Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
tchalla
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
66 ms
·
871.
▲
by
tchalla
6y ago
Most situations aren’t black and white. Crypto (in particular Bitcoin) is an investment asset. Every asset comes with risks. The way to adjust for the risk is to have the right proportion of your investment into the asset. It’s the same for
872.
▲
by
tchalla
6y ago
I also want to make an additional "Thank You" note for the author on the lovely "The Illustrated Word2Vec" [0]. I wish every concept Machine Learning or otherwise would follow such a framework. [0] https://jal
873.
▲
by
tchalla
6y ago
> This costs money and is not free I have been a reviewer at Elsevier, Science and many of this journals for a better of the decade. I personally know the editors of many journals including IEEE, ACM, Elsevier, Science etc. They don'
874.
▲
by
tchalla
6y ago
> Facilitating peer-review, being selective, indexing content, archiving and copyediting does indeed cost money. Publishers only archive and index. The rest is done for free by scientists.
875.
▲
by
tchalla
6y ago
There are other OS with a larger percentage of devices installed with other app stores possible. How many privacy focused stores do we see with privacy focused Facebook clients? How many of the users exercise those privacy options and give
876.
▲
by
tchalla
6y ago
> Sorry, we can only complete resumes for experienced technology professionals in the US and Canada who've accepted our Terms of Use. Welp! Is there any reason why you are restricted to US and Canada? I can understand 2+ years requi
877.
▲
Great Impractical Ideas in Computer Science: PowerPoint Programming
(youtube.com)
2 points
by
tchalla
6y ago
|
0 comments
878.
▲
by
tchalla
6y ago
Signal can not be bought by someone unless the regulators approve it. WhatsApp was bought only after an explicit approval by regulators.
879.
▲
by
tchalla
6y ago
> So 9 cents in India is equal to 30 cents in USA. Now, that is 30 cents per GB and not 1.91$ per GB. I couldn't get this from your earlier comments. I get it now.
880.
▲
by
tchalla
6y ago
> 9 USD = 9 * 70.394 = ~633.5 INR [3] > 633.5 INR = 633.5 / 21.3 = ~30 USD PPP. Thanks for taking the time to correct me.
881.
▲
by
tchalla
6y ago
"It is better to be vaguely right and than precisely wrong".
882.
▲
by
tchalla
6y ago
Let me once again repeat what I have said and please take some time to listen and understand it * You can not compare goods across two countries by simply taking the exchange rate, you need to normalise to account for different factors like
883.
▲
by
tchalla
6y ago
It's fine to take another normalisation metric - no metric is perfect. The point was that there is a normalisation needed. Far too often I see a comment which states, "Oh! It is just 1$ so cheap!". I chose PPP because it is
884.
▲
by
tchalla
6y ago
You can not simply compare a good in two different currencies without normalisation for disposable income, value and purchasing power. The value of 100$ in India is different than in the US. So obviously, some things in India will be cheap
885.
▲
by
tchalla
6y ago
> If PPP is 21.3 (2019 data) between USA and India, a Harley Davidson in India should be 21.3 times cheaper than a Harley Davidson in USA. It isn't. The cost is pretty much the same. So is the case with iPhones. Even if iPhones are
886.
▲
by
tchalla
6y ago
> PPP makes no sense when it comes to comparing data usage between countries. We are not comparing data usage, we are comparing the value of two different currencies. The value of 100 US$ in India is different from that in the US or the
887.
▲
by
tchalla
6y ago
They won't do that for the EU but will do it for the rest of the world.
888.
▲
by
tchalla
6y ago
> But the fact that it only costs 9 cents a GB is amazing. 9cents in India would be equivalent to 1.91 USD after taking into account Purchasing Power Parity (OECD 2019). Therefore, you are in effect saying 1.91 USD per GB is amazing. You
889.
▲
by
tchalla
6y ago
You can not compare two currencies without including the parities. In terms of the US dollar, India has a Purchasing Power Parity of 21.276 as of 2019 [0]. > Purchasing power parities (PPPs) are the rates of currency conversion that try
890.
▲
by
tchalla
6y ago
> how many non-academic taxpayers are trying to read articles that don't have institutional access through their employer or local public library? A different question to ask. Why should taxpayers not have access to the articles whi
891.
▲
You Don't Always Need the Best. What You Need Is “Good Enough”
(mr-stingy.com)
4 points
by
tchalla
6y ago
|
0 comments
892.
▲
by
tchalla
6y ago
You should measure inflation for your own personal environment and check if it holds true.
893.
▲
by
tchalla
6y ago
> You don’t have to be genius to understand what is going on: USA printed trillions and gave most of it to already rich people. It's also a good time to understand "The Cantillion Effect" > Namely, when you print money,
894.
▲
by
tchalla
6y ago
Since the site won't load, an alternative - https://sci-hub.st/https://journals.plos.org/plosone/article...
895.
▲
by
tchalla
6y ago
The major trade blocks ignore these actions because they profit from it. Once they won’t, we will see “democracy values” based moral statements and arguments.
896.
▲
What Big Tech and Big Tobacco research funding have in common
(venturebeat.com)
1 points
by
tchalla
6y ago
|
0 comments
897.
▲
by
tchalla
6y ago
She has an impressive h-index (51) and citation count (12000+) currently. Her pre-2013 count isn’t that impressive as current, but good. https://scholar.google.com/citations?user=PS_CX0AAAAAJ&hl=en...
898.
▲
by
tchalla
6y ago
> The publishers filter the OUTPUT: they select what is noteworthy and (ostensibly) good science produced with those funds. You may want to check this assumption. The publishers do not filter the output. It is done for free by the scient
899.
▲
by
tchalla
6y ago
I see where you are coming from. I would like to add two points. First, I have rarely seen only one approach to "it works". There are always more than one ways to "will it work?". Second, in order to get around the defin
900.
▲
by
tchalla
6y ago
> Similarly, I've found one of the most useful questions to pose for each option is "What would need to be true for this to be a good idea?". Thanks for mentioning this. I think, it's the principle of inversion. I can
More ›