Erik's blog

Code, notes, recipes, general musings

Archive for the ‘notes’ Category

Notes from Kyle Neath’s presentation at Twitter on 5/31

with 3 comments

  • Slides
  • hashbang urls
    • are a kludgy workaround for lack of history api. Since history api is coming, they have no future. Since urls are forever, especially w/ tweets being stored in the lib of congress, use of hashbangs results in permanent support for a temporary condition.
    • break pre-existing url fragment behavior
    • result in confusing routing logic
  • “responsive web design” is adapting to client and seeming responsive to user input
    • page load isn’t just a benchmark; a page is only “loaded” when the user can scroll, read text, and click links
  • well-designed urls provide a command-line-like interface for web apps
  • all web assets should have a url, i.e., navigation should not allow access to a resource that cannot then be accessed directly via a url
  • native elements should behave as the user expects
    • do not modify common key combos, e.g., shift + click
    • take advantage of the back button, tabs, links, etc
  • responsiveness is as much about performance as perception
    • wait ~500ms before showing loader image; showing loaders immediately can actually make the page seem slower
  • ssl
    • is required now that there are common, easy ways to sniff credentials
    • a new ssl handshake is very slow, and required for each domain
    • use http keep-alive to reuse ssl connections
    • multiple parallel requests to a new domain will each have to perform a handshake; instead, complete one fast request, and then reuse the connection for subsequent parallel requests
    • github optimized its backend to 40ms latency before realizing that the ssl handshake takes 500ms
      • a case of perception > performance
      • favor science over theory, i.e., test time-to-usable in multiple regions instead of just running perf tests on components
    • templates
    • use something simple, e.g., mustache
    • avoid rendering on client and server; pick one
    • kneath prefers server-side
    • for server-side rendering, passing html back as one value in a json object allows for passing data back in other keys
  • html 5 history api
  • allows for much richer state management. See github’s new issues dashboard

Written by Erik

May 31, 2011 at 8:19 pm

Posted in notes

Tagged with , , , ,

Notes from Neil Gershenfeld’s 5/24 talk at Twitter

leave a comment »

My mind was just blown by a talk from Neil Gershenfeld, director of the Bits and Atoms lab at MIT. His team created the fab lab. Here are  some notes

Written by Erik

May 24, 2011 at 12:45 pm

Posted in notes

Tagged with , ,

Notes from Rob Pike’s 5/12 talk on Go at Twitter

leave a comment »

  • See for language documentation
  • W/in a factor of 2 slower than C/C++. Generally w/in 20% speed of c/c++ programs
  • Intrinsically safe
  • This talk has been presented before, so the slides may be online. see google io 2010 archive
  • check out article in Register about Go that quotes Odersky: “I like a lot of the design decisions they made in the language … Basically, I like all of them.”
  • built on 4 self-reinforcing principles: simple, ortho, succinct, safe
  • see axiom of choice in type theory
  • public/private hint in variable name is one of the best things about the language
  • see CSP tradition
  • uses a deterministic model, channels, for concurrency
  • the “go” keyword launches a go routine
  • “for { … ” declares an infinite loop
  • expressiveness comes from orthogonal composition of constructs
  • Go conceived while waiting for 45 min gcc compilation
  • Go app engine sdk is a complete installation vs building from source


How is the language intrinsically safe?

  • no stack overflows


  • native swig support for C/C++ progs
  • no java interop


  • no try/catch
  • uses panic/recover
  • function-level, not statement-level

Channel implementation?

  • not like erlang channels
  • passing channel over netchan is coming


  • Core team has members that believe generics must and must-not be included


  • gofix rewrites code
  • gofont pretty-prints/formats code


  • Oberon
  • New Squeek
  • Didn’t cherry-pick features to build an ideal language, but they did include elements that helped them be productive

Written by Erik

May 12, 2011 at 11:24 am

Posted in notes

Tagged with , ,

Getting started with unit testing for Node.js

with 3 comments

I’m diving into unit testing with Node.js, and my first stop is nodeunit. Luckily, Caolan McMahon wrote an excellent introduction to nodeunit on his blog. Thanks, Caolan.

I installed nodeunit via npm no problem: npm install nodeunit

All the examples in the Installing nodeunit section worked fine, but I needed to add
var events = require('events');
to the first code sample in the Testing asynchronous code section to get those tests to pass. So, the top of my test-doubled.js file looks like:

var doubled = require('../lib/doubled');
var events = require('events');

Farther down in the blog post, in the Shared state and sequential testing section, there’s a code sample with the events include in it, so I think I’m on the right track.

In the Test cases, setUp and tearDown section, I had trouble getting the tests to run. After referencing the project’s README file, I tried adding a callback arg to setUp() and tearDown(), and calling the callback, which worked. So, my code looks like:


var testCase = require('nodeunit').testCase; = testCase({
setUp: function (callback) {
this._openStdin = process.openStdin;
this._log = console.log;
this._calculate = doubled.calculate;
this._exit = process.exit;

var ev = this.ev = new events.EventEmitter();
process.openStdin = function () { return ev; };

tearDown: function (callback) {
// reset all the overidden functions:
process.openStdin = this._openStdin;
process.exit = this._exit;
doubled.calculate = this._calculate;
console.log = this._log;


With the minor tweaks above, I was able to get all tests to pass:
Screenshot of all tests passing


Written by Erik

December 6, 2010 at 11:11 pm

Posted in notes

Tagged with , ,

Notes from Ryan Dahl’s talk “On Node.js” at Cloudstock

leave a comment »

Here are my notes from Ryan Dahl’s talk “On Node.js” at Salesforce’s Cloudstock event on 12/6:

  • Google put a lot of thought into v8 performance
  • Node is a set of bindings to v8 to allow js to do non-browser things
  • “I/O needs to be done differently”
  • There’s a big difference between dealing w/ stuff from the cache vs from a network.
  • nginx is just 3x better than apache in terms of # concurrent clients x # req/sec, but nginx’s mem usage is nearly constant vs apache’s steep curve towards 40 mb
  • apache uses a thread for each request, whereas nginx uses a single thread w/ an event loop
  • it’s well known that you need to use an event loop f you want to go crazy w/ concurrency
  • but even w/ an event loop, you pay dearly for blocking processes
  • we should be writing all of our i/o using non-blocking calls
  • sleep() is a blocking operation
  • Why “transfer encoding: chunked” by default? because we don’t know the full size of the response, and we can start returning immediately
  • each connection costs 1-2 kb minimum
  • random, humorous paraphrase: … !==, not !=, i hate javascript. Coding on stage is so difficult …
  • questions
    • when is node going to become stable? we have a stable branch (0.2), and 0.3 branch will break backwards compatability. I want this to be awk. I want it to be a unix util
    • how’s the hosting landscape looking? joyent is working on a special service for node, which is in its beta. heroku has some stuff. you can always use a general vps w/ an ops layer like monit.
    • interesting applications? node is good for realtime, massive concurrency sorts of things. is a bot on an irc room which geolocates chats btwn people in a room.
    • what do you think about express? express is a web framework for node. it looks cool
    • what do you think about node being tied to v8? untying it doesn’t make sense at this point, but I’m appy w/ v8.

Written by Erik

December 6, 2010 at 12:32 pm

Posted in notes

Tagged with , , , ,

getting started with Node.js

with 5 comments

I found what appears to be a nice tutorial for installing Node on Ubuntu 10.4, so I’ll start with that.

To keep things simple, I’m going to skip the dependencies, grab the tgz file, and just try running it. Why not? Thing’s could’ve changed w/ Node and/or Ubuntu since that post was written, and I love it when packages are designed well enough to provide irrational users with informative feedback, so let’s see.

  1. wget
  2. tar -xf node-v0.2.3.tar.gz
  3. cd node
  4. ./configure
  5. make
  6. sudo make install
/home/erik/node-v0.2.3/wscript:132: error: could not configure a cxx compiler!

Nice! I need g++:
sudo apt-get install g++

/home/erik/node-v0.2.3/wscript:188: error: Could not autodetect OpenSSL support. Make sure OpenSSL development packages are installed. Use configure --without-ssl to disable this message.

Looks like I need ssl too:
sudo apt-get install libssl-dev

Configuration now passes, but I see this in the output:

Checking for openssl                     : not found
Checking for function SSL_library_init   : yes
Checking for header openssl/crypto.h     : yes
Checking for library rt                  : yes
--- libeio ---
Checking for library pthread             : yes
Checking for function pthread_create     : yes
Checking for function pthread_atfork     : yes
Checking for futimes(2)                  : yes
Checking for readahead(2)                : yes
Checking for fdatasync(2)                : yes
Checking for pread(2) and pwrite(2)      : yes
Checking for sendfile(2)                 : yes
Checking for sync_file_range(2)          : yes
--- libev ---
Checking for header sys/inotify.h        : yes
Checking for function inotify_init       : yes
Checking for header sys/epoll.h          : yes
Checking for function epoll_ctl          : yes
Checking for header port.h               : not found
Checking for header poll.h               : yes
Checking for function poll               : yes
Checking for header sys/event.h          : not found
Checking for header sys/queue.h          : yes
Checking for function kqueue             : not found

Rather than run into mysterious errors later, I’ll go ahead and install all the dependencies mentioned in the post:
sudo apt-get install g++ curl libssl-dev apache2-utils

But it doesn’t correct the problem. Doh! Oh, well. I’ll deal with any errors later.

Continuing on, I run make (it’s been a while since I watched C compile), and then sudo make install:

'install' finished successfully (0.239s)


Rather than starting with the tutorial’s translation example, I opt for the hello world example on the Node site:

var http = require('http');
http.createServer(function (req, res) {
  res.writeHead(200, {'Content-Type': 'text/plain'});
  res.end('Hello World\n');
}).listen(8124, "");
console.log('Server running at');

I tried loading this, but no response. However, I grappled with my vm’s firewall recently enough to remember that I hadn’t opened port 8124, so I update the code to use port 80, and re-launch, but still get no response. Undaunted, I tell ufw to take a break:
sudo ufw disable && sudo shutdown -r now

Upon trying again, I noticed that I had the address wrong earlier – pilot error – but before enabling ufw, let’s see if we can get a response.

Curling returns “hello world”, but I’m not able to see this server from my laptop, i.e., curl doesn’t work.

Update (11/22/10): check out Padraig’s comment below about binding to

It’s getting late, so this is a race against mental mutiny. Desparate, I stumble across another node.js + Ubuntu tutorial, and blindly install everything. I launch as per the instructions, bounce over to the browser, and … it works. Awesome. To the author of that post, nice job.

Ok. That’s enough for tonight. I’ll revisit this again later and try to understand what just happened.

In parting: Cats! In 3D! Don’t think. Just stare.

Cats Anaglyph 3D اناگلیف

Photo credit: Shahrokh Dabiri

Written by Erik

October 21, 2010 at 12:27 am

Posted in notes

Tagged with ,

getting started with Ubuntu server, continued

with one comment

I’ve been playing around with an Ubuntu 10.10 64-bit server vm for the past couple days. My previous post ended with me unable to ssh into the vm. Today, I read a post on serverfault that got me thinking about trying to ping the vm. For whatever reason, I might have more success with that than ssh.

The post also mentioned turning off the firewall, which seems like a logical step, though I’ve been paranoid for so long, I’m uncomfortable doing so, even on a vm. Nevertheless, I turned off the firewall: sudo ufw disable

Restarted the vm: sudo shutdown -r now

Ran ifconfig to confirm that the ip hadn’t changed: ifconfig

But it had. It was now something like, which looked off. It didn’t seem like something I could ssh into. When I was hacking around the other day, I had tried configuring the vm to use a bridged network. I now switched back to the default NAT setting: Virtual Machine > settings > Network > Share the Mac’s network connection

I restarted the machine, and ran ifconfig again. Now it was back to an ip that looked more familiar:
… inet addr: …

From my Mac, I tried pinging the vm: ping

To my surprise, it responded. Amazing. I tried ssh’ing in and was rejected with
ssh: connect to host port 22: Connection refused.

Then I remembered I had set the ssh port to 2222, and tried again:
ssh erik@ -p 2222

Success! Amazing again. Then I went overboard and turned the firewall back on, sudo ufw enable, and restarted, but it still worked(!). I checked the status of the firewall just to make sure I wasn’t delusional: sudo ufw status

Status: active

To                         Action      From
--                         ------      ----
2222                       ALLOW       Anywhere
22                          ALLOW        Anywhere
80/tcp                     ALLOW       Anywhere

This is madness, but it’s working, so I’m not complaining.

To wrap up the ssh piece, I want to configure ssh to use a public key instead of a password. I’ll follow the SSH/OpenSSH keys tutorial on the Ubuntu wiki.

I created a new ssh key, ssh-keygen -t rsa, and save it in a file called ~/.ssh/erik_rsa.

Then I copied the key to the vm:
scp -P 2222 ~/.ssh/ erik@

Ideally (for me), the vm would now just automagically prefer public key authentication. I gave it a shot and immediately tried ssh’ing into the vm, but it still prompted me for a password. Oh well.

I edited the sshd_config file to uncomment the line AuthorizedKeysFile %h/.ssh/authorized_keys and set PasswordAuthentication to “no”, and then restarted sshd: sudo /etc/init.d/ssh restart

I exited the vm, and tried ssh’ing in again, this time specifying the location of my shiny new ssh key: ssh erik@ -p 2222 -i ~/.ssh/erik_rsa

Wow. It worked. I’m always astounded when things like this actually work. Mac prompted me to enter my passphrase, and then I was in.

Now that I can ssh and I have a firewall, it’s time to move on to the next step of the security wiki … hmm … well, before I dive into denyhosts and fail2ban, I think I’ll play around with the real reason I wanted to get an OS up and running, to install node.js.

Written by Erik

October 20, 2010 at 10:32 pm

Posted in notes

Tagged with , ,