harper

this was originally how i hosted simple sites on app engine. It works wonders.

I certainly don't think your post was for naught. The flask app is a great way to do it.

If i didn't already use git in my workflow, i would prefer this approach.

I use my www as the origin for the static media. I assume that cloudfront files stay in the CDN until they are invalidated. I haven't tested how long they stay. I imagine that if there are no requests, they go away after awhile.

You do have to pay for cloudfront usage. I am not sure the breakdown of custom origin vs s3 is - but i imagine it is similar.

Here yea go: https://github.com/harperreed/drydrop

The 30th works for me. would love to meet up.

This is great. I can't wait for the other vendors to follow suit.

This should benefit anyone who is looking to use the awesome everyblock data in their local news/data/etc apps.

That was a fun hack. When i originally launched it - it proxied all the updates through the users accounts. ;) needless the users were like "wtf." heh. I removed that feature

I built something similar to this awhile back: http://www.awesomeupdater.com/

I think telling people they are awesome is a great idea. I agree that the website needs work. I like the charging for real people to do it.

threadless was founded in 2000ish. they didn't take money until around 2007ish.

37signals is about the same (time period and what not)

i love those dudes. ;)

Just wanted to remind people to come tonight.

sounds good. don't you owe me an email?

no problem. ;)

anytime ;)

There are a bunch of people talking about this. The most promising stuff seems to be coming from Sandbox industries (i have spent a lot of time with them - so i may be biased).

There are also a bundle of VCs that are talking about this as well. If you want email/jabber me at harper@nata2.org and i will introduce you to peoples.

he is also a really nice guy.

I had just finished reading "This is not a game" about crowdsourcing crime solving using ARG techniques. It was a decent read. When i heard that something terrible had happened to the threadless tech dept. It seemed like a great way to possibly find the alleged crooks.

Luckily the dudes who remain at threadless were quick on the execution and had already appealed to the community. They are attempting to use the same method that MIT used to find the red balloons (i.e. a tiered reward for reporting info).

Here is some info from threadless: http://www.threadless.com/news/536659/Identify_these_Christm...

Here is photos from scottv's flickr stream: http://www.flickr.com/photos/scottvdp/sets/72157622921219371...

whoops ;)

This guy reverse engineered the google tasks ajax calls and created a perl module interface for it:

http://github.com/nickspacek/Net-Google-Tasks/

Works decently. Could easily be replicated in any language.

I will be there.

It is certainly an amazing situation. it took a long time and a lot of work. but when i was there it was clarity. i felt like i could see forever. ;)

I think it was a bit of coincidence and a bit how we grew up doing threadless and needed to graduate to move on. it was hilariously timed though.

i am a bit late to the game here. but i have used FDC for a year or two and have been very happy. Have one dedicated unmetered 100mbit box. it owns.

They also have GREAT support.

Cumulate Labs (cumulatelabs.com) was probably the first vector graphics app. It was purchased by autodesk and is now project draw (http://draw.labs.autodesk.com).