Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Notion AI – waiting list signup (notion.so)
361 points by antouank on Nov 16, 2022 | hide | past | favorite | 321 comments


There are hundreds of "use AI to help you draft text" apps out there, and it really feels like a "feature, not product" category of thing. So I'm glad to see a larger player like Notion integrate this feature.

That said, every implementation (including Notion's, which admittedly I haven't dug into very much) seems to be same stale "predict the next sentence" interaction, which reminds me very much of the early web's "it's like a store, but online!" vibe. In the sense of "I built this because it was straightforward to build" and not "I built this because I think it best meets what users need". To be fair, building the former can be a prerequisite to building the latter.

But honestly, most writing comes in editing; where is the help for editing? For rephrasing, or for making text flow more easily, or for reorganizing an outline and having the text update to reflect that higher-level reorganization? Where is the support for finding other text you've previously written that might be appropriate?

As someone who writes a lot (granted, in an academic setting), "first draft" is definitely part of the process, but it's just one piece! I suspect that in a few years today's tools will feel like ancient chisel-and-stone implements compared with what will come online. Can we get there already?


We've been building a new kind of collaborative text editor (https://www.orchard.ink) keeping all of this in mind! We're trying to reimagine writing workflows, focusing on creative commands for editing that are easily accessible through a Cmd-K. The generated suggestions are displayed as Google Docs-style comments that you can cycle through until you find one you like.

Wasn't planning on making this public so soon (it's only been 2 weeks), but thought people here would be interested in checking it out. Got a lot of changes coming, so stay tuned! Happy to discuss at kevin[at]village.dev


I tried typing something. I don't understand what it takes for one of the commands to actually get AI involved in my text. Either that or the button click events weren't firing. I noticed the template lists were all blank; do I need to make templates somewhere first and then start writing?


I think we've fixed the bugs you're referring to—could you give it another shot? You can trigger commands by pulling up the cmd-k menu at the bottom of the screen.


I know Grammarly gets a lot of hate but their value really is in the editing process. Pro features are awesome and provides you with helpful tweaks based on your tone preferences.


> reorganizing an outline and having the text update to reflect that higher-level reorganization

Can you elaborate on this particular point and how you would see it work in a collaborative doc/workspace?

Outlining is very underrated as a method of editing and organization, and most outliners are often designed for offline and personal use. I can see a lot of potential in adding automation to how you can organize and connect multiple nodes across outlines.

P.S. David Pierce wrote an in-depth review [1] with some interesting use cases. He was the one that originally broke Notion's story on WSJ a few years back.

[1] https://twitter.com/pierce/status/1592882875254595584


> For rephrasing, or for making text flow more easily, or for reorganizing an outline and having the text update to reflect that higher-level reorganization?

Current models can kind of do this, e.g. you could give some text to GPT-3 and ask it to do this. Not sure what the limit on the input length before results degraded would be. But yep, would be nice to see this out there in a nice interface and working well.

> Where is the support for finding other text you've previously written that might be appropriate?

Semantic search could be used for this, e.g. finding blocks of text similar to what you've already written

But yep, this stuff isn't all nicely packaged in one place


100% agree.

The editing process is a lot harder to build around than the more straightforward "continue this text for me". Partially because the models are really suited to do the latter out of the box, and partially because there's more nuance and subjectivity on the latter and it's harder to make it work well today.

That said, totally doable. (We put most of our effort into AI for human editing and control with Sudowrite.)


These are all great points but Notion is not an AI company. They are most likely using gpt3 or equivalent in the backend to provide a (rather gimmicky) feature, that's it.


Similarly feelings with the ML-based code writing tools. Where's the help with debugging??


Should focus on summarizing existing notes that nobody is reading instead of generating more noise.


would be amazing to have a suggestion like "looks like whatever you're writing has already been written in this workspace"


> ...I can't take it anymore, the guilt is eating me. I've had an affair with your wife...

[looks like whatever you're writing has already been written in this workspace]


This sounds like a beginning of a sci-fi mystery story. Imagine this happening, but on a personal workspace with single user.


I noticed Grammarly just entered the AI summarization space for email. https://www.grammarly.com/recap. I haven't tested it yet, but saw the Gmail banner last week.


I found that in my current job I don't so much need a summarization service, but a translation service; what does the writer actually want me to know and to do? I got an email the other day with a lot of brain farts, grainy screenshots, and I had to read it five times just to boil it down to "change this tracked URL to that one".


Yes please.

I’ve been toying with the idea of making a browser extension that summarizes pages into ~5 bullets, <20 words each. And maybe a picture.

Someone please make this.

Too lazy to read walls of text. Too lazy to build the extension myself.


This already exists on Reddit as a bot that says “the best I did was reducing this article by 87%, here it is…”

I imagine you just need to wrap that in a browser plug-in. Or maybe already exists.


This is done by autotldr[1] which is powered by smmry[2]. Smmry is pretty neat actually !

[1] http://autotldr.io/

[2] https://smmry.com/


Sadly it seems to fail bad on things that aren't new articles.

Try for example https://en.wikipedia.org/wiki/Functional_programming/ as an input to smmry - you don't get the idea


I mean, it works ok. I think the issue here is structured documents like that wikipedia article. If you used smmry on each section, leaving the headers and structure alone, you'd probably get a far more 'coherent' result, and it's not surprising that the tool is built for "single block" content.

It would probably not be very hard to build a mediawiki smmrly extension that does just that.


We're testing something that addresses this need right now. Email me if you wanna try the beta (amit @ sudowrite)


It exists in various forms and you are not the only one that desires this.

https://tldrthis.com/

https://www.summari.com/products/chrome

Etc etc


title, subtitle, and thesis statement from each paragraph.


This sort of hits on what I was thinking when I saw this feature: already, nobody reads anything. And now, nobody has to write anything either.


Now middle managers can create even more busy work while doing even less actual effort.

Of course no one will lay them off for that.


Sort of unrelated (but someone at Notion may actually see it here): PLEASE NOTION, FOR THE LOVE OF GOD, make the tap targets for "to-do" lists larger on mobile devices. I can't stand making a to-do list in Notion if I know I have to use it on mobile. 9/10 times I tap the box to check it off, and it goes to edit mode instead of "checking it". It drives me crazy. Make it more mobile friendly, or make it an option to make it larger.

It literally is a SIMPLE fix, and I don't say that lightly. Since it's a web app you can just add a basic CSS media query for mobile devices to make the padding/sizing larger for mobile devices.


They're in the business of pushing new features. This is why problems like this (and many others) will go unfixed for a long time. Long gone are the days of prioritizing fixes in Notion (which BTW their "performance fix" before they hit hyper growth was abysmal).


The day bug fixes drive sales you will see companies prioritizing them. This isn’t impossible either, you just have to make “no open bugs” a marketing term.


Would be better if Notion did it, but if you're on Android using Firefox, then install uBlockOrigin and add the following filters:

  www.notion.so##.checkboxSquare:upward(2):style(width:30px !important; height:30px !important;)
  www.notion.so##.check:upward(2):style(width:30px !important; height:30px !important;)


For me the number 1 reason of not using notion for all my note taking needs, is that the app is super slow on my device. Granted it is an old LG G6, but there are also apps, which are quite responsive.


Every time a new easy-to-use content generation AI tool drops, the collective blood pressure in Google's spam detection department goes up by 1% :D

This one should be promising!


I think that department has already had a stroke.

For the last month or so, on a daily basis, I have been getting the same spam messages (subject line: "You've been chosen”) in my inbox, often classified as important. No amount of reporting them as spam seem to help.

However, what ended up in my spam folder are emails from Amazon informing me about winning $100 in their sweepstakes. (When I found them it was too late to claim the prize as confirmed by Amazon's rep.)

I get that spam is hard, but weighing an account owner's signals more than other signals seems like a good strategy. (And if it is not, please do correct me!)


> Amazon informing me about winning $100 in their sweepstakes.

Is this for real?


Enter your credit card details to find out.


Yup. Look up: Small Business Prime Day 2022 Sweepstakes


I have been getting those emails too, and I noticed that they get automatically send to spam ~5 minutes after I receive them.

I seem to be getting a spam email of this kind almost every day for the last ~6 months. Maybe the spammer found a way to prevent fingerprinting by Google and each batch of emails only gets automatically marked as spam after so many people manually mark it?


It has been the same for me. A small but noteworthy fraction of legitimate emails go to Spam and illegitimate emails come into Inbox; in spite of marking "spam" or "not spam" multiple times.


>Amazon informing me about winning $100 in their sweepstakes

Are you joking? Surely this is not something that exists.


It did exist. Look up: Small Business Prime Day 2022 Sweepstakes


Meanwhile my mailbox decided to send every single email to spam.


Another day older and a deeper neural net, one step closer to a dead internet.


It's going to be real fun when more than half of "user" generated content on the net will have been generated by AI content farms: forum comments, code, videos, pictures, entire websites.

I honestly think this is how we reach this "dead internet" meme. We're probably less than a decade away from this.

EDIT: here's my bold prediction. In the next 10 years, a very successful content aggregator (imagine Reddit, Instagram or Tiktok), where content is exclusively and openly AI generated, will reach the Alexa top 100 ranking.

We will have HN users saying we're too old and we just don't get why it is more fun to mindlessly get your dopamine fix from scrolling and liking AI generated images.


Arguably, we're already there. Most of the Internet is content marketing, and that content isn't meaningfully different than AI output. That it's generated by protein bots instead of silicon ones is just implementation detail.


> Most of the Internet is content marketing, and that content isn't meaningfully different than AI output.

I argue that it is. Human content farms don't create long-form articles, artwork or deep fake videos by the truckload, in minutes, from just a little prompt. And humans are relatively expensive.

AI generated spam will be orders of magnitude cheaper, faster and thus incredibly more common.


Loaded 16 terabytes of AI code, and the scrum master said “Well, bless my soul”


I'm terrified at the thought of all this hollow, low-on-details content that's about to hit the internet. Interestingly, if this degrades Google's search result quality, it could be an opportunity for someone else to create a new search engine.


> I'm terrified at the thought of all this hollow, low-on-details content that's about to hit the internet.

Honestly it feels like "hollow, low-on-details content" hit the internet at least 5 years ago. Developments in indexing and searching have incentivized writing of low-quality content that makes it to the top of results, and here we are


Honestly, no reason why we will have the same search experience as we do today in 5 years.

I expect search engines to be more like an AI assistant.

I don’t think Google can pull that off. Not because they don’t have the expertise for it (they do) but because their organisation is built around monetizing ads and an AI assistant might not work with it.


The past 5 years Google went downhill. If they lose the ability to search, users and ad revenue follow.


I think it will go the opposite way. Spend some time physically inside your local library.

All the search results are from experts, and have been proofread and stuff. None are ad-supported, and very few are paid-placement.


> I expect search engines to be more like an AI assistant.

> I don’t think Google can pull that off.

Does Google Assistant not count? Searching for things with the assistant is certainly one of its main features.


I'm thinking more like some of the demos I've seen of Adept AI, where you type in (or speak) a complex command and it does the job for you. Like telling the assistant to "find flights to New York between 15th and 18th December that are under $250".

Regardless, if the future of search is less a list of pages and more a set of focused, customized results, I can't see how Google will work in that context. Google, so far, has shown zero ability to monetize anything that can't be papered over with ads.


Already buulding one. Assuming content like this will be ad-supported (what is the other way to monetize it? nobody would pay to read it) Kagi is already successful at downranking sites with a lot of ads.


What will it search for?


Why is this spam?


If your site has a lot of fresh, unique content on a topic you appear to be an authority about related search terms and your search engine rank goes up. So pair a bunch of AI generated pages with ads or e-commerce and you’ve got a spam-driven revenue model and we have a worse internet full of regurgitated AI content.


I think their point is that if both real humans and spambots are using AI to generate their articles, then detecting which one is spam becomes that much more difficult.

This is where I see natural language AI causing a real kerfuffle. We're probably already reading bot spam articles without realizing.


If you are reading about the results of sporting events, you've been reading bot spam articles for several years, at least.


The local newspaper where I lived until this summer used software called “Lede AI Sports Desk” to automatically generate high school sports articles with headlines like "High School A edges High School B in snug affair 54-52" and "School C drums School D in sound fashion 48-4".

This is clearly not actually using any process that could be called "AI" with a straight face since it's obviously just playing madlibs with different sets of vocabulary based on the scores and margins, but the results were invariably stupid and repetitive headlines and articles.


To be fair, the special case of the Turing test where participants only discuss recent sporting events seems substantially simpler than the general one. I'm not totally convinced you need GPT-3 level capabilities.


that is fair; my only real complaint is that if the software were just maybe 5% more clever, the headlines it generates probably wouldn't look as bad as they do.


Using facebook's latest https://galactica.org/ i can get wiki pages that are pretty informative summaries of topics that don't yet exist on the net (though not very good yet).

A lot of the blog posts i read here are manually generated blogspam. These timelines these will converge, so i don't see a huge difference .


> both real humans and spambots are using AI to generate their articles

Are you suggesting that humans do not create spam too?


AI is a massive force multiplier.


Of course they do, they just don't have the same throughput as a bot network.


> both real humans and spambots are using AI to generate their articles

But humans create spam too!


It isn't. But tools like these make actual spam, and pseudo-spam (e.g. SEO content) easier and quicker to produce.


I was a user of Notion for a while but ended up moving to Obsidian, which solves a similar set of problems without the bugs. I’m seeing in this announcement that rather than focus on making a better product or fixing things that are broken, they’re jumping into the AI hype train. Seems like I made the right decision to switch to Obsidian, and over the last year have really enjoyed it.


Obsidian is basically perfect. Huge fan.


Basicaly perfect unless you want a special character in the title of the note.


That’s because your data is stored locally, so special characters in the title = special characters in the filename, which isn’t possible. Small but very worthwhile tradeoff.


Sure it is, what is this Windows? Put emojis in your file names, it’s fun.


How is this a small tradeoff? You don't use punctuation signs when writing(even if it's just a title)?


The file title is the file name so what ever's invalid for a file name is invalid as a title.

But there are some work arounds: - Disabling inline title and setting the document title yourself with what ever characters - Using aliases (in the YAML front matter) so you can search a document with it's special character title, and auto-set a link's custom text


Isn't Notion supposed to be an internal workspace for your company?

I understand the idea of cynical marketers wanting to use GPT-3 to churn out keyword nonsense to game Google Search.

But why would you want your employees reading/writing AI generated nonsense in your company's workspace?

In any case, since GPT-3 is open to everyone now, I think if you're building an AI copywriting tool (eg. Jasper and the 1,000+ others) you're going to have a rough go of it. It's now a commodity feature, not a standalone product.

As evidenced by this move from Notion, the GPT-3 API is just going to be bolted on as a feature in whatever tool you're already using.


The first use case in the demo is generating a blog post, so I’d assume that they’re trying to win the “cynical marketers wanting to use GPT-3 to churn out keyword nonsense to game Google Search” market. I know of some marketing teams that use notion for content management.


Congrats on launching the feature, which I will never use. If someone from Notion is reading this, I humbly suggest a few other features that might actually make an impact.

- Offline mode. My data is mine, my tables, templates, notes, have no reason to live in your server.

- Improve performance. Notion is getting slower with every release.

- Full Text Workspace Search.

This are features customers actually want. Show some customer obsession and cut the crap with that GPT-3 non-sense. You are helping spammers to automate their workflow, on creating irrelevant posts that match search keys.We already have copy.ai and a ton of other services that do just that.


I am a raving Notion fan. It's one of the few products that I just adore.

Maybe my tolerance level is high, or maybe my hardware is good, but actual performance is acceptable.

I do hope that Notion does improve offline mode. A single page works that is already loaded works wonderfully. Syncing has been non-problematic for me. I wonder why my phone and mac can't cache all the pages indefinitely, which would allow me to work completely offline.

Also, search needs work. Both search from the "search button" and the search that pops up when I type "@bla" are becoming less useful with each additional page in my workspace.

Finally, where are tags? Please please please add the ability to add ad-hoc tags to each block and page, the same way I can add comments to each block and page. And then give me the ability to view all tagged items as a kind of virtual page, or a type of database.

I'm at a point where I type things like tagIdeasToTry in the actual text and then use search to find all blocks that have that tag. But that requires mental effort, and fighting with the search functionality to get back results. I've even considered @mentioning special pages and then using backlinks as a way of categorizing content, but that only works at page level, and is clunky.

Notion, please help an addict get his fix.


You can use tags on pages that are listed in a database, right?

I’m not aware of any block level tag though.


Tag-like constructs do exist in databases. Select and multi-select work like simple effective tags within a database. They work nicely.

You can't use tags on pages that are not in a database.

Block-level tags would be a godsend.


>You are helping spammers to automate their workflow

Exactly, I can't remember a single time when AI-generated spam articles that you find on google search frontpage was useful


> - Offline mode. My data is mine, my tables, templates, notes, have no reason to live in your server.

I couldn't agree more than I already do.

Without this "feature" I will never use Notion. Offline mode is a minimum for me to even look at such an information handling product.


"offline mode" is the reason i prefer obsidian to notion for personal note taking.


Offline mode would impact the most users, but they are purposefully avoiding this feature because it's not sexy. As a paid Notion user, it's infuriating that they spent engineering time on this stupid AI non-sense in front of things that paying users have actually been asking for.


Yes on all three points! Especially search. Notion’s search is so painfully slow. Whenever I try to link a row to another row the loader sounds for 2+ secs, even if I’m searching a table with <10 rows.


If you actual read what the AI prints when it's asked to create a blog post. It repeats the same thing over and over again.

You can sum up the whole blog post in one sentence.

Notion AI helps you be more productive by understanding your work habits and providing you with suggestions on how to improve them.

It writes like how I did when I was a kid in elementary school writing English papers lol.


But it understands your work habits! Also, it uses machine learning to understand your work habits! This makes you more productive by understanding your work habits.


Yeah I thought at first the video was skipping or rewinding or something when I tried to play-pause-play my way though it as it kept repeating essentially the same thing over and over.

Sounds like some next-gen Lorem Ipsum.


Maybe they should focus on fixing bugs. The editor has significant problems. It happened to me that I literally couldn't copy stuff. I had to download as txt and then copy. Or sometimes you're trying to select a sentence and it selects the whole block, and all these annoying little quirks that I find incredibly enraging.


This is essentially innovation theater for a fairly lackluster core product. Adding AI for a feature no asked for is like straight piping your Honda Civic to make it sound like a race car. At the end of the day you're just driving a loud Honda Civic.


You're onto something with "innovation theater". Companies that overhired need to occupy all their new hands. Well-defined initiatives are already staffed so the extra's are thrown into big, ambiguous initiatives that executives don't pay attention to but have hope for. No attention means no career growth, so the overhired PM's/engineers/whoever put together a performance to steal the spotlight.


If you're having AI write your posts and coming up with your ideas, you're either a spam site or a failing business.


On iOS this link opens up the Notion app which says “this content does not exist”. Removing the app was a quick workaround.


> On iOS this link opens up the Notion app which says “this content does not exist”.

Here I thought that was a meta comment on the notion of AI.


Now I feel like feeding "this content does not exist" to Stable Diffusion in true Magritte fashion


Honestly the web link to iOS app connectivity across the board seems completely broken these days. I always just end up in the home screen of the app in deep linking to.


Opening in private tab also works (at least in Firefox)


AI is the new 3D printing. Cool novelty that will undoubtedly change the world somehow, but I hope we avoid letting it become this creation that is played out within 5years, overapplied, and shoved into any process whether it improves the product or not because its "the future". I'm about two AI support bots from snapping.


I know this wasn’t your point, but I’m a 3D printing enthusiast and I follow what’s going on in the industry.

I wouldn’t consider 3D printing played out, but I would say that it’s become more boring in the last few years. This is a good thing because it means that it’s being sustained by real work (e.g. medical, prototyping, micro-manufacturing) rather than hype.

I do agree that there was a hype phase, though, and the AI hype phase will probably be similar if not bigger.


Where do you see AI support bots? Don't think I ran into those yet.


Every chatbot on a webpage or in any app.


Those are not AI, those are MCQ. I think I've seen just one that allowed you to have a "real" convo.


I've seen the benefit of AI for completing sentences, or lines of code, and I can imagine the benefit to creative writing where things do not have to be factually correct, however writing in Notion is typically writing about something in the real world.

The problem with AI text generation is that it does not have that real world context. A great example is the first example on this page: writing a blog post announcing Notion's AI generation.

The AI text generator does not know the feature set, and therefore either the content is going to lack any detail about the product, or it's going to have incorrect detail, in both cases providing no value.

What are use-cases for this that people will actually benefit from? Creative writing - sure. Brainstorming? Maybe for very vague and basic topics that don't need any business context. But more than that?


> What are use-cases for this that people will actually benefit from? Creative writing - sure. Brainstorming? Maybe for very vague and basic topics that don't need any business context. But more than that?

Looking at my Linkedin feed: lots of lazy people excited about possibility of "creating content" like "100 blog posts on a given topic in a minute". So, in short: seo spam. Lots of seo spam.


Look into Jasper, the company was founded 18 months ago and just raised a 125M series A. The more impressive bit is they have 80k paying customers (their cheapest plan is $30 a month).

Who uses it? Mostly content marketers. GPT-3 is actually quite handy when writing because you can give it a rough outline of something and it can do a pretty good job making it sound good.

As for the lack of context, there's no fundamental reason you can't give these tools information to work off. In fact most of them ask you to.

I'm quite bullish on AI assisted writing - it feels like the internet in 94. For now no one understands it, but the fundamental value add is enormous.


It seems to me that the main benefit is the ability to give a prompt, and be given a starting point.

An h1 (or h2 I suppose) on the site is "Let Notion AI handle the first draft."

That makes sense! That's a lot easier than starting with a blank page.


I could see a use for having the AI describe the generalities (some of which could be provided in the prompt) and then have a human go in and fill in specifics or correct inaccuracies. Either way it amounts to a laborsaving tool, it's much easier to read than it is to write.


> it's much easier to read than it is to write

I see your point, although I'm not certain I agree. It's very easy to read what we hope to see, especially when we know the topic, rather than what the text actually says. This is why proof reading is so hard for the author to do. Copilot has a similar issue, pushing more effort from code-authoring to code-review, which I'd suggest humans are probably worse at.


Yeah, agreed, I would say it's much easier to "rewrite" than to write.

Even if the AI is writing nonsense, you still benefit by virtue of Cunningham's Law.


I'll use github copilot as an example, it pulls context from your surrounding code files and code in the same file. Something like that gives it the context your thinking of and a similar technique may be able to solve that problem here.


I think there's a big difference between matching up variable names and the context of what you're working on in a file, and saying "brainstorm 5 marketing strategies for me" like in the examples here. With code, most of the context Copilot needs to do a good job is a) written down and b) parseable in very well defined formats. With business contexts most of the context won't be written down, or won't be in a structure that is useful, or won't be parseable.

Let's take the example of the marketing brainstorming. Let's say you're a super organised small business and you write down a report on every marketing experiment you try, and influencer marketing didn't work for the product, so you write a report on why TikTok creators are not a good direction for marketing.

To know not to suggest influencer marketing, Notion needs to: have all your documents, know that document is relevant, have the "idea" (let's brush past the fact this was essentially written on someone else's brainstorming list and learnt from), know that the idea is related to that other document(s), know that the conclusion was negative sentiment in the way that matters, and then know that because it's doing brainstorming, lists of ideas should probably be filtered by things that haven't worked in the past.

Maybe it's not the worst thing if it generates "influencer marketing", but if you ask it for 5 and it always gives you the 5 you've done before, then you start having to engineer your way around its stupidity – give me 10 ideas, or give me 5 ideas that don't include xyz. Is it providing value over, say, googling "small business marketing ideas" and reading an article full of ads titled "Ten things that will transform your small business marketing". That's low value content, but it seems most likely to me that this will produce largely similar results.


First off sorry totally missed you had replied and you took time to write a thoughtful response. I think your skepticism is completely warranted, but it sounds like a space where it’s worth a shot. If I’m wrong no biggy this is low risk, but if I’m not that’s a lot of upside.

I’ve seen copilot do some truly novel code synthesis, more than just matching up names. I’ve also seen GPT 3 generate compelling marketing copy and many other things so I’m admittedly biased in favor of potential solutions like this.


This is how Skynet takes over, isn’t it? We thought it was going to be fighter planes and attack satellites. Giving the AI overlords the ability to subtly seed ideas into planning docs the world over locks in a more bureaucratic apocalypse. I need to brush off my screenwriting software and ask the AI to write a script with the prompt “Office Space meets Terminator.”


Does this mean they are training a model on their users' notes? Or does it all come from a separate model e.g. GPT3?


Yes, they imply they will.

Notion may, however, use your Content or Customer Data to improve and train Notion’s own models.

https://www.notion.so/Notion-AI-Program-Terms-c0066e30039041...


Under the "Improvement of services" section, it says:

"We do not and will not use your Content or Customer Data to improve or train our models unless you give us express permission to do so."

Maybe they changed this in the past 7 hours?


Is opting-in to the Notion AI beta not "express permission"?


Yup just took a look and they changed the text


I will ask support and report back. Having my private data fed back to other users via "AI" is a deal-breaker for me and my organization.


Ok this took many back-and-forth emails, but I eventually got this from support:

> Marvin from Product Operations here and happy to clarify. The answer you're looking for is: no we are not using your data to train our models.


I am wondering the same thing. From their website: "Your data is encrypted at rest and in transit." (see: https://www.notion.so/help/security-and-privacy). I am not really sure what that means.


Decoding for all "we take your security seriously" pages on the web:

In transit means https (like your comment getting sent to this page).

At rest means the underlying store uses a key, like S3, or even HDD encryption. It's encrypted on write to the disk, and decrypted on read. This may be fantastic for when nation state black ops sneak into the data center and remove the server's drives for forensic analysis, useless for anything else. (To be fair, it's useful on a laptop. Not so useful in a data center.)

Neither of these means it's encrypted between reading disk and sending over HTTPS.

Neither one means employees, servers, or clients, can't see your data.


Thanks for the clear and detailed response! It's what I feared it meant. It's a pity they need to use such deceptive language, I find.

I really liked notion as a product. This is kind of a wakeup call for me that using the cloud pretty much means giving up control - and not just theoretically.


Even if they used GPT-3 they could have easily fine-tuned on their data[0].

[0] https://beta.openai.com/docs/guides/fine-tuning


Is "their data" my data?


Looking at their privacy policy page [0], they mention that customer data is used for "Developing new products and services and improving the Services". So by using the service you give them rights to fine tune on your data.

[0] https://www.notion.so/Privacy-Policy-3468d120cf614d4c9014c09...


Writing is one field that I think won't be able to achieve the 'human' touch anytime soon. Art? Sure, as we have already seen with DALL-E and the like. Music? Doubtful (look up Kandrake-Z for an example), but not sure if it will expand beyond plunderphonics and noise. But writing? Meh.

I don't know if anyone even uses content-generating AIs like this for writing - but I'd be glad to change my mind by seeing some hard numbers. I mean, except for incomprehensible content-generated articles that pop up on searches only because of PageRank's flaws.


“AI” generates bulk structure, humans adjust “at the edge”.

Everyone is building Mechanical Turks to nudge users to add artistic nuance AI lacks so far.

Personally I prefer using the libraries directly rather than gift my work to startups I’m also paying subscription fees to.

Startups are basically double dipping; let us train based on your behavior! …for $9.99/mo per user. Oh and we own the outcome!

Hoping to get some useful open models and data sets for game world generation soon!


Even their examples seem better off as just templates of different speech patterns. For example Minto pyramid is used, https://untools.co/


Thought this is an April Fool's joke kind of thing, but apparently they're serious?

That's said, the spell checker is nice and the translation would be helpful too.


Wouldn't a more helpful product direction for this be to use AI to make notion pages shorter? this seems like bloat for bloat's sake.


As others have said, they should focus on improving their product by fixing annoying bug and adding the features people are actually waiting for instead of more impressive but way less useful stuff like this. I left my own notion in limbo about a year ago because supposedly Google Calendar integration was just about to roll out, and that was the killer feature that would make my workflow really useful. Then nothing happened, and I barely use it because it's still "unfinished". Any recommendation for something similar but a little more powerful and hackable for a developer oriented workflow ?


having the notion app on my phone means that clicking the link opens the app and gives content not found.


Classic issue with Universal Links, assuming you're on iOS. I implemented Universal Links for an app a few years ago. I think many developers think it's trivial to implement, follow the one-page guide to setting it up, and think the feature is complete. In reality it's much more complex to do for an app of any real size.


They probably need a wildcard in their routing to open a webview for stuff like this they can’t handle in the app. But yes universal links are tough to get right. Easy initial implementation and then surprise bugs forevermore :)


How long until AI generated content becomes a political agenda? Not sure why this soulless stuff is being embraced by everyone but I see it as the nuclear spam bomb for the entire Web.


This looks like the folks at Notion had nothing better to do, frankly.

Poems? Really?


Highly skeptical of AI applications like this, unless you have a personal active learning model that you can refine over time. Otherwise, it can only really answer broad canned questions like "10 ways async communication is more productive". As the other commenter pointed out, seems like it would be way more useful to use AI to summarize existing content and suggest links than to pretend like it can do the work of an employee with zero context.


I work for a company in the blogging space (not AI related), and we are seeing huge uptake on AI content generation. Something tipped in the last month or so.

For most bloggers, the hardest part of a creating a piece of content is starting it - staring at the blank screen. At best, they are using AI to jumpstart themselves, then they can flesh it out, rewrite, etc. Human nature being what it is though…

Overall, I think it is effective enough to be a complete disaster for the internet.


I'm skeptical too, but at the very least this gets buzz with somewhat limited development and maintenance effort. Summarizing your content and building personal models is expensive, hard to debug and support, and it opens the possibility of attacks from competitors who "don't process your data" or "build models off customer data" (even if properly secured/isolated.)


Who's bearish on AI used like this? Seems like everyone is packaging the cutting edge work/research and putting into this same "bucket".

Will it go the way of 3D tvs? Interesting for 2 years then gone? We'll see.

Not to mention these companies are 100% training their "bot" with your data. Notion will learn from user's notes. Replit will learn from their new "bounty marketplace" they launched. Github from the repositories. Etc.


It’s a spam machine. That’s all it is. That’s all it will ever be.

I get why it’s fun to make a system like this, but does it really never occur to anyone that no human wants to read AI generated copy?


This makes me think twice about using them to store my data.


That page is now missing. Maybe they’ve changed their mind. Generative AI is one of the few sectors that gets VCs still excited these days, but the text version of it is fraught with unresolved problems. Try getting rare code or the inauguration poem in it for completion and see what comes out. Spoiler: it’s either direct copy or gibberish (content that any good editor would remove) that wastes an ever increasing chunk of humanity’s time. The avid users of generative text today are LAZY and don’t clean up or improve the content - they are compensated and rewarded for word counts and that’s exactly what they get in droves from GPT3. If Notion implements generative features, it needs to label those notes generated, to prevent its systems from filling with crud nobody cares to read over time. If they do content summaries, will they stop with that.

That said, generative AI is a feature, that will go everywhere soon - it will be the next copy paste. If developers and product people are not careful, it will also invoke regulation that demands generated content to be labeled to remain legal (or you will loose the reader’s trust in your content).


> That page is now missing.

Did it open in your Notion client when you clicked on it? Paste the url manually to your browser and it opens.


The article zips by kind of quickly, so I transcribed it:

> Introducing Notion Al

> Notion Al: Your New Digital Assistant

> Introduction

> Notion Al is a new artificial intelligence feature from the productivity app Notion. Notion Al is designed to help you be more productive by understanding your work habits and providing suggestions on how to improve them. In this blog post, we'll introduce you to Notion Al and show you how it can help you get more out of your work.

> What is Notion Al?

> Notion Al is an artificial intelligence feature that is designed to help you be more productive. Notion Al uses machine learning to understand your work habits and provides suggestions on how to improve them. For example, if you tend to procrastinate on certain tasks, Notion Al will suggest ways to overcome that behavior

----

> Promotion Ideas

- Make a video showing all the features of Notion Al

- Create a blog post highlighting the benefits of using Notion Al

- Give a presentation to potential users about the advantages of Notion Al

- Reach out to social media influencers and ask them to try out Notion Al

- Give free trials to users who are interested in trying out the software


To anyone who is frustrated with Notion's official client:

Notion recently opened up their entire API correct? This would leave a market for 3rd party clients that are snappier and innovation without having to leave the notion platform as a whole correct?

Full disclosure: I'd like to gauge interest in a product like this as I'm interested in working on solving it.


Real-time collaboration via the public API is challenging. (I work at Notion)


Do you have any insight into what the hell they're thinking by putting out this feature before addressing basic stuff like offline mode? This is literally insulting to paid users, pumping out stuff that no one asked for and no one wants while basic feature parity with other products continues to suffer.


“Offline mode” is a bit of an ambiguous term. We regularly ship behind-the-scenes improvements to data synchronization and caching systems. These systems are much more reliable today compared to a year ago. However, we don’t have a way to explicitly manage the local caching policy.

Are there other collaborative wiki products that implement your ideal “offline mode”?


At this point just the ability to cache everything locally for read-only access while offline would be a huge improvement.

And ideally, feature parity with Apple Notes, Google Keep, Evernote would be nice. But I would be thrilled just to get offline read-only access to everything.


any other notable challenges, and for those challenges are there any workarounds? Additionally, any timeline on improvements to those public APIs?

and am I correct to assume the public API is the same API for functionality that notion itself uses or are there unexposed features that only the first party client gets?


The public API is substantially different from the internal API so we can provide a better developer experience with more understandable and consistent interfaces that match product concepts, rather than exposing incidental implementation details. We only build public APIs needed to enable a set of use-cases; right now "develop an alternative editor" is not one of the use-cases we intend to support - so there's no active work toward things like realtime collaboration APIs. If your client decides to use the internal APIs, it may be blocked as part of our anti-abuse countermeasures.


Lots of people don't care about real time collaboration


Having watched the video, the kind of content Notion AI generates is the kind I typically stop reading after a few sentences. By then it becomes clear that the article has nothing to say, was written by a non-expert, and is really nothing more than filler. It can't help me solve a problem and won't offer any new insights.

That said, it wasn't too long ago that the writing style itself was easy to identify as AI-generated, or even borderline grammatically incorrect.

Notion AI does a pretty good job of mimicking an HR-level understanding of a topic. And that's progress.

I just wonder what it will take to get to a point when the output is of the quality I'd associate with human subject-expert writers who know how to engage with a reader.


Generative AI, especially for writing, is really good at producing mediocre content.

I think people will quickly realize that poor AI-generated content is the new spam and we’ll see a surge of both algos and people prioritizing human-created content as a result.


Off topic: Any suggestions for a Notion-like knowledge management app with decent handwriting support except OneNote? I need handwriting in order to memorize and structured data (tables/boards/lists etc) in order to be efficient.


I have yet to see a text-suggestion tool that is as good as GitHub Copilot in terms of speed and quality. I’ve looked at Lex from Every, and Galactica , not impressed with either of them. I can almost get better results by just writing a comment section with Copilot.

https://lex.page/

Mem.ai recently raised $20M and they supposedly have AI-assisted writing but I haven’t tried it. But I think it’s really lame that even after all that money raised they still don’t support Math/Latex notation.


Imagine if the training set includes personal data similar to copilot and particular licensing... that would be horrendous. Notion should at least disclose how the training set is created.


That was my first thought too.


This is pretty cool!

I started working on GPT3 integration for Notion [0]; but I guess that was obvious to have AI integrated in Notion.

The question is when will we get AI generated images in Notion? Well, then I can shut down my side project [1]

[0] https://twitter.com/kiru_io/status/1572290853103865856

[1] https://slashdreamer.com/


Sad to know that instead of a proper API or offline support the notion team has decided to spent their time on AI.

Notion turned 8 this year and it still doesn't have the aforementioned offline support, no repeating dates and events and no plugin support.

Why companies prioritise a product nobody asked for instead of features which a large majority of users are vocal in support of is beyond me. I can only hope a competitor will force them to shift their focus.


Notion is a VC backed company.

Less snidely, it's probably a lot more interesting to poke at AI integration than building another REST API. I bet there are more programmers putting their hand up to "explore that area" at the stand ups.


Since when do programmers get to work on stuff they want to work on?


I doubt they walk into work every morning and pick from a big board on the wall, but if you don’t have some influence over the kind of projects you’re working on, you should get a new job.


Since forever , every place I've worked has had insane amounts of autonomy for programmers


Well, looks like that is about to change:

https://news.ycombinator.com/item?id=33620662


Yeah we've lived in a good job market. Can't assume it'll last forever but giving programmers some autonomy is likely the best way to move a software company forward so I wouldn't be surprised if that continues


> Less snidely, it's probably a lot more interesting to poke at AI integration than building another REST API

People love to work for developer focused companies with a lot of freedom, but in all honesty most paying users do not care in the slightest what you are interested to put your hand up for.


Because AI generates attention and therefore new users/subscribers. Stuff existing subscribers want is a lower priority. Also, the developers probably think AI is cool to work on and offline support isn't.


This will be a problem until user retention grows more important than gaining new subscribers, even if they only stay for a month.


Agreed, their search (literal string matching only) was outdated long before they were ever a company, but they do this instead of sticking one mid to senior level DS on that?

This is also almost certain to lead to more documents, not better ones, making the abysmal search experience worse.


Unfortunately every product focused AI/ML/DS team out there is currently churning out these thin wrappers over GPT3, I know because my PM can't stop pushing this nonsense.

The issue is that because a lot of non-technical people haven't played with GPT3 too much, there is huge initial engagement with these "features" so the stats around usage look great, even though anyone with a few brain cells can identify that the root cause of this is the novelty effect.

PMs also get the mistaken belief that, because writing is hard for most people, it must be a big blocker for these marketers. But for a marketer writing copy is like writing code for a seasoned dev. Similarly copilot is fun to play with, but I don't know any developers out there who seriously use it (save your replies I know some enthusiasts must exist). Additionally, unlike copilot, these GPT3 copy outputs always only look decent at first glance, but contains enough oddities that they are rarely if ever useful in their displayed form.

Expect floundering "AI" teams to churn this for out en mass at any remotely content related SaaS company for a while.


(I work on Notion)

As of November 8, Notion supports recurring events in databases: https://www.notion.so/releases/2022-11-08

Curious what you think is missing from the API? https://developers.notion.com/reference/intro


Seriously, they're ignoring the thing that they do well.


Because Notion AI is not built for developers ?


Then who is it built for, exactly?


I use large language model for work and use Notion daily.

while I like the "AI" part (the large language model), think it would be more interesting and productive to use same backend for full text semantic search & question answering or summarizations.

But it is cool to see Notion trying this way, kinda curious to see the results when so many people have access to this type of generative model.


Tapping this link on my mobile phone automatically opened the Notion app which led to a "this content does not exist" error.

Maybe the AI can fix it.


uninstalling the app seems to make the link work properly


Notion is painfully slow - if I use a notes feature I need it to load instantly. I find obsidian to be much better for this reason.


You can do AI on Obsidian with the GPT-3 AI writer.


Thanks do you have a link to a guide or video or webpage where I can check this out?



I'm very wary of all these companies building their offerings on GPT-3. It's dependent on OpenAI until it's cheap enough to host your own instance, and then they wouldn't have a proper service to offer anyway.

Also, GPT-3 completion is pretty hacky, and anyone who needs it on an information organization pipeline is not taking it seriously anyway.


This is cool. Not sure about the privacy implications of this. Nowadays anything with AI after it makes me think twice before using it. Especially if it's as critical & personal as a notes app. Who knows in what ways they'll be using the collected data (it's obvious that they will collect data to improve the AI).


Is this trained or fine-tuned on the content of existing Notion customers like GitHub Copilot?

The FAQ is unclear:

> Any information used to power Notion AI will be shared with our partners for the sole purpose of providing you with the Notion AI features. We do not allow any partners or 3rd parties to use your data for training their models, or any other purpose.


The privacy policy / terms is clear that while they don't share data with third parties, they help themselves to the data for their own models. Misleading of them to leave that off in their FAQ. And it's opt-out, not opt-in.


This is awesome! i was searching for a writing tool integrated into Notion just yesterday

Together with Notaku [0] i can now create blog posts and help articles much faster, choosing Notion for my websites CMS was the right idea

[0] https://notaku.so


I will join others to conclude that, being VC backed it what makes you do things like this and is a signal that company will work on more shallow features and focus on growth for growth sake.

Notion is solid app and idea and it would be a shame to became just one more thing that is just noise.


The sample generated blog post for Notion AI scrolls by pretty quickly, but appears to be (obviously) completely unrelated to what Notion AI actually is. How is this any more helpful than just having a template? The language model has no clue what your new product is.


Not sure what Notion is using under the hood, but it makes me think about how good GPT-3 is... What will it be like in 10 or 20 years when this tech is dramatically better and is commoditized, available everywhere? It's hard to grasp.


Seems like a pointless toy feature that will be left to rot when people realize that making it useful is actually really hard.

There have been some cool developments in AI recently, but it's just not free magic, and it might never be.


I can't wait for this trend of companies forcing AI "features" into their systems to be over. I get why they're doing it though, It's just such a mundane and safe way to market your product.


Imagine being a Notion user that’s been wishing for sensible things like offline mode or performance improvements on large databases/docs and finding out they’ve spent their time and money on this instead.


This is why I love Notesnook and what they are doing with their app. Focusing on privacy, performance [0] [1] and their users. I am still amazed by the speed with which they work on different things. Definitely worth following.

[0] https://twitter.com/notesnook/status/1591048432931975168 [1] https://twitter.com/notesnook/status/1588602471290572800


What's your connection? Most of your posts and comments are about Notesnook


God help us. The quality of content on the web is about to get even lower.


I wonder, if it can be generated by an AI, maybe it shouldn't be written down. Current AI can only generate what most people already know, and the purpose of writing is to express something new.


I’m curious to know what the profitability of a feature like this is?

Are generative text models and image models cheap enough for consumer products or do you need to charge a premium?


Definitely locks the user in, pushes engagement up, probably pushes the creation of new "blocks" up, etc. Defends against google docs doing it to them?


Google already has AI based autocomplete in things like Gmail and docs. It's not very useful of course but technically others are catching up here.


My sincere fuck you goes to all the miserable transhumanists who give up all their human essence, including creativity, to their corporate-controlled AI overlords.


The intro video definitely reads like satire. This is either a very early April Fool's joke, or just a silly toy that Notion is marketing for some reason.


Got a FB ad yesterday inviting me to join xyz to get my very own AI friend to interact with, for a monthly fee of course. With picture of alluring young lady.


w.t.f... fix the tables instead, let me copy a row without having to duplicate the table, turn into a database, copy the rows and paste in original table


in my humble opinion Time Magazine should make AI their annual personality of the year.

It will either be zelensky or putin but that would in my opinion be a mistake


It's funny because people can just use other AI tools that do the exact same thing already with out Notion AI. So many services do this.


I need a better full text workspace search instead !


So Notion learned how to use a GPT-3 API? Ok, cool.


Would've been interesting to have AI in music engraving software... but alas, this wasn't about Presonus Notion


The demo videos are going way too fast, it's almost impossible to read what they say and there is no pause button.


Interesting timing, mem just raised 23.5 mil at 110mil post money valuation from OpenAI for ML assisted notetaking.


I write my blog posts in VS Code and GitHub Copilot did the same thing. It completed many of my sentences...


Notion may, however, use your Content or Customer Data to improve and train Notion’s own models.

Nope, I'm out.


Im a long time Notion user... I wish they would focus on features that the users want.

How about custom colors ? :(


I'm starting to think more people take Roko's Basilisk seriously than I realised.


Please stop calling this AI.


why?


Because it's not artificial intelligence by definition.


Good name but feels mediocre to use. Hard to use it intentionally.


Horrid, horrid, horrid


GitHub Copilot moment


Great. A noise machine that will make people feel "productive" while producing little actual value.


"Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."

https://news.ycombinator.com/newsguidelines.html


Fair enough.

Based on the two examples I looked at on the posted page, one was a brainstorm ideas list for how to promote Notion AI which was fairly obvious. It did not include anything novel, just literally spit out what a quick Google search for “ideas to promote a new product” would have. I suppose you could say that it saves you the search but if you aren’t someone who understands promotion inherently then you’d need more info in the first place and if you are then this list is unnecessary as you’ll already have a promotion strategy in mind.

The second example was the AI writing a blog post introducing Notion AI. This is what I really referred to as a noise machine. The reason you need a blog post like that in 2022 is not for people to read but for Google’s indexer to read so that it gives your page and site all the points and you get better SEO. In other words the whole concept of this has gone off the rails where we have computers writing articles for other computers to read, which will ultimately be fed to another “AI” to learn from to write better articles for computers to read. Why are we perpetuating this circle jerk? Wouldn’t it be better to fix some core issues?

Lastly, I think the tech that allows you to turn a sentence or two into a coherent article is quite exciting. But I think this is somewhat the wrong application for it.


Much better comment—thanks! One for the 'rebound' list - https://news.ycombinator.com/item?id=23507259.


A great tool for the overemployed so they “write” blog posts as a side hustle.


Honestly even the examples on the promotional page are quite bad.

I’m a big believer in AI as an accelerant for working, especially after playing with Stable Diffusion. But these examples are like almost wasting my time not making me work faster.


Like all office suites


plagiarism on steroids


invite code for lex.page anyone ?


With all of the exciting directions knowledge management has been moving in lately, it's disheartening to see that Notion's product vision is already so uninspired.

Particular in the corporate space, the problem of finding relevant knowledge (and keeping that knowledge up to date) is a really hard problem. I think competitors like Mem are going to eat Notion's lunch here (categorization and perhaps tagging of stale content is a much more reasonable application of ML than this).

... further, Notion's performance is so absolutely awful that my very small company had to stop using it. Latency is incredibly relevant to note taking and writing apps.

The marketing page example is also so contrived: "Write a blog post introducing Notion's new AI feature." How would the AI even know what that feature even is? Where's the context? It seems like this just proposes static solutions to dynamic problems.


Signal boosting in case anyone from notion is reading: please address performance, it is practically the only thing I care about. I love the product but it is almost unusably slow with my tiny amount of data. Cache more stuff on my device, avoid blocking network calls where possible, etc, please!


Yes, performance is absolutely awful. Not just individual (it’s a heavy react app), but it’s not really possible to use search when you have more than a couple hundred pages. I swear their test fixture is a site with no more than 5 pages and they all use Mac Pros so they don’t notice any performance issues.


I have a single markdown file that stores all my work notes for the year, and recently I imported it into Notion. My partner raves about how much she likes Notion so I thought I'd try it.

The files can take up to three seconds to load at times, and seemingly caches nothing on my phone so when I want to look at my notes on the tube, I can't load it.


Yes, I’ve been using Notion for years, but recently started migrating everything to Obsidian because it feels so much faster.


(Notion engineer here) I'd love to hear more! We're focusing a lot on performance and want to hear where it hurts the most. If you (or anyone else reading) are comfortable sharing, drop me a line at carlo AT makenotion DOT com


Agreed. I use a few filters which take seconds to update, sometimes disappearing and then redrawing on any change in the tables. And recently some of my filters and views have been changing, leading me to have to tweak and adjust some of my dashboard views, shaking my confidence.

The filters and relations are so powerful and I really love the flexibility: example, “show only tasks linked to projects with an active status.” But when they keep disappearing and redrawing, I just feel slowed down in the app lately.


seems like they wanted to launch an "AI" feature as fast as possible, so they took the easiest route which is basically just creating a UI wrapper around GPT-3 for generic text generation

that's what it looks like to me at least, I don't see anything here that 50 other generic GPT-3 tools don't already do. It's just embedded in Notion


I think they would consider it a success if the existence of that tool drives people to spend more time in Notion

Doesn't have to be revolutionary. It helps to be "in just 1 place"


true, plus I imagine it gives them potential for some usage based pricing to increase revenue from power users


I like to call this "AI (or ML) as a blunt instrument". You just plug in the model, map inputs to outputs in the most straightforward way.

In some cases that can work, but generally you want to massage the inputs and outputs a lot more than that.


Agreed about the performance. I even think this feature is cool (I don't use Copilot to write code, but happily pay the $120/year just for the time saved writing docs (the kind colocated with code, like rustdoc, JSDoc, etc) and unit tests, so I'd like a trained model fuzzy logic bot helping me with notes/docs too).

But there is no feature at all that could excite me about Notion now unless they fixed the performance first.

I stopped using it due to performance also.


> Particular in the corporate space, the problem of finding relevant knowledge

This really is incredibly hard. A large fortune 500 company I worked for this summer had a whopping 35000 confluence pages. There is no navigating a knowledge base that vast. You couldn't even clean it up, because there was no way to know if someone would need some piece of fringe information on a rarely visited page any time in the future.


Wonderful if everything is in Confluence though. Most companies have knowledge in various forms free floating through e-mail, sharepoint, sharefile, onedrive, teams, Onenote, wikis and so on.


I worked at a place that used confluence as a knowledge base. There was nothing that a git repo of txt files and grep couldn't do better. The search function was useless on top of being incredibly slow. Maybe they were not using it correctly I don't know.


Git, text files, and grep are a horrible user experience for anyone who isn't already experienced with those tools.


That's what wikis are for but IMHO wikis are annoying to setup and customize, and good luck convincing people to document their knowledge

I wish we had a librarian on staff on my team to just keep data organized


This is… a really good idea? Do any tech companies do this? I’ve had tech writers on staff but never a trained librarian.


It's also a matter of vocabulary and writing about your knowledge. It's almost impossible to find somebody that solved the same problem you currently have, because somebody that solved it didn't write about it. We're kinda OK searching code nowadays, but we're not able to discover architectures or abstractions properly.


I'd bet my had that most of it was meeting notes


I desperately wanted to tackle knowledge retrieval problems when I worked at Slack, to the point where I built prototypes during nights/weekends to try and get internal traction. Slack easily has one of the most valuable and important datasets in the world for training corporate agents.

IMO, the most useful rejects were:

- Subscribe to a search query. Get an RSS-like digest of Slack messages, external document updates, CRM interactions, etc.

- Auto-responder I called "escal.ai," named after Slack's internal naming convention for help/support channels. These channels were prefixed with #escal-<subject>. I built a variation of https://arxiv.org/pdf/1901.05415.pdf to handle asymmetric retrieval tasks, where the answer to a question was typically a Slack post in another channel.


I think I find the examples they give actually quite worrying in an age where we are dealing with hard problems around moderation.

"10 ways async communication is more productive" - alright, so you have an AI answering a leading question. Are you going to take the output as given and publish it as-is or spend time verifying and tweaking it? If you do the latter, how much productivity have you gained from rewriting a generated post compared to writing your own draft?

For automatic summarising and translating, okay cool, that really could be handy but this is almost commodity tech at this point.

I don't like my negative, pessimistic response here but I don't need tools that make it easier to write or create, I need them to keep things organised and up to date as you've mentioned. I need a better search that can help me find something that already exists, or actually point me to that article if I'm writing something that starts to look like a duplicate.

Can the AI say "this 'setting up the environment' post looks really similar to [some other setting up the environment post] which was written 6 months ago; is that what you're looking for/do you want to edit that?"


"If you do the latter, how much productivity have you gained from rewriting a generated post compared to writing your own draft?"

In my experience, AI providing the activation energy to get a first draft is highly valuable.


Particular in the corporate space

This is EXACTLY what people in the corporate space need! AI to help them generate nonsense corporate-speak so they can get back to work!


I'm still surprised that no one has just mined all of a company's e-mails to make a database:

- strip off all attachments --- these are archived on a central server w/ version control and links are passed around

- all e-mails are converted into threaded discussions in an hierarchical organization

Then, use an expert system to review and tag all messages and facilitate finding the correct forum thread/file.


I was trying to fix the problem of getting relevant knowledge and staying up-to-date with my startup https://www.gainknowhow.com/ . It never took off organically and I only sold it to a couple companies. My core idea in GainKnowHow was to structure data as a graph of skills. I think the graph data structure to store knowledge is better than how a lot of tools store knowledge today. I still think it is a great way to store knowledge and use it myself to train people, but the graph data structure was a leap too far for many people who implement knowledge systems at companies.


Some time ago, a Notion engineer was telling HN that their old principle of hiring slowly was blocking growth. This was around the same time Notion started launching half-baked feature like https://www.notion.so/help/guides/creating-links-and-backlin... which, last time I checked, still doesn't have the smoothness of Roam or Obsidian.

My gut says that every engineer they hired in the name of growth is trying to get their own little feature in to show impact. They prematurely scaled the company and it shows in the loss of focus.


FWIW my small company has zero issues using Notion, what you're describing sounds like a situation specific to your use case, not a generic Notion issue.

But more fundamentally, it sounds like you just don't believe the product works, which is fair, just not a very helpful comment. "How would the AI even know what that feature even is?" by learning about how its documented in the rest of Notion.

Fundamentally though, your comment has serious "Uber? No way that's going to work! Why would I get into a stranger's car?" vibes.


>Particular in the corporate space, the problem of finding relevant knowledge

I'm eagerly waiting for the day that I don't get frustrated with Notion's search. I'm not sure if I actually even bother with the search that much these days. I just try to drill down the page structure to find what looks relevant.


They could have used HuggingFace models to embed texts, put them in a vector search engine and serve results.

- models here: https://sbert.net/docs/pretrained_models.html

- fast vector search here: annoy, faiss, Milvus, Elastic, Pinecone

It's like the hello world of current day NLP.


This is exactly what Mem does using Pinecone and OpenAI: https://www.pinecone.io/learn/mem-semantic-search/

Almost any self-hosted ANN library will do if you're dealing with <1M vectors, but the indexes at these note-based applications tend to balloon very quickly into the hundreds of millions and billions. Given the popularity of Notion I expect their number of (would-be) vectors to be monstrously big. That's where rolling your own gets painful (or downright not feasible) and where Pinecone shines.[1]

[1] https://www.pinecone.io - Disclosure: I work there.


Thanks for the shout-out! For folks interested in playing around with vector and/or hybrid search: Milvus is open-source (https://github.com/milvus-io/milvus).


> How would the AI even know what that feature even is? Where's the context? It seems like this just proposes static solutions to dynamic problems.

The whole feature is kinda cynical in that it already kind of implies that because nobody will read it, you shouldn't put too much effort into writing it either.


They need to do offline first with online sync, rather than online first. But they simply won't do it.


What is the difference?


Writing goes first into your local storage, then gets send to your remote storage, rather than vice versa. Same with searching, search local storage rather than remote one. Everything would load faster, latency would be less, could use it offline and so on.

On the other hand, Notion wouldn't have full control anymore, usage stats might be hard to get when some users are offline, and so on, so unlikely Notion would do such a thing. Certainly their investors wouldn't like it either.


(I work at Notion)

> Writing goes first into your local storage, then gets send to your remote storage

We do this already; writes are queued in SQLite or IndexedDB first and then asynchronously batched and streamed to the server. Our apps (Mac/Windows/iOS/Android) cache pages locally, and all platforms search a subset of documents locally in addition to server-side search. We’re actively improving this area as well.


Can you shed some light on why something like Google Docs is able to operate this way but not experience the lag or latency that your app does?


Typing latency in Notion is generally worse than Google Docs because our renderer & input handling is less efficient than Docs. Docs has substantial lead over Coda, Notion, Dropbox Paper, etc because they use a custom layout engine for presentation that is entirely separate from the input surface. Notion, Paper, etc instead use the Web’s built-in but flawed API for rich text editing called “ContentEditable”. In ContentEditable, generally the rendered output DOM nodes are also the input surface. This is much less expensive to implement, but comes with a bunch of limitations and constraints.


Thank you for the response, it was very illuminating. Is the reason you choose the built-in API instead of building a custom layout engine just due to complexity?


The latency. Accessing browser storage is slow, live syncing is slow.

Maybe ask the Google Docs team? They seem to have low latency online-first doc management figured out.


The only thing that is making me not switch to Mem is lack of tables. I really heavily on tables for notes. And that has not yet been implemented in Mem.


What did your company end up using instead of Notion?


Why is this all so buggy? I don’t know why people like Notion so much; I discover bugs literally every day using it with our team (we are paying members with 2 companies).

So another lame one; I click on this article in iOS safari and it pops open my iOS Notion app saying ‘this content doesn’t exist’. Well done! I will add it to the list, in Notion.

It is the system with the highest popularity while also having the most bugs that I used in a long time. But not for long.

It is also slow; started using some similar (but less feature rich) products and things like Joplin are so nice and snappy that I tend to forget the missing features.


My view is that Notion is an utterly mediocre product with great marketing that also benefited from the rise of "productivity porn" videos on YT.

It probably finds success in enterprise because it can replace a bunch of tools (confluence, jira, google docs, wikis) if you try really hard so maybe it makes financial sense. But in my experience it is not better than any of those in a 1:1 comparison. Okay, probably better than Confluence.


I use Notion and don't think I was swayed by any marketing. For me, I can write a document with great formatting in a fraction of the time it would take me in Google Docs. All personal stuff I use Obsidian but for work I can type up something and share it quickly.

Re: Performance. I only use it in the browser after suffering from the Electron app. It's fast but I am also on an M1 Mac.


For me this is huge. No more random formatting on google docs from team members, notion removes this from the picture and ppl can focus on the content.

Also the ability to nest documents, insert tables that are easy to manage and again formatting is a solved issue.


That's funny, it's all the bells and whistles on Notion that my team complains about. Markdown or Notes.app is way better for composition. Notion is where content goes to stop being iterated on.


I'm sure your team already knows this but you can write in Markdown in Notion. It auto-formats everything. I'm sure it's missing some things and one of them is anything past H3(H4, H5, H6).


It’s more about the affordances. I can’t trust notion not to drop into some command system. Therefore I need to keep my monitor turned on when typing. If I can’t type with the monitor turned off, it must not be a very reliable word processor.


I like to describe Confluence as "a wiki written by someone who has never used a wiki". I feel the same way about Notion and everything it tries to do.


I wonder if it is still impossible to toggle checkboxes on/off in rapid succession without crashing the page. We discontinued using Confluence partly because of this issue in 2019, and just out of curiosity I just checked the ticket again now - closed in 2021 not because it was fixed, but because "we are closing this bug to focus on our upcoming roadmap for all Confluence users."[0]

That seems to be endorsing the exact behavior some users are complaining about with regards to Notion in this thread, leaving so many bugs unresolved while pushing new features.

[0]https://jira.atlassian.com/browse/CONFCLOUD-56972


It is not mediocre. It has flaws but it’s not bad.

It has experienced hyper growth. It is what slack is to irc in some ways.

It’s far better than google docs, confluence and wikis. Jira doesn’t do it.

I’ve looked at many options, some more developer friendly. It’s the best right now.


We started with Notion with a team of 3 people.

Just after 3 months our notes have grown too complex to the point of it crawling every time we load a page - especially if there are more than 1 person on theh page. We ended up switching to Confluence + Google Docs (depending on the use cases). Notion definitely has a nice UX but for any meaningful large document it's utterly slow - that was 2.5 years ago so maybe they've improved but our contents are now stuck in JIRA and migrating out will take too much effort :)


We address this by moving old parts of the big document into sub-documents

But, in general still find Notion to be way better than other solutions for both taking down and rearranging information.


Could you name a better designed product that has the flexibility and feature set Notion has?


That's like asking "can you name a product better at being Notion than Notion?".

Rephrase the question as "what problems do you have which Notion is trying to solve, but you think there are better solutions on the market?".

Personally, I prefer plain-text where I own the content, so Obsidian works great. When I need to collaborate on something in realtime, I usually prefer a more focused tool, like Google Sheets, Trello, Excalidraw.


For that feature set and flexibility, I can't. However, for individual use cases I can.

Note taking: Bear, Apple Notes, Obsidian

Project management: Jira, Linear, Height

Wikis: can't think of any

IMO the flexibility and feature set Notion offers makes it a jack of all trades, but master of none tool. Maybe the sum turns out to be greater than parts for some people but for personal use I haven't enjoyed it. Fwiw I felt the same way about Coda. So it's a general comment on this category of "all in one" doc tools.


> Note taking: Bear, Apple Notes, Obsidian

Bear is really bad. The fact it uses some kind of Markdown but not really is insufferable. Can you even make proper tables in Bear now? It wasn't even possible for _years_

Apple Notes is good enough but doesn't support code blocks properly, that's a deal breaker for me. Also it can't really be shared with people not on iOS/macOS. It's also really bad for writing structured documents: you get 2 title levels, and you can't generate a table of contents.

I didn't fully try Obsidian because I didn't like its general UI. I really liked the graph philosophy it tries to push but couldn't make it work for my notes.


I think Bear is great. Bear provides the exact UI for notes that I was looking for. I take simple straightforward notes as if I was writing on paper slips with a Zettelkasten-like organization. Apple Notes is the only other app for mobile and desktop that has that level of simplicity. In my opinion, if you need advanced formatting and editing features you're not making a note, you're making a document.


I'm not in any way associated with Notion but I find Notion way better than what you've listed _for me_.

I love Notion because it makes it easy to organize lots of notes and has a good search feature. The databases are super useful, I have a bunch of them and I use the template buttons to auto-create new entries. I also use the template buttons to create new entries for my daily and weekly planning & review to make sure I check off all the things I need to do.

For example, I write blog posts on programming, startups, and mental health. I have a database with ideas for each of those three things. Then I have a centralized database where I link to those other databases so I can see which ideas I'm most inspired to write about in a centralized queue.

If you're interested in things like meal prep, using the gallery to show recipe idea pictures, then you can link a database of shopping needs to a database of meal ideas to automatically create shopping lists and meal prep ideas.

I acknowledge that some people have the opinion that you can spend too much time futzing with your tool and not doing the work enough. I respect that for some people something as powerful as Notion is a distraction. For me, my mind races and I take tons of notes and do tons of planning and I find Notion very helpful for organizing everything in a way Apple Notes could never be. I think people need to be respectful of differences on this topic.

I'm someone who's always hated Jira at work because I felt like we spent more time planning than doing work. But with Notion, I use it very quickly for the most part, there was a learning curve but it mostly just helps me out and stays out of my way. The keyboard shortcuts are amazing too, I like automatically adding different colors to my documents using them.

I do like Bear for Apple Watch and I still think Trello has a better mobile UX for boards/cards, so I take notes in those two apps then organize them into Notion. I also think it's funny Notion advertises habit tracking on their SF billboards since I think habit tracking is one of the only use cases Notion is terrible for (disclosure: I'm building what I consider to be the Notion of habit trackers).

But overall I think people are being incredibly uncharitable to Notion in this thread. Maybe it's a little overhyped, maybe performance could be better, maybe some people prefer more minimalist tools (my wife still likes pen/paper and whiteboards over all these things). But to me, Notion is still an incredibly well-designed tool that's incredibly powerful.


I was recently giving Coda a look and it’s got a lot of the same or similar features.


I would rather have multiple apps that do a smaller set of things, but do it well... than one app that does everything, but is bad at all of them.


It's a bit of a different model, but Guru is pretty cool.


True, the Productivity Porn people is the worse aspect.

It's about noting, saving, prioritizing, organizing every little thing. It's the "wake at 4am people, run for 30 min, meditate for 20min and drink quinoa juice" people. The ones that feel the need to be clockworks of effectiveness.

And of course, the more you automate something, it will start getting in the way after a certain point.


I agree with everything. Notion is essentially unusable for me.

My “favourite” bug is that when you import a markdown document, if you have a piece of inline monospace text that starts with a hyphen, the hyphen will disappear. So, something like this:

  Use the `-v` option to enable verbose mode.
Will render like this:

> Use the v option to enable verbose mode.

A-a-and that's how you get a lot of confused people wondering, if your documentation just sucks.


In Notion’s defense, copy/paste didn’t really work at all until very recently. It’s improving.


That’s quite surprising to read for a note-taking application that launched six years ago.


How is that a defense of Notion?


I have to use Notion for work sometimes and it makes me so angry. It just... it's painful trying to do anything with it. It's like the entire thing was written by junior devs who thought "hey, we can do this cool thing" without ever bothering to consider _if_ it should do that thing (like hide a toolbar unless you mouse over it... but the spot it's in is blank anyways, so now you don't know it's there unless by luck AND you don't save space). And then they implement it poorly. So there's a lot of poor ui choices that barely work. Arg.

I will play the devil's advocate to myself though... I know people that love it. I'm just not one of them.


Notion is Sharepoint for zoomers. It’s an okay wiki with slow UX, underwhelming collaboration, and a slew of tacked-on over-engineered features that have better alternatives elsewhere.


Any alternatives you'd recommend? I'm not happy with Notion at all, but I haven't been able to find a good alternative for my own personal use:

- Sync across all devices

- Hierarchical notes

- Note-level sharing for collaboration


For personal use I rely on GitJournal. It plays really well with obsidian + git on laptop. I haven’t automated 1-click note-level publishing; instead I have a personal website that renders markdown.

For collaboration I default to Google Docs instead.


I use Notion extensively and I have to say I have no idea which bugs you're talking about.

After searching for a good note taking app for years, Notion has been life changing for me. Honestly the only 2 issues I have in mind are:

- the sidebar shows collapsing arrows even for pages with no subpages

- there's no way to open 2 pages side by side, so I have to open 2 windows and put them side by side

> It is also slow

It's... really not? Clicking on a page takes like 300ms to load, and once you're editing a page, there's nothing that can even be slow


> once you're editing a page, there's nothing that can even be slow

Create a new notion doc. Create a json formatted codeblock. Paste in the ~250 lines from https://gist.github.com/keeguon/2310008

Add a new made-up entry. Back-space it. Add it again.

I typed for ~10s to do that. It took notion _5 minutes_ to catch up to my typing. I am not making this number up.

Multiple minutes of lag between input and it rendering is the worst I have ever seen from any text editor.

This is for a ~250 line document, which is admittedly in a json codeblock... but still, 250 lines is nothing for a computer, how do you make it take 5 minutes to append a character to a ~9000 character string?

Of course, you'll say "nothing that can be slow" did not refer to codeblocks because everyone knows notion's codeblocks are unusably slow.

Codeblocks are the most striking example, but other things are slow. Let me give examples:

Time to type "cat" into the notion search box (from entering characters to it rendering the actual text) on a large notion workspace - 3.5s

Time for the search to complete for "cat" - 20s

Input lag for just typing regular text on a ~3000 line normal document - 350ms

350ms may not seem like a lot of input lag, but it is very noticeable to me. It causes me to make typos and generally makes notion feel bad to type in. No other webpage I use has input lag this bad.

Now, admittedly, that input lag isn't on every page for me, only about half the notion pages I edit are that bad, but it still makes me dread ever typing text into notion.

Note as well, this is not on an m1 macbook, just a 2 year old i7 (mobile) processor running firefox. Perhaps I need to upgrade my hardware so that I can run this particular text editor.


>once you're editing a page, there's nothing that can even be slow

No, once you're editing a page, there's nothing that should even be slow. Big difference. I've found that if you have a page with more than a few hundred elements (i.e. if you have an outline for a book or something) then Notion becomes unbearably slow.

Also, if you type too fast on the mobile app, Notion will just straight up lose keystrokes. Like, for example, I'd type something like "afford", and what would show up was "a <pause> rd".


I like Notion - I think it's a good product and I think what they have built is really cool.

But OP is right - it's so buggy. Every time I create a new page it says the page is in the trash. Every time.

The release of Teamspaces was sloppy at best.

Some basic computing functionality like highlighting and attempting to copy just doesn't work sometimes. It's so frustrating. If you accidentally grab the block and not the content, you're screwed.

There are also a bunch of usability issues. For instance, I want a DB but I don't want it to show in the sidebar.

Again, I agree, it's great, but it's full of bugs.


> It is also slow

I wanted to build a culture of documentation in my organization, and asking many people to use an app dozens of times per day - performance matters!

We ended up with Slite and are very happy 3 years and 1,000 documents later.

Slite is not nearly as feature rich as Notion (and that's totally fine for us!), but it does what it needs too extremely well.


> Why is this all so buggy?

> started using some similar (but less feature rich) products

You may have answered your own question here.

At least in my experience, PMs at "feature rich" companies love to push for more new and exciting products (and what's more exciting than AI!?) and find it really annoying when devs talk about fixing bugs or improving existing processes for customers.

I'm really hoping one positive result from the current economic turmoil is a return to focusing on delivering value to customers. I remember in years long past when product people viewed solving a customers problem as the key path to a successful company. But I guess years of cheap money and forgetting about the very idea of profit have made people realize that the real customer is VCs.


One issue there I think is that Notion has had a number of investment rounds, according to crunchbase [0] they've gained $343.2M that way; that plus paying customers means they have a lot of money to burn. And investors don't like it when a company just sits on that money, they want results, they want to see growth, they want cool demos of cool new features or products.

I mean you can rebuild the essence of what Notion is with less than $1M. They don't need that money to build more features; I'm convinced that writing the code is the cheapest part of product development these days, despite developers earning five figure sums + bonuses + stock options + $15 / serving premade fruit smoothies from a squeezed bag.

[0] https://www.crunchbase.com/organization/notion-so/company_fi...


If you like performance and have a team usage, Slite should be an interesting choice to explore.

Disclaimer I'm the founder, hundreds of team come from notion every month for this reason. Our search is much more powerful and lightning fast, navigation and doc loading is super fast as well, we put extra care there.


I'm looking forward to the coming arms race: crappy articles and memos expanded and mostly written by software vs automatic summarizers that strip out the padding and puffery and just leave any useful part.*

This "limitless power of AI" (their words) addition to Notion is a great example. It's a kind of anti-compression.

* Ah, turns out this second feature has been baked into unix for decades. Most of the email I get and pages I am offered as output of a web search can be summarized using this handy shell command: `echo < /dev/null`


That future is already here. Recently I've been researching a few physical products and online services, and Google search results returned multiple "articles" from various websites that seemed to be AI generated, with some summarized pros/cons sections that seemed to be from other sources.


There already are posters on this very site openly admitting in their profile they use GPT-3 to write their articles and they get decently upvoted. Confronting them on it didn't go too well...

Perhaps I'm just a Luddite and the only idiot that's fighting against AI-generated content invading the Internet.


I like the idea of "anti-compression"

There are a limited number of ideas with consequence, and we used to read carefully written books about them (or at least some people did)

Now everyone just reads junk on the web that is recycled by computers from other sources, and quite often wrong

e.g. on nutrition, health, finance, etc. I even see it in my parents, who both got Ph.D.'s in the 1970's (when that kind of degree actually meant something). They don't read books, and read the web instead.


> I like the idea of "anti-compression"

I learned this from business books: even the (incredibly tiny number of) good ones are mostly a few pages long. But you can't sell that in an airport, so they are expanded to the size where they can get a binding and go on a news stand shelf.

Really no different from the padding in clothes laundry to keep people from burning holes in their clothes.


This is kind of like how people are working on AI to get their claims submitted and accepted by health insurance companies in the U.S while meanwhile there are companies building AI to help insurance companies reject claims.


Ugh ... yeah corporations were already "slow AI". The bureaucracy had a mind of its own, encoded in forms.

Now when you add "AI" running on computers, it gets so much worse ...




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: