Another Experiment

For the last couple of days I’ve been experimenting with a variant of Autofocus 2. This was a system which I developed in response to the perception that Autofocus had difficulty dealing with urgent tasks once the list had grown beyond a certain size - which it inevitably did.

Unfortunately Autofocus 2 suffered from the same problem in reverse. It was great at dealing with urgent tasks - which is why I still use it for panic lists - but not so good at dealing with non-urgent tasks, particularly the more difficult ones which tended to congregate at the beginning of the list.

So I’ve been looking at how I can improve Autofocus 2 so that it keeps its ability to deal with urgent tasks, but keeps the work going on the beginning of the list as well.

I’ve found a very simple change which seems to solve the problem. Just divide the list into two. Here’s how:

1. Draw up a list of things to be done. Keep it reasonably short at this stage. My first attempt was 23 tasks long and so far it’s working fine.

2. Draw a line at the end of the list. Everything that you enter or re-enter from now on goes after the line. 

3. You have now two parts, the Old List and the New List.

4. You scan from the end of the Old List back towards the beginning of the list. When you come to a task that stands out as being ready to do, you do it. Then delete it and, if there is still work to be done on it or if it will recur again within the next few days, re-enter it at the end of the New List.

5. Now scan back from the end of the New List in the same way. When a task stands out do it and re-enter it as necessary.

6. If no task stands out continue your scan in the Old List as in 4.

7. Continue in this way, scanning for one task in each list alternately until there are no more tasks in the Old List. When this happens draw a line at the end of the New List which becomes the Old List, and start a New List.

8. Please note carefully that if no task stands out in a List, just carry straight on to the other list. But if a task does stand out, once you’ve done it you do not scan further in that list but move immediately to the other list.

The rules sound more complicated than they actually are. In fact it’s a very simple procedure.

If anyone wants to accompany me in this experiment, please do so. Your comments are welcome.

P.S. Make sure that 1) you always scan backwards from the end of a list towards the beginning. 2) you only pick one task before moving on to the other list.

One last point. When the Old List is down to just a few tasks and none of them get selected on repeated scans of the list you may either delete them or move them to the end of the New List (which will then be free to become the Old List).

Comments

[Caibre65]

AF2 is one of my favourite methods so I'll give this a try.

[Paul MacNeil]

I'll be happy to give this one a shot. I gave up on the Nov. 29th system pretty quickly -- there's an oppressive feeling deep in my chest sometimes when a system doesn't work. I get that feeling if I go back to a straight-up GTD or even on occasion going back to my Franklin Planner. Too much overhead or something. Before that, I had been using skedpal and loving it, but there was getting to be too much work in setting it up and entering tasks with the right variables. I missed the simplicity of the autofocus systems where you just wrote stuff on a list. Then I tried your "same old thing" system and it has been really amazing! I've been meaning to write about it. It seems to have all the benefits of Random (melting procrastination especially) with more control over what I do. I love it! And I'm able to melt my list like wax and I feel great. So I'll try this new variation of AF2, but I think I have a new standard to return to now. Thanks!

[Brendan Howard]

Dude. I've been GTD-ing for years and years and I finally started dreading the long "Next Action" lists broken out into contexts. Your todo list with a simple rule of three, picking the first on the list then asking a question to find two more ... it's genius. Thanks, man.

[Mark Forster]

Paul MacNeil:

<< there's an oppressive feeling deep in my chest sometimes when a system doesn't work >>

I know that feeling only too well!

[Alan Baljeu]

I'll give this a go. Seems quite compatible with how I'm organizing my tasks now.

[Cameron]

Interesting idea. Going to try it. Just when I thought you've uncovered nearly every type of scanning algorithm you go and invent a new one.

[vegheadjones]

Trying it today...

Mark did you try the "skiing" method I laid out on Dec 6? http://markforster.squarespace.com/blog/2020/11/30/experimenting-with-a-new-system-day-2-disaster.html#comments

I've been working it since 12/2 and really like it. I am curious enough to try this AF2 S(squared), but may slip back into ski mode tomorrow.

[Mark Forster]

vegheadjones:

<< did you try the "skiing" method I laid out on Dec 6? >>

Yes, I did, but not quite the same as you because I was using it with one long list rather than the various different lists you are using. My first impression that it works very well was confirmed. I want to give it a go sometime with several lists, but at the moment other projects are calling!

[Alan Baljeu]

> AF2 S(squared)

I'm calling it AF2x2. It's similar in ways to AF4.

[Voluntas]

DualFocus

[JD]

Hello Mark,

A question on the task processing in the Old and New list.

Does a task need to done on the either list when moving from New List to the Old List and from Old List to the New List?

My reading of the rules is that you don't have to, but then I was confused when I read the last paragraph where you say:

<<One last point. If no task stands out from the Old List you should consider either deleting the remaining tasks or moving them to the end of the New List>>

[Mark Forster]

JD:

Sorry, I realise that it is rather confusing as it stands.

It is designed for when the Old List is down to just a few tasks and none of them get selected on repeated scans of the list. In that case you should consider either deleting them or moving them to the end of the New List (which will then be free to become the Old List).

I have re-written it so it is clearer what I mean.

[skeg]

Mark

I have been trying this and as you say its easy to implement. The only thing I find difficult is to always to remember to scan backwards when used to adding new or unfinished items forward. As you only ever take on one item in each list does it matter if you do it forward.

[Mark Forster]

skeg:

<< As you only ever take on one item in each list does it matter if you do it forward. >>

I've tried both and they both work ok. I slightly prefer backwards myself, but if you find forwards more to your liking then that shouldn't be a problem.

[skeg]

Mark , thanks for the response. Its interesting that even after all thr systems you have developed it still comes down to balancing urgent and less urgent items. When I first read the rules it made me think of AF4 rather thsn AF2 given Old and New lists. I have always been more keen on AF systems than FV systems because to me AF rules largely take the decision making out of your hands whereas FV meant (to me at least) that you were always asking yourself the question of what to do next. Given the time of year unlikely to be able to use any system over next couple of weeks but will give this more of a go. It reminds me of a version on AF1 I used years ago. On 20 line pages I would have a rule that I would only do one item per page but always had to do one. Similar in a way to the Ten Task system. Anyway thanks for ALL the systems, there is something there for every one!

[Alan Baljeu]

I wasn’t happy with moving the whole list to Old and starting a new list empty, so I’m making an amendment: When the Old List is clesr, instead of drawing a line at the bottom, I’m drawing a new line more towards the middle. You decide if that means one day back or all-but-one, or whatever. Anyway, urgent items remain in New, and some previously neglected items get to the bottom of the Old and will get attention.

[Mark Forster]

Alan:

Yes, that sounds as if it would work well. Let us know how you get on with it.

[Len Black]

Mark, thanks for the autofocus systems! Happy new autofocus year!

[Mark Forster]

Len:

And happy new year to you and everyone!

[Alan Baljeu]

One day later: It seemed to work fine, but I wasn't satisfied with the balance of tasks: I want to do more New and less Old. But I am inspired, and after thinking through the details, the next iteration I will try is: Serial Two List. No description provided yet.

[Mark Forster]

Alan Baljeu:

<< I wasn't satisfied with the balance of tasks: I want to do more New and less Old. >>

I don't quite understand why that is a problem for you. Using my rules in this blog post, you are scanning backward. That means that when the New List is the smaller, the tasks you come to first in the Old List are quite new. When the Old List is the smaller, then there are relatively few tasks in it, and you may well by-pass it under Rule 8 without choosing a task.

So whatever the balance of the Old and New lists, you can choose all new tasks (or all old tasks for that matter).

[Alan Baljeu]

Despite the words “please note carefully”, I didn’t notice that rule at all!

[Tommy G]

Hello all.
Hello Mark

Thanks for share this idea.
I want to ask if this system better compare to FVP? Since you mentioned in older blog that FVP is the best system

Thanks again for sharing your thought and experiences

[Colin]

I'm enjoying trialling this method as a new year's experiment, and getting a mild sense of glee in seeing things move below the line and the list above the line shrink down. So far so good.

[Christopher E.]

I’ve been using this system with some effectiveness. I do need a push to work on the older tasks, which this system does. However, I find myself usually needing to spend more time in the latter part of the list. Is there a modification to the rules you would suggest to help me spend enough time in the open list without losing some pressure on the closed list?

[Christopher E.]

Replying to my own question:

I came onboard the Mark Forster Project around the release of AF4. Now that I look at AF2, on which this “Another Experiment” is based, I think AF2 might better fit my needs. I’m going to try that!

[Mark Forster]

Christopher E.

A modfication you might try out is to divide the list into two as before, but instead of doing one task in each part in turn, you continue circulating round one half until nothing further stands out, and then move to the other half and do the same.