Hacker News new | past | comments | ask | show | jobs | submit | skorgu's comments login

This comment section is the most absolute proof I've ever seen that absolutely nobody reads the article before commenting.

Hint: it's not about integer types at all


Then please enlighten us and tell us what you feel the article is about rather than a snarky "rtfm" comment.

Because I started reading. I got pretty far, and it was _still_ talking about integer types. The article is very long, and I gave up, not having any idea why I should keep reading. So saying the article isnt about them "at all" is wrong. Maybe, the int type thing is an intro to some other thing? But thats a long intro. So, tl;dr, at least not all the way. Why don't you tell us the point so we have desire to actually read it.


Conveniently, there's a table of contents at the start that hints at the later topics:

  Different kinds of numbers
  Conversions and type inference
  Generics and enums
  Implementing traits
  Return position
  Dynamically-sized types
  Storing stuff in structs
  Lifetimes and ownership
  Slices and arrays
  Boxed trait objects
  Reading type signatures
  Closures
  Async stuff
  Async trait methods
  The Connect trait from hyper
  Higher-ranked trait bounds
  Afterword
So 17 sections, 16 technical (excluding the Afterword) and only the first two focus on the int/float stuff. So to answer your question:

> Maybe, the int type thing is an intro to some other thing?

No, it's not an intro, it's just the first two sections.


> I hope I was able to show, too, that I don't consider Rust the perfect, be-all-end-all programming language. There's still a bunch of situations where, without the requisite years of messing around, you'll be stuck. Because I'm so often the person of reference to help solve these, at work and otherwise, I just thought I'd put a little something together. [emphasis mine]

As you can also see from the table of contents at the top of the article page, it goes through a great breadth of topics. And you can use that table of contents to jump to the afterword.


Since you don't seem to have commented on the article without reading it, my comment was not addressed to you. I'm not the article police; if you find something too long to read, decide not to read it, and move on with your life, we have no quarrel.

Honestly, I don't even mind the folks who do comment without reading. Comments are free, do what you like. I just find it mildly hilarious that 100% of the comments are about (the first) 3% of the article.


Reading all of it is a big ask for a blog post that is this long


Then why bother commenting?

I don't understand the urge to comment on something one hasn't read.


I did read the beginning of the article and I wasn't too impressed. So I went through the comments, looking for bits of wisdom that might hint reading the whole article was worth it. Still, I wasn't impressed.


I see multiple people commenting on why they didn't read the whole thing, which is a perfectly valid topic of conversation.


It is categorically about integer/numeric types in part. Why do you feel that "at all" is an appropriate qualifier?


I'm curious how this can both avoid the average-of-averages problem (presumably by using the original full-rate data to compute multiple aggregates) and also supports backfilling. Is there a danger of the full-rate data expiring and having a different behavior for backfills past that horizon? Or am I wholly misunderstading both these features?


(NB: post author here)

Great question. We support average of averages by storing the intermediate state of the aggregate (for average that's the sum and count) so we could cleanly re-aggregate.

Eventually, we'll be able to incrementally update the aggregate if we backfill even if the raw data is no longer available. That's not implemented yet though, so backfill only updates the aggregate if the raw data is still around by re-computing the intermediate state of the aggregate off the raw data for affected buckets. For most cases that isn't actually an issue since most people have a longer data retention period than backfill horizon.


Thanks for the answer! I'd love to know more :) Also, I'm not following, how you guys deal with issues with unique counts? For example, lets say you've got 100 unique visitors on Monday and 100 on Tuesday. The unique visitors for both days might be anywhere between 100-200 and averaging counts between days doesn't work.


Not sure about this specific implementation but normally you handle this with approximations that support merging. i.e HyperLogLog You can merge 2 HyperLogLog counters to maintain proper distinct counts.


Yep!

And in fact, that's exactly what TimescaleDB supports - things like hyperloglog to support approximate count distinct, including as part of continuous aggregates. [0]

This blog post - "How PostgreSQL aggregation works and how it inspired our hyperfunctions’ design" - provides a really nice description of how our the API design of some of our analytical functions are motivated by the ability to "split" processing into the "pre-aggregation" and "finalization" steps, with the blog post focusing on the example of percentile approximation. (I think it was on HN a while back as well.) [1]

[0] https://blog.timescale.com/blog/introducing-hyperfunctions-n...

[1] https://blog.timescale.com/blog/how-postgresql-aggregation-w...


Awesome, thank you!


You avoid average-of-averages by storing multiple summaries. For example, you don't compute and store average, you compute and store sum and count.


The flipside: you can also be lucky enough to have borrowed a ton of money just in time for the crash to drive everyone else out of business: https://www.vox.com/new-money/2017/4/5/15190650/


I'm ... shockingly interested in this bank letter ghostwriting thing. Could you say more?


There exists a message board called "Credit Cards and Consumer Debt" within the Motley Fool. I commented there for a few years on a variety of subjects, under a pseudonym (which is laughably transparent to anyone who knows me but please don't take that as a challenge, HN).

I was originally there because my credit report suddenly got $100k+ of debt added to it due to a series of errors. I researched the CRA and FDCPA and, in the process of doing so, found that forum. Many folks with different fact patterns needed relatively similar resolutions to mine or ones which were predictable based on information I had access to, so I started writing replies like "Yeah just write your bank and tell them $FOO." It turns out that the types of folks who end up deep in credit card debt often have some difficulty in banging out a quick professional letter. I didn't, and I had a lot of free time on my hands, so I drafted perhaps a few hundred letters.

This is one of my more esoteric hobbies, but it has been occasionally useful over the years.


Congrats on the launch!


AIUI, RSUs are treated as normal W-2 income when they vest.


Right. Your cost basis is the price at vesting, so it's the same as if you bought the stock at market with cash for the same price.


so is the capital gains tax only on the profit from the stock? wouldnt it be better just to sell instantly and not have so much of your money tied up into one asset anyway? Sorry if these are basic questions


Yep and yep. Not basic at all, this stuff is only simple once you already know how it works.


What about ion engines? I'm extremely not sure I have this math right but NEXT claims 4190 second ISP [0]. Plugging that into the rocket equation I get:

  iss_mass = 419600
  sep_isp = 4190
  delta_v = 5748
  htv_payload = 3310
  falcon_launch_cost = 62000000
  total_mass = iss_mass * math.exp(delta_v / (9.8 * sep_isp))
  fuel_load = total_mass - iss_mass
  dragon_launches = math.floor(fuel_load / htv_payload)
  total_cost = dragon_launches * falcon_launch_cost
I get:

  Total vehicle mass: 482,646.89 kg
  Fuel to add: 63,046.89 kg
  Dragon 2 launches: 19
  Total cost: $1,178,000,000.00
Not what you'd call cheap but at least in the realm of the possible. I'm assuming the ISS' solar panels (~120 kW [1]) is sufficient to power the thruster, that a dragon launch costs no more than a falcon launch (certainly false), and that the ion engine is already on board the station.

  [0] https://en.wikipedia.org/wiki/NEXT_(ion_thruster)
  [1] https://en.wikipedia.org/wiki/Electrical_system_of_the_International_Space_Station


If there's only one engine then that's 236 mN thrust according to your [1]. Lets call it 4.12 N - put a few thrusters on or improve the technology.

The station mass is 419,455 kg. F = ma so the acceleration is 1 µ G.

It will take a very long time to get to Mars that way.


About 18 years assuming I'm using the right formula:

http://www.wolframalpha.com/input/?i=acceleration+formula&ra...


It's not a direct oil subsidy but roads are ~50% subsidized: http://usa.streetsblog.org/2013/01/23/drivers-cover-just-51-...


Car ownership is extremely broad across all income groups in the United States, see http://www.newgeography.com/content/002666-how-lower-income-.... The overwhelming majority of income taxes are paid by high earners, with the the top 10% paying close to 70% of all income taxes. If roads are subsidized by income taxes in addition to gas taxes, they are being paid for in an extremely progressive way. Effectively, roads are provided for free to over half the population.

Or maybe I misunderstood and you are trying to make a different point?


Half the population doesn't pay gas taxes? You're aware that they do. That was a silly thing to say. Gas taxes may be trivial to the portion of the population that pays most of the income taxes, but I doubt the bottom half thinks of them that way.

You need to show what portion of the road cost is paid for by income taxes before you can claim the rest.

I don't understand why you brought any of this up. A subsidy is a subsidy no matter who pays.


You are correct that I made an error in my calculation, but my point still largely stands.

I took the parent at face value for the claim that 50% of roads are funded by use fees. That means the other 50% are paid for with general income taxes. Nearly all income taxes are paid for by the top half of earners. If we assume that everyone drives about the same amount, then the bottom half of the country gets the full value and use of roads while only paying for 25% of the total cost.

I bring this up because the parents point seemed to be that road subsidies are an implicit subsidy to oil companies. But people need to get around somehow, and the current way of funding roads should be perfectly compatible to conventional morality.

At the same time, I oppose all subsidies and can only imagine the amazing methods of transport that would exist if the government did not actively interfere and manage this market.


Tesla doesn't benefit from this, too?


I don't know why they sell the 3mgs either because nothing less than three of them has any effect on me. Thank god for 10mg tablets.


Hey, that's a really useful reference, thanks. I actually learned something about OCaml from it!


Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: