Hacker News Reader

Comment by subhobroto | Hacker News Reader

My following suggestion is a very unpopular one, but it is this:

If you are a student, use the money you would have donated to StackOverflow to purchase a book or a tool that will help you to learn better and become an expert.

If you are an expert, contribute your time to StackOverflow (instead of money).

Money is easy to get compared to content. What helps StackOverflow grow is quality content.

I am also interested how they approach the subject of money:

> How much money we make is a direct proxy for how much we are helping our users. We focus on maximizing how we help users in order to make a great business. Very few companies have done this successfully, and we take great pride in the fact that we’re one of them.

Personally, I would love to see how this works out and I wish them well. Hopefully they can prove this model works well. Users need this model to work well for other businesses to adopt.

View on HN

>>book or a tool that will help you to learn better and become

I've tried that before, many times. Unfortunately when you are trying to do real work those books are many times useless. Example, I kept trying to build a newtonian physics simulator so I got a Statics and Dynamics mechanical engineering book. The book was a text book for Mechanical engineering students and yet the examples they teach you are for toy problems. I could not generalize them to real world scenarios. To actually build the physics simulator I had to scour the internet until I found something that actually explained what I needed in an easy to understand manner [1]. And that still was not enough. I still had to use more online sources to find out all the other pieces, like the separating axes theorem for collision detection.

So no, I would not rely on books anymore, at least not exclusively since they often do not provide a full picture. One last thing, books seem to be getting progressively worse in my experience. All of this is anecdotal of course.

[1]http://chrishecker.com/Rigid_Body_Dynamics

I have the same experience with books. I don't buy non-fiction for learning purposes* because it never worked. I tried with an IPv6 book which I read a few chapters of; I tried with another book or two that I don't even remember; I was gifted two books, one about making games and the other about programming paradigms, neither of which I ever read more than a few pages of... and yet I'm perfectly proficient by just playing around in spare time and looking up stuff I need and stuff I run into.

* I do buy non-fiction for amusement and casual interests though, like Predictably Irrational.

As an aside, buying a print reference book frequently feels counterproductive.

Even if I'm going to get a formal reference text, instead of online resources, I probably want a digital copy. Otherwise, I find myself going "I wish I could do text search for this thing..." and wondering why I didn't get the version that actually allows that. When I read for pleasure I enjoy the physicality of books; when I read for technical reference I demand the interface of a computer.

For books without ebooks from the manufacturer but for which I can find an illicit PDF or djvu, I usually buy the print one and read the digital version.

I've contributed a fair amount to Stack Overflow, but it's actually pretty hard in the niches I care about. Not because the questions are hard, but because people answer them so quickly: it's hard to find a question I can answer that hasn't been addressed well. That's a great problem for Stack Overflow to have, of course, but I do wonder if there's some easy way to monitor specific tags without having to constantly refresh the page...

Well if there is no work left to be done, so be it I guess? I don't think you should feel bad about that :)

There is, if you use an RSS reader. There are feeds associated with each tag. Try clicking any tag, and you should see a link where you can get an RSS feed of all the new questions that come in for that tag.

It's been there for a long time. ;-)

> Money is easy to get compared to content. What helps StackOverflow grow is quality content.

I do, >10k rep on the security.stackexchange.com site. I just think it's an amazing platform that I'd like to support with hosting and development costs.

> I do, >10k rep on the security.stackexchange.com site

That's amazing! Thank you for your contributions. I meant to imply that donating money is usually the least fruitful way to contribute, in my experience. It should be your last resort; definitely not the first one.

Example: One way SO can really gain from your security expertise is for you to think about how anonymous answers can be supported. That will be far more valuable than $$$.

I posted about this here: https://news.ycombinator.com/item?id=12961447

I would love to read your opinions about this crazy idea!

I have to agree. This happens on forums as well. I've Google searched something before and found people discussing my problem on some random forum. Maybe I know something that might help, but its a hassle to create an account to post.

I bet that happens on Stack Overflow too. People who'll never bother to create an account find a page about a problem through Google and they have something valuable they'd add it if didn't prompt them to sign up first.

I wonder how much trouble it'd actually cause spam wise.

I suspect there would be a lot of challenges with it but perhaps letting high rep users validate anonymous contributions would help with the spam. Reveal them only to authenticated users and you remove the incentive for SEO spam.

Contributing is wonderful, but those internet points (reputation) screws it up. An edit would give you 2 points if you have a rep < 2000, and contributes towards badges anyways. People would make pointless edits to a post. See http://stackoverflow.com/posts/21023161/revisions for example. There is no change between the original answer and the latest revision after 3 edits.

Excellent insight. I am hopeful SE will figure out a way to stop/reduce this arbitrage opportunity.

That example is all on Emil Vikström, who unilaterally accepted a bad edit from Akshat (with minor corrections of his own) after two reviewers had already voted to reject. That's two edits. The third edit was undoing the bad change they accepted. The undo was also contentious, with 2 votes to reject and 3 to accept.

But, Akshat has only 18 points. Even if you're lucky enough to have worthless suggested edits accepted, it's not a very good way to earn points.

Reply on news.ycombinator.com