is too shallow to keep the water temps down. This is due to damming, irr
878.
▲
by
seadan83
3y ago
Ater Chernobyl birth defects in humans significantly increased. Similar in animals. Usually those defects (not an expert,just my understanding) is in bad formation, rather than novel formations [eg, an arm fails to develop properly rather
879.
▲
by
seadan83
3y ago
I feel there is a huge over indexing on Clean Code and self describing code. Essentially are we suffering from a journeyman problem where the commentary we focus on are written by those that are new and have trouble understanding basic synt
880.
▲
by
seadan83
3y ago
From the data that I could find, it's perhaps at best 10% of all tech companies that even go public (and very possibly as few as 1%). It's far more reasonable to assume a company is the 90% or 99% of companies that do not go publi
881.
▲
by
seadan83
3y ago
Mock vs a fake. The difference lies in where the behavior is specified. In a fake, the behavior is hardcoded or simple. With a mock, it is each test that says what the behavior is. The two scale differently. Mocks are per test, fakes are on
882.
▲
by
seadan83
3y ago
Talking over other people and being throttled down is not censorship. Being talked over, is censorship. Trolls don't have to be right, just louder. Their goal (from my perspective) is to pollute the information space. With large enough
883.
▲
by
seadan83
3y ago
Can you provide a reference where I can learn more?
884.
▲
by
seadan83
3y ago
Living standards in the US are reverting to the global mean well enough on it's own Life expectancy is not even top 20 and dropping. Social mobility is also not top 20, cost of healthcare is up there, poverty is up there.
885.
▲
by
seadan83
3y ago
(B) is an interesting take. Who is to say "the other" guy is not already polluting to their max. If one country lowers its emissions,it is what aboutism to then suggest it must be a zero sum game and that other countries would inc
886.
▲
by
seadan83
3y ago
Code is fast to write, slow to modify. That 10x dev is just in the initial write phase; anyone coming in later to modify that code will appear to be slower.
887.
▲
by
seadan83
3y ago
One more thing that occurred to me, most US retirees still own cars, and still pay car tabs! That is kinda the issue, payment for driving is spread across US society rather than being the burden of those that use cars. A strict counter-exam
888.
▲
by
seadan83
3y ago
"Low-income Americans face higher payroll tax rates than rich Americans. Americans with less than five-figure incomes pay an effective payroll tax rate of 14.1 percent, while those making seven-figure incomes or more pay just 1.9 perce
889.
▲
by
seadan83
3y ago
> Additionally, if you are not using a car for commute, you're likely to be in the lower tax brackets and thus not paying (much) in taxes anyway. Doesn't this directly contradict your point that car commuters are subsidized?
890.
▲
by
seadan83
3y ago
> The fraction that doesn't work is either too young to pay taxes anyway, or they had used road commutes before they retired. Everybody else are within the margin of error. The argument that either everyone was already a tax-paying
891.
▲
by
seadan83
3y ago
Only a fraction of the US population work and thus commute. Of those that commute, that 80% do not have equal road usage. Even within that community, there is a subsidy going on. So, yes, there is a ton of subsidization going on (and this i
892.
▲
by
seadan83
3y ago
> 1. It has a higher CO2 footprint than small/medium EVs. If mass transit _and_ also EVs are incentivized over petrol cars, that is not bad [1] > 2. Transit forces people into smaller and denser housing, resulting in suboptimal l
893.
▲
by
seadan83
3y ago
Mr Zuckerberg cited a study showing new employees that started as remote performed more poorly. Does anyone know if the level of new hire training and onboarding is the same? My experiences with onboarding is you generally sit with someone
894.
▲
by
seadan83
3y ago
I guess also worth mentioning, modern forestry uses tree farms and has been for a while. Those tree farms burn differently than natural forest and are super dense. Getting more logging into the mix, or more wood mills is not necessarily a s
895.
▲
by
seadan83
3y ago
Agree with your points, though your premise is incorrect. Fire suppression started in the 30s [1] and became industrialized following WWII [1] https://foresthistory.org/research-explore/us-forest-service...
896.
▲
by
seadan83
3y ago
Certain areas burn with differing frequencies (naturally). Eg, coast range burns every few centuries while cascades and sierras it is more like every other year. Sorry for no references, so please have a large grain of salt with those state
897.
▲
by
seadan83
3y ago
Indeed. Though modern fire suppression is a huge factor. AFAIK fire suppression became industrialized following world war II (bomber type planes able to deploy large amounts of water, better radios and respirators). Even today, the mandate
898.
▲
by
seadan83
3y ago
Be careful, for smoke you want a N99. A N95 will help, but don't be fooled that it is adequate protection.
899.
▲
by
seadan83
3y ago
Ait is bad enough, skipping indoor exercise is probably best unless running a quality air filter. My learning of that was after taking days and once weeks to recover from smoke exposure. I don't think the gravity of how bad it is reall
900.
▲
by
seadan83
3y ago
Java by default allocates a healthy percentage of available RAM. It releases memory if a large percentage is unused, and it only runs full GC when a large percentage is used. The trick is therefore to decrease the initial max memory. Other
More ›