Quick Stikkit backup

Update: Any future updates to this will be happening here.

Now that I'm starting to get a lot of information into Stikkits, I wanted an easy way to back them up. What I've come up with so far is a Ruby script that utilizes the Stikkit API to backup each of my stikkits to individual text files. I end up with with a big directory of grepable text files that I can use if stikkit.com happens to be unavailable. cron is my friend and runs the script for me each night.

Here's the script, really simple and just a few lines longer than the example on the Stikkit API page (remember to put your API key in there):

#!/usr/local/bin/ruby

require 'net/http'

backup_directory = "/server/homes/alm/stikkits/"

page_number = 1
continue = true


while continue == true
    response = Net::HTTP.start('api.stikkit.com') { |http|
        req = Net::HTTP::Get.new('/stikkits.txt?api_key=xxxxxxxxxxxxxxxxxxx&page=' + page_number.to_s)
            http.request(req)
    }

    page_number += 1

    response.body.split("\n\f").each do |stikkit|

        stikkit_file = File.open(backup_directory + stikkit.split("\n")[0].gsub(/\W/, '') + ".txt","w+")
        stikkit_file << stikkit
        stikkit_file.close
    end

    continue = false if response.body == ""
end

I know that there are probably a million better ways to do this, but I just like having them all backed up as text files. It just works for me.

How I'm using Stikkit... so far

Update: Any future updates to this will be happening here.

I had started playing around with Stikkit a while back, but had sort of forgotten about it until this post at 43 Folders sort of sparked my interest again.

After a year or more of trying to get myself all gtdified, I still found myself without a really "trusted" system. Actually my system before now (and maybe still) would probably cause a lot of people to jump off a bridge because it changed on a weekly/daily basis just depending on what notepad was laying around or worse yet, what super-duper-cool website just launched promising to get your organized.

After toying with Stikkit I decided this was the one electronic system I was going to use to get organized. At least for this month. So I set out to devise a plan to try and get my system in place.

This is what I've come up with so far after a month or tweaking, broken down into different sections. Turns out it is pretty close to what I read recently here linked from Lifehacker. Actually you'd probably want to run there and read that, because it is a much better explanation. I'm not really adding too much to the discussion, but just wanted to get it documented somewhere.

Projects, Next Actions, & Ticklers Oh My!

Every project I'm working on gets a stikkit. Each one of those stikkits gets a #project tag and a nifty aka that I will (hopefully) remember.

Now you might ask what's with the # in front of project? I use that because it makes it show up at the front of the list on the tag page as shown below. I also add the # in front of all of next action contexts for the same reason.



I give my projects a title that contains a couple of conventions. First I always prefix the title with P: because it helps me visually pick projects out when I'm looking at a long list of stikkits. Second, I always put the aka in parenthesis at the end of the project name. It just seems to help me remember the aka if I'm then creating a new stikkit that is related to the project.

Creating a new project might look like the following:



Someday/Maybe things are created in the same way except without the P: and tagged with #someday.

Now that I have a project created, I eventually will need to drum up some next actions for it. Each next action is also a distinct stikkit created as a "to do". Creating a to-do in stikkit is as simple as putting the - character at the beginning of a line. For me it looks similar to the following:



So the first line has the - at the beginning to make the stikkit a todo. The second line I will put Project: {projectaka} if this next action is related to a specific project. Any notes, phone numbers, web links, etc... follow that. Finally I tag the stikkit, making sure that one of the tags is the context (in this example #online).

Since I included the {jeepac} in this stikkit, I can easily see on the right which project this NA is associated with:



Now if I go back to my project's stikkit, I will see any next action's associated with it:



Being able to view a Project's stikkit and immediately see all of the associated next action's is huge for me. However, I've got some projects that have many other things like peeps, ticklers, and just plain old stikkits full of web links or other reference material. As long as I include that {projectaka} somewhere it gets linked back to my project.

Speaking of ticklers, stikkit is awesome for this. Create a stikkit referencing a date and/or time and have it remind you.

Call accountant regarding taxes at 10am on 03/19/2007 remind me

Good times.

The ease of creating these reminders was finally what broke my addiction to Backpack, for a while I couldn't live without my Backpack reminders.

General Information Capture

I'm using stikkit as a big fat inbox. I have the nifty Stikkit! bookmarklet where I can shoot things right over when I'm browsing. I also use the e-mail function of stikkit a lot. For quick notes and stuff I want to process later I just fire off an e-mail to my super-secret stikkit e-mail address being sure to include @inbox in the body somewhere so I can quickly find it later by browsing the inbox tag.

You can also shoot a text message to this address. I have used this a few times, but not as much as I would if I didn't have such a pathetic mobile phone.

Navigation

One of the biggest problems I faced when I started getting a good many stikkits created was finding an effective way to getting to them. Bringing up the whole list of to-do's (by clicking on the check-mark icon up top) presents you with a pretty scary list of things. However, on the right of the to-do list page is a field where you can enter in a tag and just list to-do's associated with that tag. This makes it super simply to grab a list of items by context.

I ended up creating bookmarks to this page for each of my contexts so that I can with one click see my next actions with the #online context (or whatever):



In The End

I love stikkit. I do find it to be a bit slow sometimes, but overall I really think it has helped me get more organized. It took some playing around and trial and error but I think so far it has been worth it.

Next on my list of things to figure out is some good way to back things up. I've got a lot of info in there now and want to make sure I can back it up on a regular basis just in case it all goes boom.

tumblr

A couple of weeks ago I discovered tumblr (via Merlin's). Mine is found at austin.tubmlr.com. I really haven't had a whole ton of time to look into it, but thought it was nice looking and easy.

Today I see where they have added new functionality that allows you to import other various feeds (del.icio.us, flickr, etc...) and have that content post to your tumblr. Neat.

It is the idea that I toyed around with for a while and finally got roughly working over here (stupidly called my "feedbucket"). My original inspiration came from checking out Erik Benson's site quite some time ago.

I'm going to start slinging some stuff there and see how it goes. May free up the time I was spending/wasting playing around with the "feedbucket" thing.

Update: After playing around with Tumblr for a while I've decided that I really like the idea. Mainly because you aren't restricted to attempting to come up with something brilliant to fill up a traditional "post".

Lots of OpenID action...

Since my last post on OpenID there has been all kinds of buzz going on about it. I've compiled a list that I was going to post, but you'd just be better off doing a quick search at places like del.icio.us and technorati. However, I will highlight a few that are the most interesting to me.

1. DHH (creator of Rails) posted this about how he added OpenID authentication to their forthcoming Highrise application. It's only 37Signals, but at least some web-application providers are taking notice.
2. This post by Dan Webb came across my google reader today. A great post about adding OpenID authentication to your Rails app.
   
3. I saw this changeset hit the Rails trac today. A Rails plugin wrapper for OpenID.
   

Thoughts on OpenID?

This is a short post I posted on Geeksouth's Forum on 02/18/2007. OpenID is something I'm closely watching and have some ideas for with some applications I'm involved with.

What do you guys think about OpenID? I’ve just started looking into it slowly over the past couple of weeks and like the idea. Like many, I’m getting sick of setting up usernames and passwords all over the place (we are all a bit prey to the useless account), and would love to have something like OpenID that I could use to handle this if I wanted.

This past October, Technorati announced support for OpenID. Livejournal has been OpenID enabled for some time now, AND if you have a Vox blog then you can use that as an OpenID (where you see an OpenID login box put in your Vox url… example: mine is mrmoody.vox.com)

Also recently announced was a site called idproxy.net. This site allows you to use your Yahoo ID as an OpenID (if that was something you’d like to do). Microsoft has also announced plans to support it.

There are already libraries out there for pretty much every development language which easily allows OpenID authentication for your application.