nnythm

Engineer on finagle, used to work at tumblr when we started running into problems with finagle. In large part it was because we didn't understand how it worked, and were doing things like reinventing different pieces of finagle, like load balancing and service discovery.

I think saying colossus is "significantly better performing" is a bit of a stretch, although I haven't seen it in the past year, I think the main win was that you called epoll faster than netty, which I think was because you were using a faster timer. What are the performance comparisons you've done recently?

MVCC has been around for at least thirty years, but it's interesting that we have seen more databases with this feature recently.

This is good advice. Another good tool to better yourself is reading code. This lets you think about why people have written code the way they have, and whether you agree with how they did it.

I have also had this happen to me--I am part of a fairly large company on github, so this means that suddenly I don't see anything in my strema except for posts about that company, so I had to unwatch nearly everything in it, which I would rather not have to do.

I feel like a large percentage of hacker news is posts exactly like this.

"Want to be productive? Stop reading about how to be productive!"

I personally feel like Hacker News would be better if these posts stopped being bubbled up to the top.

It seems like you get most of this stuff from a basic abstract algebra course. How does category theory help me beyond what I got from modern algebra 101?

edit: this is from the perspective of someone who knows a little abstract algebra, and wonders if it would be sufficient to brush up on my algebra to get these ideas, which seem a little familiar, or if category theory is really important to get these ideas. I'm not trying to be a contrarian.

this is the correct pronunciation.

I had a similar experience, where I found that none of my computers, neither my six year old laptop, nor my one year old laptop (!) could support GNOME 3, which has essentially prevented me from upgrading my fedora from 14. Since Fedora is currently on version 17, this makes me sad, and ruminating about which OS to jump ship to.

I was skimming this article and thought that the premise was that first person shooters were switching to using the up, left, right, and down arrows instead of wasd. I felt the lede + nut graf (was there even a nut graf here?) was misleading--it was completely about the buttons, not about the identity of the gamer, the game industry, nor society in general.

I am interested in articles about female gamers and how they deal with the unpleasantness surrounding the gaming industry, but I wish that blog posts were held to a slightly higher standard--I am a fairly fast reader, and had to double back to reread large parts of the article to recalibrate as the article started talking about being directed to games with fuzzy animals and facebook games instead of shooters, when I expected the author to be complaining about the wasd keys opening different menus rather than moving around the character.

This seems to be a pretty good argument for paying for better quality journalism. I bought a copy of GQ, not exactly the highest brow magazine in the business, admittedly, and the articles weren't riveting, but it was efficient to read them, because they were written carefully and in the traditional journalistic style, with a lede, nut graf, and transitions. Furthermore, it's a magazine FOR MEN so most of the words were Anglo-Saxon and avoid the errors that Orwell rails against in "Politics and the English Language." I never felt like I was being jerked from one idea to another, and I understood the point of the article if not from the titled or the images associated with it, then at least from the first few paragraphs.

TL;DR: May there be truth in advertising, if bloggers can't write a proper lede, we will have to pay for news.

Using a language with static typing means that there are a lot of tests you don't have to write. Do you remember the units trick from physics? If you forgot which equation you're supposed to use, keep on using identities until you're in the units you're supposed to be. Static typing checks that your units are correct for free.

I agree, writing Java without powerful refactoring tools is a bore--that's why it's great to use Eclipse's "Generate Getters and Setters based on Fields" refactoring tool.

Interestingly, I've never really felt the pain of not having a good REPL for java, although I often use it in other languages. Although Java isn't interpreted, unlike the other languages you mentioned, Scala is both interpreted and compiled, so it's not a great argument for why it doesn't have a good REPL. I think eclipse's very strong autocomplete somewhat obviates the need for it, because I know that I mostly depend upon bpython/ipython for the easy access to docstrings and information about arguments.

I wonder if part of the reason why I've never felt the need for a REPL in Java is simply because there isn't one built in, so it isn't a tool that I ever reach for. REPLs are great for prototyping, and for learning, but I've found that the more I learn about a language, the less I tend toward using the REPL, so perhaps it's simply that I got used to being able to edit my code instead of having to type it all in again.

Interesting. My experience with Java is that the only thing that gets in the way of programming in Java is the verbosity, which an IDE with powerful refactoring tools (eclipse, IDEA, eclim) gets around nicely. I would argue that in many ways, Java is easier to program in than a lot of other currently hot languages because there is so much tooling around it (although I would trade basically any python package manager for ivy any day), it is statically typed, and its behavior is generally boring but extremely predictable, and not hard to reason about, excluding the threading library which is not super fun. What have you found is frustrating about using Java?