Explore automated deploys from master

[tofumatt]

IT shouldn't need to push new versions as they do; maybe all it takes is to automatically git pull from time-to-time?

[fwenzel]

They do that for dev / stage instances, certainly.

[tofumatt]

We actually don't have one for this site! ^_^

• tofumatt (Sent from mobile)

On 8 May 2015 02:15:14 BST, Fred Wenzel notifications@github.com wrote:They do that for dev / stage instances, certainly. —Reply to this email directly or view it on GitHub.

[pmclanahan]

Since this is a generated static site we should be able to have Travis CI build it and push the site to S3 and host behind CloudFront. We're doing that very thing right now for Lorax (shapeoftheweb.m.o).

[tofumatt]

Any docs as reference? I don’t mind hooking all that up if there’s an example.

[pmclanahan]

No docs for that project so far outside of the Travis docs for deploying to s3. The trick will be regeneration on a schedule like it's doing now. I think @Osmose did something for that for the http://mozilla.github.io/webdev/ site.

[pmclanahan]

But like you said in the issue, likely this would be easy for IT to setup. Just pull from git every few hours or even daily and it'll update the next time the scraper runs.

Since everything is fleeing the datacenters though, perhaps it would be good to do something like s3 sooner than later just to make this be one less thing to worry about for the mass migration. Though I've also heard rumblings of integrating this site into bedrock.

[tofumatt]

Well, the bedrock integration has been a rumour for years, so I wouldn’t build around that ^_^

[pmclanahan]

Fair enough :)