Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
MrPowers
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
17 ms
·
121.
▲
by
MrPowers
5y ago
Land taxes are based on the value of the land, not the size of the land. The property tax system already performs land value assessments. Land taxes are highly progressive. Note that land taxes are only assessed on the value of the land, n
122.
▲
by
MrPowers
5y ago
The deadweight loss of taxation is much lower for a land tax than an income tax. The deadweight loss is the economic resources allocated to complying with the tax. The armies of tax lawyers would be able to perform other economically prod
123.
▲
by
MrPowers
5y ago
All programming languages should have a PyArrow-backed DataFrame data structure. After working with DataFrames in Python or Spark, you'll never want to go back to manipulating nested arrays. Ruby, Go, Scala, Java should all build thes
124.
▲
by
MrPowers
5y ago
Do you have any resources or particular qualms with Thomas Stanley's research techniques? To me, it seemed like he was drawing conclusions based on data. Interested in why you think it was "crud masquerading as science".
125.
▲
by
MrPowers
5y ago
The book The Millionaire Next Door provided data to support "Seek wealth, not money or status". The book was written by finance professors who examined the characteristics of actual millionaires. They found a lot of people who ap
126.
▲
by
MrPowers
5y ago
Tesla can grow and sell tons of cars and the stock price might still fall. Cisco Systems is a great example of a fantastically profitable business with a stock price that's still below peak. It's an incredibly successful company
127.
▲
by
MrPowers
5y ago
I agree with the sentiment that comparing Tesla's market cap with traditional car companies is inaccurate. For certain companies, market capitalization is a good proxy for enterprise value, but as this poster is mentioning, that's
128.
▲
by
MrPowers
5y ago
I built daria ( https://github.com/MrPowers/spark-daria ) to make it easier to write Spark and spark-fast-tests ( https://github.com/MrPowers/spark-fast-tests ) to provide a good testing workflow. qui
129.
▲
by
MrPowers
5y ago
Here's a quote from Reynolds, one of the creators of Spark: > The primary issue I can think of comes from the lack of pattern matching. Kotlin’s language designer left out pattern matching intentionally because it is a complex featu
130.
▲
by
MrPowers
5y ago
I am excited about Scala 3. Scala is a powerful tool that can make teams hyperproductive. Here are the tooling shifts that can broaden the Scala 3 userbase: * Community switching from SBT to Mill * Community agreeing on automated code for
131.
▲
by
MrPowers
5y ago
> In 2019, Uber claimed $4.5 billion in global operating losses (excluding the US and China) for tax purposes – in reality, it brought in $5.8 billion in operating revenue This sentence makes me feel like the article author does not unde
132.
▲
by
MrPowers
5y ago
Thanks for sharing! I actually chatted with No Starch Press and met with them in their offices. Really nice people and amazing vibe. I am working on a PySpark book and will reach out to No Starch again based on your comment.
133.
▲
by
MrPowers
5y ago
I wrote a couple of books recently and have some tips for aspiring authors: * Start by blogging. Try to get thousands of daily pageviews with an average time on page > 5 minutes. Objectively verify you're able to write content tha
134.
▲
by
MrPowers
5y ago
Good to know, thanks for clarifying! Parquet predicate pushdown filtering is even more powerful than I thought!
135.
▲
by
MrPowers
5y ago
Yea, haha, that was worded too strongly. Should have said "more folks are using Parquet for data workflows and adoption seems to be increasing". CSV is better when you need something mutable / human readable, so they'l
136.
▲
by
MrPowers
5y ago
Deequ is the big data / Spark alternative for similar functionality in case anyone is interested: https://github.com/awslabs/deequ Looks like a great project and wanted to highlight why projects like these are muc
137.
▲
by
MrPowers
6y ago
I agree with you. Some tax law is from lobbyists & some is from tax nerds. Congress doesn't have enough time to read laws (and they simply aren't qualified for tax law - some requires CFA-level understanding of accounting &#
138.
▲
by
MrPowers
6y ago
Definitely don't agree with "Transfer pricing requires pre-emptive approval from the tax authorities involved". Maybe you're thinking about an Advanced Pricing Agreement. Transfer pricing is the pricing of goods and se
139.
▲
by
MrPowers
6y ago
To be clear, the "esoteric mechanisms" are invented by some relatively low paid IRS tax nerds and then signed into law by congress (people that certainly cannot understand the rules and probably don't even bother reading them
140.
▲
by
MrPowers
6y ago
> you need to actively look for loopholes, and get access to tax lawyers and accountants in order to do something like a 'double-Irish' What is the alternative? Multinational companies have to file tax returns in tons of diffe
141.
▲
by
MrPowers
6y ago
Not sure I follow. If US Multinational charges license to Italian Subsidiary, then the pricing impacts the profit allocation between the two entities. The IRS will say "the license should be really high" so the profits in the US
142.
▲
by
MrPowers
6y ago
I normally find the Hacker News audience to be balanced on issues, but for tax policy, seems like strong opinions are held without a basic understanding of facts. In this instance don't have strong opinions if: * You have no idea what
143.
▲
by
MrPowers
6y ago
There are international agreements on taxation. Tax authorities can also work with taxpayers and agree on the transfer price for goods and services in advance. Tax loopholes typically refer to ambiguous rules. In this case, the IRS was ex
144.
▲
by
MrPowers
6y ago
Python's packaging is a lot better with Poetry. Use Poetry and build the wheel with poetry build. The lockfile lets you create an exact virtual environment. Like you mentioned, wheel files are not nearly as customizable as JARs. You
145.
▲
by
MrPowers
6y ago
Yea, JFrog announced the May 1st shutdown of Bintray on Feb 3rd: https://jfrog.com/blog/into-the-sunset-bintray-jcenter-gocen... 3 month warning that they'd blow up the ecosystem. I have no idea why a company tha
146.
▲
by
MrPowers
6y ago
Java has gotten better over the years (borrowing good ideas from Scala) and is maintainable. They make a huge effort to keep Java backwards compatible, even across major releases. I depend on some ancient Java libs and they still work just
147.
▲
by
MrPowers
6y ago
It's more unstable that ever with Scala 3 around the corner (which is basically a new language) and the May 1st Bintray shutdown that'll be massively breaking for the ecosystem. SBT builds were failing the other day from the Bint
148.
▲
by
MrPowers
6y ago
From what I've seen, tax loopholes exist mainly because of incompetence. The Section 482-7 regs were written by the IRS to control how intellectual property is shifted offshore. The valuation methods outlined in the article were quest
149.
▲
by
MrPowers
6y ago
Land taxes are different than property taxes. Property taxes are on the building, land taxes are just on the land. Land isn't a perfect tax, still needs to be levied as you mentioned, but it cannot be hidden and incentivizes land owne
150.
▲
by
MrPowers
6y ago
Yes, you as an individual can get tax credits. If you own foreign stocks, you pay foreign taxes and get a US tax credit for the foreign taxes paid. Or if you lose money on a stock (and realize the loss by selling), you get tax credits. So
More ›