Out of the blue the other day, Tana announced it was building a “new Tana” product that is AI-first and collaboration-first. The existing product will be around (they say) and supported, and there will be a migration path from the old to the new product, though the new product is a new subscription. (Mixed messages on how much data will be migratable.)
Feels unstable. “Oh, by the way, our new product (delivery date unknown) is the future.” Also, I am not a interested in collaboration or AI-first.
I’m looking for a stable, well-supported, replacement. Ideally a graph database and custom data structures (like Tana’s supertags). Web or local, but prefer local. Migrating data out of Tana is not likely – supertag definitions there are totally lock-in – so import is not much important.
Funny how these PKM apps keep adding stuff until they’re so confusing that the app and team behind it lose their aim and direction. Then they start from scratch again. Same with Mem, Kortex, etc…
One that has not fallen into this trap is Roam. RR keeps to its core principles and has matured into a rock-solid tool true to its aim- connecting ideas.
Yes I was disappointed to read that as well (having been a Tana daily user for over 18 months).
Considering the transparency with which the team communicates and operates, expecting support for the current stand-along single-player version to continue. If not, I plan to move fully into Obsidian (as it plays really well with my Claude setup - Tana does as well which will make this a smoother transition if needed).
And Tana Outliner is not even part of the main website except for a banner at the top routing you to that page.
This does not inspire confidence that the original Tana app will remain in anything close to its current form. And that’s really disappointing because what they built was pretty unique, capable, and required quite a bit of study to master. That’s a huge letdown to people who invested that time learning Tana.
I think the original Tana is going to remain similar its current form, just not grow, i.e. it won’t get the benefit of whatever easy onboarding they come up with for Tana agent, an the mobile app is likely stuck where it’s at as far as features/UX.
The AMA tomorrow should be clarifying.
If someone wanted to leave Tana, I’d probably advise them to see what they could get done with Obsidian and bases/front matter. It would depend on how deep they went with Supertags. I’m not a huge Capacities fan but I wouldn’t advise staying away; i think you just have to try it to see how it feels for you.
I’ve been using Anytype since the start, and it’s excellent in many ways. No outlining, so unlike Tana in that way and doesn’t have Supertags, although you can simulate some of that functionality.
The development path has been a bit rocky - the team has struggled to settle on its core audience (as do many others). Support for new users isn’t as good as it should be, but it’s not that hard to get started. The whole vibe is less “cult” than Tana - no gushing evangelists, but is IMO a bit less exclusive.
The big thing with Anytype is privacy- if that’s important, it’s worth some time
Capacities has not taken VC money, but it’s chasing the same path to profitability (AI assistants) as all the other VC boys are doing. The Privacy Policy does not feel assuring at all. I dislike the idea of notes as objects in Capacities—too complex to me, though probably not complex at all to a Tana user.
Both AnyType and Capacities have an iOS app that is unusable, or barely usable, depending on who you ask.
Unfortunately there’s no good outliner on the market at the moment.
I would say AnyType is second only to Tana in this particular regard.
I suspect we are at a time point where AI is being integrated into all walks of life and the transition is going to be messy. Online services like Tana and others do not want to be left behind. Also, collaboration and sharing are features that can attract business and lock-in.
The best alternative that is not likely to go a similar way to Tana is Obsidian with the dataview plugin and using meta for your supertags. It might be worth exporting a small sample and testing it in the various available products.
Edit
The irony would be if no one or very few move to their new product. It’s a risky business, especially if they are not even going to create a clean and full migration path. That may well alienate them from their current user base.
And that means death - especially for an app intended for power users.
Frankly I think it’s a huge negative that may affect any future Tana products. One of Tana’s main draws has been how passionate their staff is and how often they make innovative Youtubes showing what their app can do.
Pulling the plug like this reminds me of Abandonware gimmicks Google has pulled in the past - but those were free apps.
If Tana is willing to lead people on like this and convicne them to invest both money and time in their app and then suddenly development stops - then why should anyone risk investing time/money on their next app?
Interesting how opinions differ. My experience of Anytype is:
UI is good, but still needs refining to make workflows smoother.
Documentation is poor
IOS app is fine - I use it frequently, although not for major editing projects. It has functional limitations which are a problem for some people (maybe for the majority)
It’s not an outliner and has never attempted to be
On the cultishness - my own experience is that Tana is in a different league. There’s a constant stream of evangelistic content from the Tana influencers extolling the virtues of Tana the AI coach.coworker.personal assistant/researcher/meeting manager or whatever: Youtube is full of them and Slack is full for invitations to workshops/seminars/walkthroughs and whatnot. Among Anytype’s shortcomings is that there is hardly any body of work to show new users what it could do for them. The team produced some starter videos about 3 years ago, but since then, not much.
I like both products very much; I don’t like their focus on collaboration, which leaves us solo knowledge workers on one side. I understand that they need to go where the money is. I’ll probably be saying goodbye to Tana sometime soon - it’s clear that “Tana Outliner” is a sop to pacify a user community that’s being dumped. I have some hope for Anytype, but I might be kidding myself. I’ll stick with it because it’s essentially coal, and sync is fully encrypted, unlike most (all?) of the others.
In the end, if you need to think in the long term. something like Obsidian is probably the way to go: data is in a robust, recoverable format and you can operate without an active internet connection. I don’t get on with it, but that’s me.
Just got off the AMA. The new Tana looks quite good! More powerful in some ways with the editing tools. Not as wild an outliner (though still does outlining.)
Outliner they are serious about continuing to develop and they shared some of their roadmap for it.
There will be an import from Outliner to new Tana, i.e. you won’t be able to view the same data with both. So a decision there.
They’re hoping to get new Tana in people’s hands late summer/early fall. Interested to see Tana Outliner’s release nodes between now and then.
I was initially skeptical but I do like the demo of the new Tana, which looks like a more powerful version of Mem. I’m not interested in meetings or collaboration, but it would be useful to be able to create documents, outliners, kanhan boards, and canvases in one app for different purposes.
Tana and Capacities are among the few apps which allow users to “type” their notes. I subscribed to either of them at some point. Ultimately I prefer Tana due to the speed and UI.
From their description I am not sure the new Tana meets your stated use case. Indeed the description is of an app that has a totally different use than the original Tana Outliner:
Not to be too cynical, but I wouldn’t make any significant decisions based on a demo, especially of a software product that won’t be available for some months and especially especially one coming from an outfit that has, for all practical purposes, failed to deliver on its major promise.
I understand very well why the Tana team might have been forced into its pivot, but that doesn’t change the fact that it is a pivot away from what it promised.
I would say a couple things. First, they will still have a dedicated team working I. Tana Outliner and emphasized that they will be bringing new features to it as they have in the past. Second, do the new Tana they also emphasized that by meeting they were referring to a whole host of things including you just communicating with your PKM solo - so the new Tana, while good for collaboration, does not require it. While I was very taken aback by the initial announcement, I was very reassured that Tana Outliner which is what I may keep using, will continue and receive new features.
They’ve clearly stated that these are two products. Supporting two complex products (albeit mostly on the same technical foundation), with different target markets, seems to me to be an unsustainable business in the long run. Plus the internal risk of product teams competing for resources is real, despite the gloss management puts on the problem. I cannot imagine that the two-product scenario will last past this year, even if the company survives.
Public info indicates Tana has raised $25 million – on the basis of being a single-product company. Not much especially since they are starting over with a new product which will crank up the burn rate. The next product is not 100% new, but focused on capturing a segment they haven’t focused on.
