Checking backups

We sent an email to some of our Kickstarter backers yesterday, giving people a chance to check out Micro.blog again if they haven’t been active recently. The platform has improved so much in the last year. It’s possible that it didn’t meet someone’s needs a year ago but will today.

One of the links in the email was to the 12 days of microblogging help page, which links to 12 blog posts I wrote in December to highlight how people are using Micro.blog. Unfortunately, most of those links no longer worked! I scrambled to figure out why some of the links on my blog weren’t working, finally discovering that I had broken them myself when testing the Medium.com import feature across multiple test blogs.

In the process, I decided to pull a complete backup of Micro.blog’s database from 30 days ago and use it to fix some of my old blog posts. This was a great exercise. I check the backups from time to time, but it’s rare that I need to go over specific data by hand and actually restore anything.

Last week, Myspace revealed that they had lost all hosted music from 2003 - 2015. I don’t think we can be too careful with the responsibility of hosting people’s blogs, so I’m glad I had this chance for a real-world test of Micro.blog’s backups. (I also recommend enabling the Internet Archive feature under Posts → Design, and occasionally downloading a complete archive of your blog from Posts → Export.)

Bruce Steinberg

Any chance automating backups is on the road map?

Manton Reece

@Bruce I'm hoping to bring back the automatic GitHub mirroring in some form. But downloading an archive to your own computer really needs to be automated outside of Micro.blog.

Bruce Steinberg

Am I right that’s something I could do with the JSON API? And would it be possible to do the same with my favorites/bookmarks?

Manton Reece

@Bruce No, it's not really set up to do from an official API yet, because the download link is generated and emailed to you (and then updated on the web). Would require a little bit of web crawling to automate currently.

Bruce Steinberg

“GET /posts/all” wouldn’t do it?

Manton Reece

@Bruce That will return what you see in the timeline, but not enough for a complete archive of your site (for example, not the text of long posts).

David Reed

@vboykis: This is the most terrifying page I’ve read all week. From Creativity, Inc. by Pixar’s Ed Catmull. twitter.com/vboykis/s...

Manton Reece

@davereed That's an incredible story!

Jef Holbrook

Good update. I like what you’re doing with micro! Learned about it from Core Int.

Manton Reece

@jef Thanks!

Jef Holbrook

micro.blog reminds me of how fun Twitter was in 2008 (before things got mean over there). Thanks for making this service.

Manton Reece

@jef Thank you!

Manton Reece @manton
Lightbox Image