ishansharma

https://ishan.co

I rarely blog about interesting problems that I think could help others and sometimes gaming.

The article has a refers to a few sources towards the end including:

  One such study found that children who grew up in homes with between 80 and 350 books showed improved literacy, numeracy, and information communication technology skills as adults.

May I ask the $2000 desktop configuration with 2x the speed?

Of course Apple chips won’t work well for gaming, but what other benchmarks will this $2000 desktop win?

Re: India. It’s very common. Large IT firms like Tech Mahindra, Cognizant etc do it. I had offers from a company where they wanted around 1 year salary of signed check. If I leave before 2 years, they use the check.

Some companies go a step ahead and keep your original documents (Degree certificate etc) so you can’t apply elsewhere after accepting offer. It’s ridiculous.

I agree with most of your comment and it's making me rethink some of my opinions. But is this accurate:

Wealth creation isn't zero-sum, just because someone builds a business and creates wealth doesn't mean they're taking it from others.

I'd say that's exactly the model that some of the startups are taking. Uber and Lyft come to mind. A part of their wealth is coming from the pockets of workers who have to work for lower wages than before and get hit by depreciation that they didn't account for when earning.

I admit that politics is mixed into my opinions here.

It's just that based on current political climate, I think it's hard to keep politics aside.

My thoughts about pg have essentially gone through this flow: - Start reading some of the old essays in college, agree with most of the points - Start working, the essays slowed down. I read old ones once in a while, still mostly agree - Start following pg on twitter. There's some stuff I don't agree with but it's alright. - See a lot of stuff that I don't agree with [1]

(in between somewhere, I got convinced that keeping politics on sidelines is not an options)

Then I see the essay on wealth tax. My read is that main argument is that the tax compounds. But I don't see any alternative arguments. So I see his position as against the transfer of wealth and societal equality and I don't agree with that.

Maybe I'm being too opinionated or not reading deep enough, but that was my line of thought.

https://twitter.com/Pinboard/status/1130241349935149056

It’s my opinion but that was a very one sided view. And I need to read more on this but I don’t think the way he pointed it out is how anyone is proposing wealth tax.

Couldn’t agree more. I read all his essays in my early 20s and became a huge fan. Not just startup advice wise, but general big picture stuff wise too.

Then I saw his recent tweets and especially the essay about wealth tax, I was a bit disappointed. But as you said, his position is well earned. But I don’t see any useful real life advice from him anymore.

Out of curiosity, what could have Microsoft done to cause this?

I'm not aware of the acquisition details. But did they make GitHub switch to Azure or make significant changes to infrastructure?

I've seen several people blaming Microsoft for GitHub outages and am trying to understand why this is a common thing.

Can you explain further? I’m curious how Bitcoin benefits from it?

Ah, the planned obsolescence argument. It used to be invoked all the time with software (updates are making it slower) but Apple just released software updates to fix GPS on several older iPads.

Maybe, they are focusing on other things like thinness and light weight which results in making gadgets less repairable instead of a big planned conspiracy.

I use default Mail.app for macOS. I tried the fancy clients like Spark etc but settled on Mail finally.

I prefer desktop clients because I have 6 email accounts and managing all of them in tabs is just a headache.

I'm pretty sure the ads are not put up by the author by WordPress.com. According to their pricing page[1], they "sometimes" display ads on blogs to cover costs.

https://wordpress.com/pricing/

Or apps for that matter. Having used the app for 6 months, I just keep a browser tab open these days, no use of feeding all the RAM to Slack.

This is a huge clickbait. Does not mention that fine is in Australia and fine is just $7 mil.

I am not aware of Readdle doing anything bad or having a bad reputation. How is it bad for privacy, just asking because I’m a regular user and would switch if it’s not good for privacy.

Spark for macOS and iOS can do something similar for any mail accessible via IMAP/POP/Exchange.

I guess the point of the article is exhorbitant fees. An ice pack and a bandage in ER don’t have to cost $5k plus.

Thank you. I will consider doing that. I will have to create or find a relevant dataset that's small enough for those concepts but I guess that'll help rather than jumping straight to random forest or gradient boosting.

Thanks for sharing. I will take a look at Cython implementation, that should help me learn a few things.

Since we are looking at entropy with respect to class (whether we want to eat or not eat), entropy for blue & kit kats is 0 as we don't want to eat them at all. So there's no randomness.

They are included in the IG formula but since they are multiplied by 0, they don't have any effect.

I am considering doing a more programming focused part, where I take a big dataset and implement a decision tree using Scikit Learn. After that, I will look into ensembles.

If you have any specific questions about ensembles, feel free reply here or send me a mail to contact(at)<myusername>(dot)com.

Ah, I guess I could have used a Gist for that. Thanks for the suggestion, let me see if I can update the article.

Thank you for the kind words. My goal was to make the explanations as simple as possible.

Using them to get a feel about dataset is an interesting use case and I haven't tried that. It does sound interesting though. Something to do with my next ML problem I suppose. :)

Just skimmed through your article(class final exams tomorrow), it is informative. Thanks for sharing.

Hi,

Author here. Thanks for the links. I originally planned to try a bigger dataset but didn't want to make the article too long, so left out the optimizations.

I had to do the same exercise for my ML class and had similar results. Random pruning (mostly) worked great for me.

IMHO, While DTs do overfit a lot, they are a great starting point for beginners because of their (relative) simplicity. Better to start light and then introduce the math heavy neural nets and SVMs.

Duckduckgo. It's good enough now, I am slowly moving away from Google products (only Gmail left). Rarely, I will use !g to go to Google but it has gotten lesser and lesser recently.

I’m in the same boat. I don’t understand if it’s just all the smart people at Twitter don’t know how power users use Twitter or they just choose to ignore all this.

I was an Apple music since launch(till a month back). But once I tried Spotify and the playlists, I switched instantly. While Apple Music is a competitor, it's not everywhere like Spotify is and their recommendations are horrible. In all time using them, I discovered 2 songs that I liked. With 4 weeks on Spotify, I'm already at 25!

Would be great if you could post the notes. I'm definitely interested. :)

Now just putting conspiracy theory hat for a moment, but can it be the real reason behind Travis walking out? Board got to know that this would come out and got rid of him to do some damage control?