⚡️ Ideas
DanDanDan DanDanDan Aug 2, 2024

Current impression of the status of Tana

Explain the problem as you see it

Dear Tana Team and Community Members,

I would like to share my current impression of the status of Tana a few weeks after the transition to a subscription model.

To the Tana Team: Please consider this as input - it's important to me!

I first used Tana in 2023 and was overwhelmed by the features and flow, unlike any other note-taking app I had experienced.

I was fully aware that it was a test phase, and I deliberately overlooked many small and large problems since I hadn't paid a cent.

However, over time, I encountered more and more issues that at least delayed my all-in commitment. When Tana Core was announced, I was confident that, in addition to the AI features, there would also be a comprehensive update to the core functions of a note-taking app.

Unfortunately, this is where the problems begin. As I said, please take this as input. I have subscribed to Tana Core and will continue to do so.

Basic Functions:

Tana still lacks fundamental basic functions that every note-taking app must have. I must honestly say that I am more than amazed that Tana has still not implemented these, even though the community has considered them important for months.
Media Handling: Working with images and PDFs is still a disaster. Sorry, but I have to be that blunt.
Long-Form Writing: I think it's absolutely okay that Tana doesn't focus on this. Anyone who writes seriously will do so in professional word processing programs anyway. Some styling would be nice, but for me, that's absolutely fine.
Import/Export: This is one of Tana's biggest construction sites. If it's not even possible to re-import the only export back into Tana, that says everything about the state of this issue. The JSON is completely useless for 99% of all users. Attachments are not included. A bulk import is completely missing. I have 800 markdown files with many images in my Obsidian vault. Without importing all these notes, using Tana makes no sense to me. I simply can't start from scratch again. To be honest, it's a no-go for me to switch to a paid model without being in control of my notes (import & export).
Speed: The hoped-for speed boost after switching your backend has unfortunately not materialized. The loading time of the web app averages 8 seconds with a really good connection in my office. As described in point 4, I currently don't have many nodes in Tana. I don't want to know how slow Tana will become if I import all my notes.
Supertags & Search: This depends on the user. The possibilities that Tana offers are unique and breathtaking. It works and is fun. However, some functions are super complicated to implement, only understandable after reading the documentation, and often not intuitive. Sometimes it feels like I'm coding something that other note-taking apps offer at the push of a button.
Mobile App: This seems to be a bigger problem. Normally, this should be accessible after switching to a paid model. But knowing how Tana works in the backend, I can imagine how difficult a mobile implementation is. However, for users, it's an absolute must-have.
Offline: For many, this determines whether Tana will be used long-term or not. The way Tana works, I can't imagine there will ever be an offline mode. This will be a show-stopper for many. We must be honest about that.
AI Features:

When I first heard about the Tana Core features, I was electrified. The initial tinkering with them was really good. However, this has completely turned around, and in my daily work, I hardly use Tana's AI features anymore.
Meeting Agent: A great technical feature that works technically well but cannot be used in real life, especially in a work environment. No one wants to be transcribed word for word. No one wants an external bot in important meetings where it's unclear where my data goes, how it's processed, and what happens to it.
AI Chat: Unfortunately, it offers no added value. When I ask the same prompts in Chat-GPT and Tana AI Chat, Chat-GPT produces significantly better results 8 out of 10 times. Often, the AI Chat stops answering in the middle of a sentence. It was announced as "Chat with your notes." The problem is that I always have to manually specify a precise context. A simple question like "What is the phone number of customer XY" already yields no results without specifying the exact node where the phone number is stored. I don't need an AI chat for that. Simple questions like "What meetings did I have last week" produce no results even if I give the week node as context. For me, it's currently just a visual representation of a Chat-GPT window with worse results than Chat-GPT. Sorry for the blunt words, but that's my actual user experience.
Communication:

After switching to a paid model, users can expect open and honest communication. Tana still has many construction sites, and users need to know what is currently being worked on. What has priority? What is coming later this year? What will never come? These decisions affect whether we will use Tana or not. If a central feature like import/export is not yet satisfactorily implemented, I can temporarily accept that if I know, for example, that the feature will be implemented soon.
Roadmap/What's Next: Look at how openly and honestly Capacities communicates about the current status and upcoming features on their website. I myself work in the communications field, and Capacities does this almost perfectly. At Tana, I have absolutely no idea what you are currently working on and what is planned next. I don't need exact dates, just an overview: What has priority for you right now?
I hope you can see that Tana and its team are important to me; otherwise, I wouldn't have spent half an hour writing this down.

Best regards,

Daniel.

Why is this a problem for you?

Mentioned in the long text.

Suggest a solution

Open up a clear communication-channel on your website.

⁨1⁩ ⁨Comment⁩

100%. I know this isn’t a specific feature request, but if I were to title it I would call it “Polish your app to at least feature parity with competing free apps, before selling it as a paid app”

Tana feels half-baked right now and I think there’s similar sentiment on Slack.

The AI Chat in particular needs to have feature parity with either Notion AI (RAG-based Q&A) or Poe/TypingMind (an unopinionated LLM client).

Right now it’s as basic as someone’s overnight hackathon project built with the Vercel AI SDK. Actually, worse than open source hackathon chat UIs I’ve tried: it doesn’t follow the structured output of Tana nodes format 100% of the time, and doesn’t accept multimodal input.

Not to mention they decided to include Dall-E and charge for it, when even ChatGPT offers it for free. I think if this is Midjourney or Flux Realism LoRA level of quality, it would be worth charging for. Otherwise it should be free as in beer.

Charging for Tana AI right now feels like a ripoff, even though they’re just passing on the cost of the OpenAI API to us.