Profile cover photo
Profile photo
Aegis Lin

Aegis's interests
View all
Aegis's posts

Post has attachment

Post has attachment

Post has attachment

Post has shared content
Agents, we want to thank you all for your continued enthusiasm and support as Ingress turns 3 years old.  As a token of our gratitude, all Agents that reach Level 3 by midnight Pacific Time (PT) on Tuesday, December 1st, 2015 will earn a special Vanguard Medal.  The Vanguard Medal will be awarded based on the following tiers:

Bronze Medal: Reach Level 3 and above
Silver Medal: Reach Level 9 and above
Gold Medal: Reach Level 13 and above
Platinum Medal: Reach Level 15 and above
Onyx Medal: Reach Level 16

Additionally, for the next two weeks (from November 17th to December 1st), each Portal action will generate double Action Points (AP).  For example, capture a Portal to earn 1,000 AP (instead of 500 AP). 

Fluctuations in the Portal Network have caused the Hack output to be significantly boosted across the board as well.  

Share your favorite Ingress memories using the hashtag #IngressYearThree .  It’s time to Move, Agents.


Post has attachment
Google awesome

Post has attachment

そして以下のコードをclang 3.2 (x86)以降でコンパイルしてみて。恐ろしいことが起きるぜ。(まあreturnがないのはスペック的にnon-definedなので、死んだとしても文句を誰にも言えないけど)


int foo()
  for(int i = 0; i <1; i++){ //non-sense but should not hurt, right
  //how could missing the return statement become a big deal?

Post has attachment

Post has attachment
This is awesome. The display of "the raw results of handwriting" is beautiful, esp. for Traditional Chinese. I do hope the same system can be utilized to release a hand-writing Notebook app. 

Post has shared content
Ever wonder about the habits of kernel developers? Who is a night-owl, and who has a 9-5 job? Who has kids?

Never fear, you can just ask 'git' (let's see how badly G+ screws up the formatting here - there seems to be no "block quote" formatting thing):

    git log --merges --committer=Torvalds --pretty="%cd" v3.19.. |
        cut -d' ' -f4 | cut -d: -f1 |
        sort -n |
        uniq -c

and there it is.

Now, look out a bit: my merge history says that I'm more active in the mornings, then take a breather around noon (recently that's my swimming) and then come back in the afternoon.

But if you skip the "--merges", it looks very different, and it looks like I do most of my commits in the afternoon, then take a break for dinner, and come back after 9pm.

Why? My non-merge commits are hugely skewed by the patch-bombs from Andrew, which seem to happen in the afternoon. While a fair amount of my merge activity in the morning is because of all the pull requests that came in overnight from other parts of the globe.

Also note that the best statistics really depend on committer times, since author dates are often skewed by who forwards the emails and where in that sequence people added a date to it. So I wouldn't trust author dates to show the authorship dates as much - they might instead be showing the email patterns of people in the chain.

I'm sure you could make something cute with gnuplot and show different patterns of the top developers.

And if you're a developer and don't want the world to see how you seldom get up before noon, you might want to be aware of just how much these things can tell people about your work habits.. Although among sw people, I suspect the "not up before noon" is a badge of honor (as well as a indication that you don't have kids ;)

[ Edited to replace the sed-script with the 'cut' lines that G+ formatting won't eat ]

Post has attachment
Wait while more posts are being loaded