Disqus - Latest Comments for collintmiller

Re: Del Close Expects That Every Player Will Do Their Duty.

collintmiller — Tue, 15 Mar 2016 19:05:16 -0000

Teaching explains conceptual systems and guides of learners to understand and use those concepts. This requires correction when students fail to recreate the concepts.

I agree hole-heartedly: Have a specific metric that you're teaching against. And then hold them accountable for failure to adhere to the metric.

Folk-improv-philosophy supposes that there is no wrong, and you can do anything you want. This is important and true, but it does not override the role of a teacher.

At it's worst, the folk-improv-philosophy causes teachers to encourage or ignore failures. A student might student fail to carry out a specific exercise. If the teacher responds "Well that's okay, not every scene needs [ purpose of the exercise ]," there has been a breakdown of the teaching role.

I've been in classes where that has happened. I've even said that unfortunate phrase in classes I was teaching.

The students are not dumb. They know they aren't landing the exercise. But it's a well-intentioned lie to encourage the student. In the student/teacher status what is says is, "I don't feel comfortable fulfilling my role by correcting you." That creates ambiguity the student can internalize to mean, "I'm so bad at this the teacher doesn't even want to help me improve."

Re: Static Showdown: Hackathon for Static Web Apps

collintmiller — Thu, 06 Feb 2014 17:16:15 -0000

aw snap!

Re: Static Showdown: Hackathon for Static Web Apps

collintmiller — Thu, 06 Feb 2014 13:54:10 -0000

Hello, this is a comment! Great innit?

Re: I do not think that word means what you think it means

collintmiller — Mon, 15 Apr 2013 04:40:40 -0000

Thank you sir.

Re: Blog of Ryan Bigg - Engines and Authentication

collintmiller — Sat, 03 Mar 2012 23:54:39 -0000

I worked on a project that used to different plugins that both used Devise in incompatible ways. Yuck.

Re: Being recruited in the USA - Pau Ramon

collintmiller — Wed, 04 Jan 2012 13:16:36 -0000

I wil gladly hire remote when I get things off the ground.

Re: Backbone.js and Capsule and Thoonk, oh my! A scalable realtime architecture? | &yet | the blog

collintmiller — Wed, 16 Nov 2011 20:56:59 -0000

Do you do anything for concurrency and conflict resolution?

I'm working on an Operational Transformation solution that leverages Backbone in a similar way.

Re: http://localhost/orchard/does-this-look-good

collintmiller — Fri, 30 Sep 2011 14:54:03 -0000

Very strange indeed.

Re: Rails 3.1 Asset pipeline rocks!

collintmiller — Mon, 29 Aug 2011 11:55:04 -0000

Do you have any examples of the sort of stuff you're doing in your models you could show off?

Re: Rails 3.1 Asset pipeline rocks!

collintmiller — Mon, 29 Aug 2011 11:52:58 -0000

" generate backbone.js classes with all boiler plate stuff filled in "

Oh snap! This is an awesome tip, I didn't make this connection, or even really thing to use ERB in my coffee.

Re: http://improvnonsense.tumblr.com/post/6097831033

collintmiller — Thu, 02 Jun 2011 21:28:47 -0000

I've also thought that #1 "yes" responses work particularly well. As an example I was in a scene that was initiated something like, "You only want me for my body don't you?"

And so I responded something like "That's the only reason anyone would want anybody."

And this feels very much like it fits the theme of your #1 situation. I feel like there is something common to all these lines.

“Wait, are you they guy who’s trying to fake that your grandfather is a citizen?”
“Are you saying you’re going to quit your job?”
“There’s something else going on. Are you trying to tell me something?”
“you’re angry, aren’t you?”
“is that MY briefcase?”

I think the common thread is what makes them so easy to brush off is that these are all in some way an accusation. And we're quite used to ducking accusations both little and small.

What I see going through my mind when I see that yes is, "Oh wait a second, s/he was just accused of something. That should be embarrassing, the accuser should have the upper hand now. But the accused just flipped my expectation of what happens in an accusation, how unusual and delightful. I don't know what will happen but I know exactly how we got here."

Re: HandlerSocket: The NoSQL MySQL & Ruby

collintmiller — Fri, 14 Jan 2011 17:23:50 -0000

I like the HandlerSocket approach. And to be entirely honest, I'm never writing SQL anyway. It'd be awesome to have handler socket integrated directly into Arel and other ORM query engines.

But I have my doubts that replacing a big distributed fault-tolerant memcached-like tier with InnoDB is going to be a great pain.

A good cache ought to be a bit more workload specific than that.

Re: Don't Romanticize Dropping Out - Edward Tenner - Politics - The Atlantic

collintmiller — Fri, 17 Dec 2010 12:56:02 -0000

Thiel waves an unlikely carrot, but knocking that down doesn't end the discussion on dropping out.

There is a college monoculture in the US. From a young age young people are told that college is the way.

That fulfilling their potential is predicated on picking an institution.

It is emotionally difficult for those who do not fit well in that mould. It may be few who will excell best outside these institutions, but minority status has historically been a poor reason to ignore the freedom of a group of people.

And it is a struggle. College certainly is easier if you choose a sensible major, focus and stuck with it. Once you graduate you are in the club, and then the fight begins. And it has become one heck of a fight in these trying times.

I'm glad I dropped out of high school and have a 5 year advantage on my peers.

As far as accomplishments and moving things forward? Some of that will happen inside an some out. Some people excel in academics. I would never take that away from them.

So please yes, let's kill the sacred cow already.

Re: Two Kinects at once!

collintmiller — Mon, 29 Nov 2010 04:32:15 -0000

How much of the distortion here do you think could be cleaned up?

I only ask because a few kinect or some similar custom rig would be ridiculously cheaper than and more flexible than a standard studio camera setup, and I would love to have that sort of thing.

Re: JPEGs with Alpha Channels?!?

collintmiller — Mon, 04 Oct 2010 23:03:26 -0000

Wicked smart. Love it

I've been missing alpha jpeg my whole life.

Re: Announcing Twilio Fund for 500 Startups

collintmiller — Fri, 24 Sep 2010 17:14:29 -0000

Great timing Dave :)

I just started hacking on a Twilio app a couple days ago focused on problems with phone-based-work scheduling.

Re: Using Bundler 0.9.4 with Rails 2.3.5

collintmiller — Sat, 13 Feb 2010 15:59:50 -0000

This doesn't work for me. Using this strategy loads the wrong versions of gems from my ~/.gem directory. It is in fact very much like not using bundler at all.

Re: The NYC tech scene is exploding

collintmiller — Mon, 01 Feb 2010 09:43:10 -0000

I'm excited by all the activity in the NYC tech scene. :)

Another thing I like about NYC as a host for a startup is the density. Density provides opportunities mere population does not.

But it can be a little overwhelming to understand how to get started as an entrepreneur anywhere; it's encouraging to see the momentum picking up here.

Re: DRY and Pythonic jQuery?

collintmiller — Tue, 01 Dec 2009 03:37:16 -0000

Hey, I've noticed, thanks to your post, that the Jabs readme file is showing those two examples incorrectly. I've updated the text at http://github.com/collin/jabs

Re: Pusher & Async With Thin - macournoyer's blog

collintmiller — Tue, 15 Sep 2009 17:43:11 -0000

I am using a modified deferrable body, the issue I'm running into is I don't want to return a standard [status, headers, deferrable_body] response.

With my proxy I want to play the "remote" response back directly over the "local" request.

My rack handler looks like this:

def call(env)
request = Rack::Request.new
# using a modified DeferrableBody so I can control the callback used.
socket = EM.connect request.host, request.port, RequestProxy, DeferrableBody(@env['async.chunk_callback'])
socket.send_data @env['http.request_body'] # plain, unparsed FULL http request
end

class TCPSocketProxy < EventMachine::Connection
def initialize socket # socket is the async chunk callback
@socket = socket
end

def recieve_data data
@socket.call data
#
EM.next_tick { close_conncetion } if finished?
end

def unbind
@socket.succeed
end

def finished?
# I have implementations of these, but need to redo them with a proper HTTPResponse parser
return false unless headers_recieved?
return true if not_modified?
return false unless content_length
return true if body_length = content_length
end
end

Re: Pusher & Async With Thin - macournoyer's blog

collintmiller — Mon, 14 Sep 2009 20:33:49 -0000

I've been doing more with this callback lately and I've found a use-case that I couldn't figure out how to do without a modification to Thin.

http://gist.github.com/187018

I'm using rack to build an http proxy. The proxy works rougly like this:

Request comes in.
Open Socket to the host and port of the Request.
Replay entire Request Body onto Socket.
As data comes back over Socket, write to Request's Async Chunk Callback.

Any thoughts about better ways to do this?

Re: Superfeedr Blog : Federating PubSubHubbub

collintmiller — Sun, 12 Jul 2009 13:36:05 -0000

http://video.google.com/vid...

Great tech talk that talks about network technologies, and ideas about a possble future where content is the center of the network, not wires or nodes.

Re: Superfeedr Blog : Federating PubSubHubbub

collintmiller — Sun, 12 Jul 2009 12:13:49 -0000

Oooh. Nice! :D

I like the parallel between the emerging content-centric network to the internet protocol and the internet protocol to traditional telephony.

New networks with their own topographies built atop the technology of yesterday.

Re: Pusher & Async With Thin - macournoyer's blog

collintmiller — Wed, 01 Jul 2009 04:34:24 -0000

@macournoyer

:) Awesome. This post and the resources linked to within have proven invaluable :)

http://github.com/collin/or...

Have it almost working with the existing Orbited client code :)

Anybody have a notion of when we can expect to have a unified asynchronous callback api in Rack? Thin is nice, but it'd be even nicer if more rack servers could be expected to run this stuff.

Re: Pusher & Async With Thin - macournoyer's blog

collintmiller — Wed, 24 Jun 2009 21:14:06 -0000

Oh excellent!
Mind if I use your DeferrableBody code in http://github.com/collin/or...