Disqus - Latest Comments for onecreativenerd

Re: How to Stop your Glassware from Getting Stolen

Ryan Witt — Tue, 15 Dec 2015 17:15:04 -0000

Offer the glass as a souvenir for $10? I suspect many people just want a keepsake. :)

Re: Django Login Required Middleware

Ryan Witt — Mon, 27 Jan 2014 01:06:48 -0000

Sure!

Re: CS For Hackers

Ryan Witt — Wed, 05 Sep 2012 18:24:26 -0000

data server: https://gist.github.com/364...

Re: CS For Hackers

Ryan Witt — Tue, 04 Sep 2012 21:01:35 -0000

6 https://gist.github.com/362...

Re: CS For Hackers

Ryan Witt — Tue, 04 Sep 2012 21:01:06 -0000

https://gist.github.com/360...

Re: CS For Hackers - Week 5 - Data Structures and Algorithms (part 3)

Ryan Witt — Fri, 03 Aug 2012 02:11:42 -0000

I found the following file format helpful for the graphs:

# src,dst,weight
0,1,10
0,4,100
0,3,30
1,2,50
2,4,10
3,2,20
3,4,60

You have to decide how to interpret though, probably need a directed/undirected loader. Or, you could just define the graphs in code.

Re: Django Login Required Middleware

Ryan Witt — Wed, 21 Mar 2012 15:14:08 -0000

Thanks! I have not updated this in a while, but I'm using `next` internally. :)
Do you have a fork of the gist I can pull?

Re: The Facebook Freaky Line

Ryan Witt — Mon, 21 Nov 2011 04:15:04 -0000

How would the "freaky line" move for you if Facebook took a more nuanced view of your behavior?

For example, a chartbeat-like reading/writing metric could lead to something like, "Robert Scoble is *closely* reading Ex-MySpace CEO resigns as Zynga executive...", or "Mark Zuckerberg is commenting on scobleizer.com".This would filter out the one-offs and provide a much more valuable signal to your friends. Of course, this value could be captured.

Does this move the line for you?

Re: How does liquidity preference affect common stock value? (Ryan) - Opani

Ryan Witt — Fri, 14 Oct 2011 02:18:54 -0000

So, would you take an average exit price, or perhaps calculate several like I did with exit preference? I also thought of tying the sliders together so they followed the data, another way to explore the relationships.

Re: What are good company tags? (Dirk) - Opani

Ryan Witt — Mon, 26 Sep 2011 19:55:08 -0000

What makes these tags "good"?

Re: http://onecreativeblog.com/post/8785145888

Ryan Witt — Sun, 25 Sep 2011 19:41:27 -0000

Haha! Well, I seem to remember spending more hours bent over the workbench soldering than you did. ;)

Re: Sarah Palin Emails

Ryan Witt — Sun, 12 Jun 2011 21:59:01 -0000

@Tagxedo, thanks! I got the table formatting wrong. Fixed now.

Re: Keys

Ryan Witt — Wed, 08 Jun 2011 19:51:45 -0000

Zach, just added some documentation for it. Thanks for pointing it out!

Re: Who are investors in early-stage data companies? (Dirk) - Opani

Ryan Witt — Wed, 08 Jun 2011 10:33:49 -0000

Check this out: http://techcrunch.com/2011/...

Re: Opani Help

Ryan Witt — Tue, 07 Jun 2011 21:10:16 -0000

like the new organization of this page!

Re: Social Graphs

Ryan Witt — Mon, 30 May 2011 17:27:35 -0000

I dunno, the twitter example doesn't really show anything about a social graph. That's more the real time web.

Re: Where is the face? (Dirk) - Opani

Ryan Witt — Mon, 30 May 2011 13:10:52 -0000

cool, works for me now. Can you replace credential with get_credential?

Re: Where is the face? (Dirk) - Opani

Ryan Witt — Tue, 24 May 2011 13:45:41 -0000

I'm getting this error:

Traceback (most recent call last):
File "run.py", line 102, in <module>
download_images(urls)
File "run.py", line 31, in download_images
eyeCascade = cv.Load('../scripts/haarcascade_eye.xml')
TypeError: OpenCV returned NULL

Re: Opani: Social Supercomputing in the Cloud

Ryan Witt — Sat, 07 May 2011 15:34:15 -0000

@znmeb you're right about the twitter project being the default access.

Opani doesn't store data, only results (images, output logs, etc.). Our current users mostly store data on private file servers or Amazon S3, although many example apps pull from public APIs.
We also support entering credentials for S3 and other APIs so you can share apps between users without embedding the credentials.

Re: The rise (and fall?) of firms in the ‘government’ ‘data’ space

Ryan Witt — Wed, 06 Apr 2011 14:51:06 -0000

Someone should supplement the crunchbase data with SEC filings... hmn...

Re: Django Login Required Middleware

Ryan Witt — Tue, 22 Feb 2011 10:19:53 -0000

you need to use consistent indentation with python. you might have copy-pasted tabs when you had spaces in the original file.

Re: Django Login Required Middleware

Ryan Witt — Tue, 22 Feb 2011 08:53:13 -0000

You could try wrapping the login view with something like:


# views.py
from django.contrib.auth.views import login
from django.shortcuts import redirect
def new_login(request, *args, **kwargs):
	login(request, *args, **kwargs)
	return redirect('/admin/')

# urls.py
urlpatterns += patterns('',
	url(r'^/login/', 'views.new_login', name='login),
)

Re: Which loads faster?

Ryan Witt — Mon, 13 Dec 2010 16:32:10 -0000

@dimok,

Thanks :)

Which version of chrome are you running? It's working for me in OSX Chrome 8.0.552.215 - 8.0.552.224

Can you describe what it looks like when it doesn't work? Do you have any chrome extensions installed? It would be good to know so I can debug it. :)

Re: Which loads faster?

Ryan Witt — Sat, 31 Jul 2010 17:34:03 -0000

You're welcome! Hopefully you will like the next version. :)

Re: Which loads faster?

Ryan Witt — Thu, 29 Jul 2010 12:06:02 -0000

Thanks, Vincent. I'm definitely with you about testing from different locations. One of the next features is an API that should allow you to collect results, and that could be easily augmented by some basic browser scripting in remote locations.

There are actually several good open-source systems for remote performance monitoring, http://webpagetest.org and http://showslow.com come to mind. These tools both provide more in-depth analysis than whichloadsfaster, so I would definitely take a look at those!

I think it would be possible to provide a better UX when sites try to framebust, but I'll have to find some time to focus on it. I was hoping somebody with expertise would comment, but perhaps the nature of the arms rase between framers and framees makes this unlikely. :)