Multiple Contexts per Task [now in TestFlight]

@msalam, I don’t know how it would work in OF2 on the iPhone or OF1 on the iPad. Although I own both, I am doing all of my testing on the Mac. I will give it a try when I get a chance.

And I agree with what you said. I simply don’t understand why (as you put it) some are concerned that “anarchy” is a short step away from multiple contexts (I am being facetious of course!).

But (playing devil’s advocate to myself) - given I work with databases, I totally understand the complexity of relational databases and many-many relationships. So I don’t much mind supporting this particular use case with keywords instead of another column in the db. As long as it works - and through the support of the search feature in perspectives, AFAIC, it does work just fine.

Someday (I think) Omni will support added metadata in the records and the world will change. :)


By the way, the term “Next Action” has been removed from OF2, so it seems we are leaving the pure path anyway. :^}

… so lets have our meta-data tag field to use as we like.

select * from omniDB where tag like '‘We Want Freaking Tags%’

I’ve been very excited about OF2. Having started managing tasks in Things, graduating to the more robust capability of 2Do, OF2’s clean UI, greater tools for control and forecasting all had me ready to ante up my money. The lack of tags, however, is an absolute deal-breaker for me. I’m so bummed! However you should do it theoretically, I often want to have a task tagged with a type (write a newsletter article), tool (do it in Mail Chimp, a person (need to talk to Beth to make sure I’ve got my facts right), and a general timeframe that I want to remember (NW for sometime next week). With robust tagging capability plus contexts, this would be easy: @mailchimp write Beth NW.

Not having this capability really limits with unnecessary inflexibility what otherwise looks like a wonderful piece of software.

1 Like

Would it be easier to think of writing a newsletter article as a mini-project?

Project name: Write newsletter article (Defer to: May 1, 2014, Due: May 15, 2014)
Task: Talk to Beth re: blog post for May 2014 (Context: Beth)
Task: Write article (Context: Mail Chimp)
Task: E-mail Beth for editorial approval re: blog post for May 2014 (Context: e-mail)

So this wouldn’t be just one task. This is actually a project with at least 3 tasks. I know I would love to see OmniFocus become flexible enough with tags that it can fit even more user types. But I guess we gotta do what we gotta do to get things done.

If this is a common activity (writing blog post) then you can probably save this as a project template with project status set to on-hold. Then when you need to write an article, highlight the project and duplicate it. Then change the defer date, due date, and the person that has editorial approval.

1 Like

You are right - in the classic meaning of GTD. Everything that has more then 1 step, is a project.
You example of breaking down “writing a newsletter” has one major disadvantage:
If you are doing this breakdown, you are forced to do it w/ every other task. BUT: Not everyone does that - including me.
I think this multiple context discussion derives from a much deeper problem the user have (incl. me). How much Micromanaging are you allowing yourself?
Tbh I find it rather cumbersome to breakdown a funky (simple) task like a “newsletter” to 5 steps. This is micromanaging and can lead to not being productive. Where is the limit? Do you check off “move left leg”, then “move right leg”, then “stand up”, then “walk to the door”, “open the door”, “go into bathroom”, “take toothbrush in your hand”, “put toothbrush in your mouth” etc. - you get the irony - in your morning ritual?
E.g. writing a newsletter could be something which can be done automatically - you do not need every single step for that - assuming you know what to do (trained as a routine/habit).
Now we come back to multiple contexts. Assuming one (like) me knows exactly what to do w/ “writing newsletter re: xy”, you could add this in a a) repeating routine or an b)SAL (Single-Action-List). Well, what is then the context? Yeah, we do not know! Imo this is why alot (the majority?) of people do not know what to do w/ contexts. We find ourselves experimenting w/ energy levels, classic GTD-approaches, etc. made a really nice a approach w/ a fresh look on contexts (quick dash, think etc.) We all know that.
Imo this reflects the need to re-invent or overhaul contexts.
At the moment I use very few contexts (body, mind, passion <> locations <> agendas and waiting for), but I also would like to see an addon like tags (not multiple contexts). Tagging could be interesting in prioritising (ABC) or tagging w/ mood / energy aso.

I also wished to have multiple contexts or tags at some point in OF2 as well. Don’t get me wrong.

But I think that we only have to break down tasks only when needed. Things like “going to the bathroom” don’t need to be broken down if it is an automatic routine or habit.

Breaking a project down is best used when trying to get unstuck at something that doesn’t seem to be moving or working on an ad hoc project that you’ve never tried before.

If one has never baked a cake, a project would have to be broken down into smaller steps. But once you get the hang of it, you can just add a task called “bake a cake for Friday’s birthday party.”

one of these days, we’ll eventually get the Holy Grail.

In the same sense of passion, Things was a no-go for me because it didn’t have the review process and Forecast view that was introduced in OF for iPad and brought over to OF2 for Mac. But it’s interesting to see very different takes in the world of task management. So, I am grateful at seeing different ways of tackling this subject.

@Sionker, I expect you’ve probably read your way through this thread. But just in case you have not, did you see this post (my own) up- thread?

I guess I’m puzzled why more of the multi context or multi tag supporters (and I am definitely one of them!) don’t adopt adding keywords to title and/or descriptions and then creating perspectives which include searches for those keywords. Why is that not suggested as a solution more often?

Not judging, just curious…

I’m gonna take a guess.

Because some users are so hard-wired into using tags as the best possible solution for them and they look with contempt upon using the search filter as a “hack.”

I’m for using keywords or hashtags in the task description or notes. But I guess some folks just can’t get past the idea of the lack of tags in OF2.

I always liked finding “hacks” to make my program work. If there is a missing feature, I’ll try to find a way to make it work.

@wilsonng, I take your point. My thinking is that support for free form text searches essentially amounts to support for tag searches (that is, if you add your own tags).

So I don’t even consider it a hack, since it isn’t hidden. It is a part of OF2, and incidentally more easily accessed than with OF1. It could be easily added when setting up perspectives, no hacking required.

That said, I support tags and even multiple contexts. I just know that’s a big deal to set up in the database, because you are extending relationships. It’s not trivial.


Hmm, I agree, it’s not a hack- but it takes quite an effort to get the whole thing implemented, compared to auto-completing tags in other Things it’s a real chore in OF…

Omni is forcing this route since day one- and for a reason. Every time I try to actually take the other route (and you might guess in what environment) I consider adding this tag and that tag- creating a bunch of tags that I am later on not assigning any value to and therefore losing myself… while your mileage in concerns of control over this might (SHOULD!) vary, I see the reason for this and love to readapt to the restrictiong of one context.

But this strategy is simply keeping folks out that excel in handling tags- why not have the option to use them? I could stay within OF for my weird experiments, then ;-)))…

yeah, I do like the auto-completion part of tags. But I guess I’ve resorted to using TextExpander so that my keywords stay consistent across programs such as Aperture, OmniFocus, and Maverick’s tagging system.


I think for anyone working in any decently large, decentralized organization this is critical feature, especially for the “pro” user. I can see either a “people” context or a tag, I don’t care which as long as it autocompletes. That would actually make the pro price tag worth it vs. just paying for features I already had in 1.0

1 Like

I don’t understand why people need multiple contexts to perform a specific task. My understanding of contexts is that is the place or person or object needed to complete a task. I am not saying that there are times when completing a task where I might need more than one task, but in those situations if I need to use/visit different contexts I set that task up as a project and break it down for each context. I could always set up Grocery Shopping as a task, but I buy items over the course of 3 or 4 stores. I could effectively buy several items in any of the 4 stores, but there are some items I buy that are only available at a particular store. Basically I then set up Grocery Shopping as a PROJECT and in that project list each grocery item I want to get with the appropriate context. The groceries that I know I can buy at any of the stores I’ll assign the context with the store name I plan to go to first. If that store is out of stock I can get it at the next store. When I am at the store I can access the context and it list the items of that store. I get the items needed. In store 2 I pull up that context and get those items. then reference the Grocery Shopping Project for any items missed from store 1 and the items from the other stores, If those items aren’t at this store, go on and repeat with the next store etc.

Too me, having too many contexts as tags is just too confusing and really blows up the size of the data base of tasks and slows everything down. I have been using OmniFocus since 2011 and did participate in the OF2 beta test so I think I have a pretty good familiarity with how to use the program despite the simplistic task/project of grocery shopping. In addition the search function pulls up anything that basically contains the words being searched. so if you want multiple tags, add those words to the description or title of the task.

I love OmniFocus because it is not cluttered and it is simple to use once you get the hang of it. Having more than one context is just adding clutter and complexity to OmniFocus.


I used to think this was an issue until I discovered three things…

  1. Nested Contexts
  2. Choosing Multiple Contexts in a perspective
  3. Using the Find Text field to filter in Notes

So for 1) if you created nested contexts right you can assign tasks to your “Phone” AND “Any Device” for example. e.g. create a parent context called “Any Device” then sub contexts called “Phone” and “Computer”. Any task you put in Phone OR Computer will also appear in “Any Device” too. You dont have to put it in Phone AND Computer for a task you can do on both.

For 2) You can select the contexts you want to see tasks for and save it as a perspective. So if you did have just Phone and Computer (and not nesting like in 1 above) you can assign a task to Phone for example, and another to Computer for example, and then choose to view both Phone and Computer if you are in Work say, and you can see both tasks. So you dont have to assign a task to both Phone AND Computer if you can do it on both devices, but add it to one of them, and select both of them when you have both Phone and Computer available.

  1. You can add multiple tags to the Notes field, e.g. ##phone## ##computer## and then create a perspective that searches for ##phone## in the “Find Text” field, this will show all tasks with ##phone## in them. So you have added multiple tags that way.

But i dont think this means multiple contexts or tags is not necessary… the issue is Omifocus is taking away peoples choice. Some people have their own way to manage tasks, and may not follow GTD 100%, and that should be fine. Its quite bad that software forces you to follow a specific methodology when everyone is different.

I’ve also wasted HOURS on trying to get contexts and things right, when a more flexible software would have allowed me to do it how I want and save time. But now I know how to do the above 3 things, thats Ok for me, I can simulate multiple contexts that way.


I think it’s pretty straight forward if you have any kind of complex work environment. I support a world wide sales force, interact with local teams where more than one member could be capable of doing a task, and have management oversight on all of these people.

Let’s say I need to talk to Fred, he lives in NZ. I’m in California. Most likely I will call him, so he goes on my call list. Just so happens Fred is in town for a customer visit I didn’t know about and I run into him in the hall. I would love to pull up my task, but it’s in phone, not Fred. I could search, but sadly there is more than one Fred in my far flung company. Having a people context or tag would be highly useful for these situations.

As a system it would also stops me from accidentally using more tags than I need as I could leverage autocomplete. I don’t have a ton of contexts because I auto complete.

I fully agree, it could clutter the interface if you put 50 tags on each item, same thing with Evernote or any other system with tags. But much like everything else in the software, if you don’t want to use it, then don’t use it. I don’t use time estimates at all, yet I can see how people would find them useful and I simply ignore them and get on with my life.


Awlogan pretty much sums up what I think is a very reasonable position: some people feel strongly that they can use Tags. It works for them. If others don’t want to use Tags, they wouldn’t have to. But why deny someone the ability to adapt an app to their particular needs, their workflow?

Even if I didn’t have any use for, or interest in, Tags, I could look at the the example Awlogan wrote and see that it would be quite helpful to him. More helpful, in fact, than any capability OF2 currently offers. I would have to ask myself “Would I really be negatively impacted if he were given that capability?”


Want to chime in and just say that we should think about WHY we want multiple contexts.

Is it only because it doesn’t fit how you want to organize your tasks?
or is it because it is disrupting your ACTUAL workflow?

Your Omnifocus is NOT your workflow. Doesn’t matter how organized your tasks are. So “I wish I can tag/context my tasks with person and place and energy level”, why? just to tag them? or does it actually really help you in DOING the tasks?

GTD is getting things DONE, not getting things ORGANIZED. I think (and I’m guilty of this too), we can easily fall into the trap of thinking that organization of tasks is synonymous with doing the tasks productively.

So, the take away is, if you want multiple contexts or tags, why? What is a REAL LIFE, scenario where it would benefit you in getting things DONE, not necessarily finding tasks in your OF. Are you constantly forgetting to do things for a specific person because you are missing a context tag for that? or you can’t find a way to see all the tasks you have to for a particular company in one glance? Can you FIX that with the current structure without adding complexity (i.e. multiple tags). I think OF is flexible enough that you can.

Don’t be rigid in thinking you NEED a certain structure because David Allen says so. Create contexts for what you NEED and nothing more. And determine your NEED from your actual WORK not your perception of how your work should be organized.


Tangled1, please read the posts above yours, I explained a real world case where a lack of tags or multiple contexts is an issue. Being organized is how I get things done, if it doesn’t match what you want ignore it.

1 Like

I’m still puzzled why some folks feel so strongly that some of us should not have an option that we feel we need, and that they will never have to see, much less use.


As I believe we said on the old forums, Omni does see a value in a feature like tags. However, so far we’ve had a bunch of work that we saw as being more valuable. Since we can’t do everything at once, in the short term adding any particular feature is, in a sense, taking away a feature we could have built instead. But we’re still working on OmniFocus every day.