Hacker News new | past | comments | ask | show | jobs | submit login
Launch HN: Quirk (YC S19) – Open-Source Cognitive Behavioral Therapy
196 points by flaque on Aug 8, 2019 | hide | past | favorite | 93 comments
Hey HN, I'm Evan. We make Quirk, an open-source Cognitive Behavioral Therapy app. (https://github.com/flaque/quirk)

If you've never seen a cognitive-behavioral therapist before, it's easiest to understand by analogy with physical therapy. Like a physical therapist, a cognitive-therapist gives you goals and exercises to do, only they're mental rather than physical. Similarly, Quirk gives you exercises where you record and challenge negative thoughts.

I work on Quirk with my brother, Koby, a former founder and marketer in both consumer products and the mortgage industry. He's really good at simplifying, branding, and explaining complicated topics to the average consumer. Personally, I'm just an engineer who used to work on Segment's developer platform.

Quirk started as a command line tool (https://github.com/flaque/freak) that I used to track my panic attacks. I've had severe attacks for as long as I can remember, some lasting several hours. I always used to down-play them, saying "they're a once-in-a-blue-moon thing." But that wasn't really true and they started happening more and more. At a certain point they got so bad that I felt hopeless; I always assumed at some point in my life they'd just stop. But they didn't. So I picked up a book on depression, which turned out to be the CBT bible, "Feeling Good" by David Burns. I started working through the exercises and saw a dramatic drop in panic attacks. (https://i.imgur.com/P29BSRC.jpg)

I started turning the CLI into a little app so I could use it wherever. But since I was having so few panic attacks and was constantly recording thoughts all the time, the project became a CBT app. And the more thoughts I recorded, the better I felt; I went from having multiple attacks a week to having two in a six month period.

I gave Quirk to folks with a "this-software-will-probably-break-if-you-look-at-it-funny" guarantee and open sourced it. But after a bit, it became pretty hard to ignore how many folks were relying on it. That was surprising; it's small and it doesn't really do much from a technical perspective. Its main claim-to-fame is that it works pretty well. Everyday we wake up to folks writing in saying that "for the first time in 30 years I feel peaceful," or that they're "happier, healthier and feel so much better about everything," or that they feel like Quirk is "their secret weapon.” We’re recommended by quite a few therapists but we’re also used by folks who aren’t currently seeing one and had never planned to before trying Quirk.

That's fantastic, but that's not really surprising: CBT has 40 years of evidence saying that it works really well, even in "bibliotherapy" where someone isn't necessarily seeing a therapist. Quirk doesn't do every part of CBT, but it does the most general, applicable-to-everyone part. So what Quirk is really doing is fixing CBT's branding and consistency problems. Most people don't know what CBT is, or if they do, they don't realize just how effective it is and how much evidence is behind it. And when they "try" it, they often don't really do it consistently enough to see the value. Quirk's current feature set is just the thought-recording exercise with some cute illustrations, evidence building to prove to you that it works, and some silly skinner-box tricks to remind you to use it.

A lot of companies solve these massive technical or structural problems. We make a little toy app that helps you feel better. But given just how approachable we could make CBT, we think we can help around 100 million people world-wide in the future. We're translated into 14 languages (kinda, it's a bit broken right now) and have quite an amazing and supportive community.

We charge a small subscription for Quirk, currently $4, though the price might change as we figure stuff out. Unlike a lot of other apps, we're quite privacy conscious; we store your thoughts only on the device and don't have any email or phone collection.

We want your feedback, questions, and ideas. If you're comfortable sharing, we really want to hear about your experiences. But mainly, we want you to know that CBT exists and that this is really something that could help you if you haven't been doing so hot lately.

Nice work! This is a great idea and something I've thought should exist since getting interested in CBT a few months back. Before even trying the app, here are some thoughts.

- On pricing: $4/mo seems like a difficult price point to build a business around, and it doesn't reflect the value you're offering. People pay hundreds/thousands per month to therapists to improve their mental health. If outcomes from using your app compare favorably to a therapist, shouldn't your pricing reflect that? $4/mo makes me think 'toy', not serious therapeutic tool. For people to be willing to pay anything for an app like this, they have to be convinced it can really work. But if they are convinced, they'll pay a whole lot more than $4/mo. I would consider freemium (so people can try it out and see the value) and then a much higher price point (maybe $50-100/mo?). You might get undercut by competing apps, but if you're clearly the best one, I think many people will want the best option and won't be very price-sensitive if they believe it can really help.

- On branding: while it's important to mention that the app is based on CBT so you can piggyback on the awareness people already have about CBT, I think referring to it as a "Cognitive Behavioral Therapy app" severely restricts your market to people who already know what that means, which I'd guess is a very small slice of the people who could benefit. I would try to lead with the benefits--feel better quickly; easier, cheaper, more effective than a therapist; evidence-based; etc.--and save the more technical/jargony stuff about CBT for supporting info.

- On stigma: I would also try to avoid positioning this as something mainly for people experiencing a 'mental illness' like clinical depression/anxiety, but rather something that everyone can benefit from and use to feel better and improve their lives, regardless of how they currently feel. I think many people who don't consider themselves to have a serious problem still engage in cognitive distortions and could see major benefits from using CBT techniques, so I would try to include these people in your messaging as well. While going from bad mood to good mood is obviously valuable and desirable, so is going from good mood to great mood. But if you make it seem like using the app is effectively self-identifying as 'mentally ill' or even just 'feeling bad', many folks in the latter camp will be turned off, so that's something to be careful about.

I'll post my comment here, because I agree with everything written above and I wouldn't be able to state it as well as @danenania.

As for pricing, I would add a second point of view: price the app in a sustainable way right from the start. Estimate the realistic number of subscribers you can get, and set the pricing so that you can continue to develop the app. Too many apps price themselves too low "because otherwise nobody will buy", and end up disappearing anyway, because they weren't able to survive on the small amount of money.

I don't have experience with consumer subscriptions. I can only say that in the B2B SaaS world, I don't think you can build a sustainable business with pricing below $40/month. The consumer numbers are likely lower, because you can get more customers with a smaller support load. But I don't think $4/month is a realistic price point.

Offhand, I feel your advice is contrary.

You suggest increasing the price because "If outcomes from using your app compare favorably to a therapist, shouldn't your pricing reflect that?", but then the next two points on brand and stigma suggest opening it up to all people who want to feel great.

I think you need to choose one. You are either a niche therapy compliment/competitor (price insensitive) or a mental wellbeing tool for a broad audience (super price sensitive).

I personally hope you keep to the lower price point. I think overall you will do more Good and reach more people. Even if you don't make a unicorn, $4/month from 5% of your market, is still $20M/month.

Also, final note, many people suffering from mental illness can not afford therapy. Even $4 would be hard, $50 is impossible.

I second your final note: $4/month can be a lot for someone with depression with nothing saved up. If they have insurance that's willing to cover it, then that might be one avenue, but then there are people without insurance. I don't have any suggestions that would be helpful while being sustainable economically.

Regardless, I'm glad that 'flaque are trying to make the world an objectively better place for some, and I wish 'flaque all the best in this endeavor.

One unrelated suggestion that I might have is to have a fixed buyout price for those that don't want to subscribe, but are willing to pay a higher one time price.

I wouldn't call therapy a niche market. Therapy apps might be currently, but if they can disrupt the current therapy model to a meaningful extent, they won't be niche for long.

While I don't have time to look for better data on this, a quick google turns up this 2004 survey via https://www.apa.org/monitor/julaug04/survey:

'A similar poll released in May, "Therapy in America 2004," and co-sponsored by Psychology Today magazine and PacifiCare Behavioral Health, found that an estimated 59 million people have received mental health treatment in the past two years...'

59 million people (in just the US... in 2004) is not a niche market! And these people are paying a lot.

The problem with trying to be a "mental wellbeing tool for a broad audience" isn't just the market being super price sensitive, it's the market being unwilling pay anything whatsoever, which leads to ads and selling personal data (yuck).

"Even if you don't make a unicorn, $4/month from 5% of your market, is still $20M/month."

Right, but how will you reach 5% of your market? I don't see much virality here, and $4/mo gives you no room for customer acquisition costs.

"Also, final note, many people suffering from mental illness can not afford therapy. Even $4 would be hard, $50 is impossible."

I agree, but an unsustainable business can't help anyone at all in the long run.

> I agree, but an unsustainable business can't help anyone at all.

10000%. This has basically been my learning experience (Koby's much more knowledgeable than I am). It's amazing how many people you can help if don't die.

Also 100% agree with all of your points! Though, you'd be surprised about how viral it currently is and also how cheap the customer acquisition costs are. (That said, I don't expect those trends to continue as we grow, they tend to get harder.)

I think he's definitely correct that we're too cheap right now, but I agree that you can still make quite a company and still be dramatically cheaper than any reasonable option. There's definitely a middle ground.

On price: $50/m would be IMPOSSIBLE for most people to pay. You are backing the $50/m with data viewed from inside your bubble. I'm part of several online groups and communities for people suffering from Anxiety and Depression and out of thousands and thousands of people there, both in the US and around the world, I'd say only 20% have the money to go to therapy.

Your data showing that "a lot of people are paying a lot of money" is already biased because it's based on a particular socioeconomic segment. Do you want another data POV?

"...only 70 percent [of americans] feel they have adequate access to mental health care." - https://www.apa.org/monitor/julaug04/survey

Please be mindful when making comments like these, as not all of us are as lucky and/or privileged as you are.

OP: Your pricing is fine, probably $9 "ala netflix", that is still low enough for most people. Awesome product!

Wanting to make a product cheap and therefore available to as many people as possible is a noble instinct—especially a product like this one. I totally get it.

But unfortunately that doesn’t change the fact that for many reasons, selling a product at a very low price point is an extremely difficult way to start a new business. Can it be done? Yes, of course, but you better be able to either raise tons of money or acquire customers very cheaply.

An app that costs 15% of what weekly therapy does and is similarly effective will help a lot of people. A company that dies because it can’t afford to advertise will help no one.

WOAH thank you for the in-depth feedback dude!

On Pricing - Agreed. This pricing mostly came out of a bit of nativity on my part. I assumed that wanting to make something accessible meant being really cheap. But we're very quickly realizing that both we won't be able to sustain this price and also that by having this price, we're basically telling people it doesn't work.

The reason why we haven't done too much on pricing quite yet is because it JUST started working as of a few weeks ago. We technically had it working a month and a half ago, but we messed up heavily with the rollout. Due a bug, it involuntarily canceled everyone on Android and then messed up our attempt at grandfathering existing users as "free-forever."

We might try freemium in the future, but at the moment we don't have the resources. In our past experience, freemium products tend to be basically two products where one is trying to sell you on the other. In a purely-paid model, every hour we spend on making a better product directly contributes to keeping the business alive. Since we're so small, we don't have the hours to spend yet.

On branding - This is super true, but when we tried to just say "therapy" or "feel better" people thought we were talk therapy and then got upset that there wasn't any talk therapy. If you say "CBT" the folks that have heard of it LOVE the product and the folks haven't we can explain what it is from a blank slate rather than trying to change any past stigma or misconceptions about therapy.

But this is something we're still not that great at. Figuring out how to do it well will be a lot of our core focus in the coming months. Especially considering our current website is basically a wall of text: https://quirk.fyi

On Stigma - This is exactly what we want to do, but haven't been very good at it yet. If you're reading this, CBT can probably help you, regardless of any specific condition.

> While going from bad mood to good mood is obviously valuable and desirable, so is going from good mood to great mood. But if you make it seem like using the app is effectively self-identifying as 'mentally ill', many folks in the latter camp will be turned off.

That is a REALLY good way of framing it. Holy moly, I'm gonna steal that.

This advice is amazing! Thank you so much! Seriously!

I think the price is great and you should keep it. It's the perfect solution for a teenager who really needs help but doesn't want to talk to anyone about it, and doesn't want to tell their parents.

I feel if something like this could ever be demonstrated to be as good as/almost as good as therapy, it should be run by a foundation with draconian data privacy rules and offered for free to everyone. This "make some programmers a lot of money and suddenly introduce us to a new future where we don't understand any of the implications of our new incentives" model is really getting tiring and often scary.

All of this ^

Some pushback on pricing: this app is a vitamin, not a painkiller. $4/mo is a perfect price for an app I may use daily for years and years with varying levels of engagement. Like Day One (price: $3.99) or TeuxDeux (price: $3/mo), both of which are quite successful.

Charge $10/mo or $20/mo and your churn will increase as people think "haven't used that in weeks, waste of money" and cancel.

Additionally, I think the term 'open-source' and link to GitHub have no place in your marketing materials at this point. You're gunning for the public now, and your support channel is going to be clogged with questions about why open-source does not mean free or how to install from GitHub.

(Also, this is awesome!)

> I think the term 'open-source' and link to GitHub have no place in your marketing materials at this point. You're gunning for the public now, and your support channel is going to be clogged with questions about why open-source does not mean free.

Agreed, basically only using the open source terminology in dev environments like hacker news. So in our screenshots and marketing materials, it's definitely not going to be anywhere.

We've expanded outside of just hacker folks at this point and some folks got really worried about privacy because they thought "open-source" meant that the their thoughts would be shared publicly.

Happy to help! I've been through a similar process on pricing with my startup so I completely understand the difficulty.

Makes sense on freemium--I think it's good to first make sure you can satisfy users at whatever price point is sustainable and then later you can open up the floodgates via freemium when you're ready for the volume.

> I think it's good to first make sure you can satisfy users at whatever price point is sustainable and then later you can open up the floodgates via freemium when you're ready for the volume.

Exactly, would much rather be sustainable first.

Though also, on pricing: it’s not just a question of what the app offers, but also of what competitors are offering, and at what price. For example: the company that did Monument Valley teamed up with some doctors to do a very similar app in this space called Moodnotes* , and priced it at a one-time payment of $5.

That is to say: I am glad for Quirk’s existence - if it gets the useful techniques of CBT in front of even one person that wouldn’t have seen it otherwise, then it’s done good in the world and should be proud, and has earned that $4/month - but I understand why they aren’t using the cost of talk therapy as their point of reference for competitive pricing.

* https://www.ustwo.com/work/moodnotes

Yeah, that's a good point. I think it all comes down to framing. Are you selling a "mood-tracking app" or "a self-administered therapy system" (that happens to use an app as one piece of the system)? The former anchors you very low (too low, I would argue, to build a sustainable business). The latter anchors you against weekly $100+ therapy sessions, which gives you vastly more breathing room.

The flip side is that customers will of course be a lot more discerning and demanding with the latter framing, as they should be. It will be a lot harder to get product-market fit, but if you do, you'll be able to build a great business on top of it, rather than just being another competitor in a race to the bottom.

I recommended Quirk to my girlfriend a few months ago. Still haven't tried it myself.

You seem to know what you're doing so my only advice is to not take VC money or you'll probably be pushed into making this app a general health app that is a bit useful for everyone instead of very useful to a lot of people, for the sake of money.

Does the app store any data in their servers? I'm hardly obsessed with data privacy but I would never, ever, ever want my mental health data out there for some data scientist to play around with and manipulate.

Facebook knowing all the dumb crap I do on their platform really pales in comparison to the data privacy issues of an app. If I get a whiff of anything negative, I'm not using it. I'm suspicious already.

No, it doesn't store you mental health thoughts on any servers. They're all stored only on your device.

The caveats are below:

We collect anonymous, aggregate (as in "user-213sdfewfefwa") events. This is stuff like: > "user-awefawewa21312aef" opened the app > "user-efaoiwejf21321few" started a free trial

These are used for aggregate stats like "how many people use Quirk" or "do people read the articles?" This is recorded by the phone stores anyway, but we do this both so we can get some more Quirk-specific stuff and so we can combine Android and iOS data.

We also collect some base information on your phone. For example, what type of phone you have or what language you're using. This is basically a requirement to debug stuff; since there's a lot of bugs that crop up only on certain OS/Phone/languages.

If your phone crashes or there's an error thrown, we use Sentry to report it. This lib will collect some basic info like phone types, IPs, and then the stack trace of course.

In the future, we may provide a way for you to share a thought with a trusted friend or therapist. This has been a highly requested feature, especially by folks who would otherwise huddle around their phone during an appointment. To do this, our plan to use an expiring, private link. That means your thoughts will be stored in raw form in a DB and then get deleted within X days. This would only happen with your explicit consent and it's only for a thought you share rather than your whole list of thoughts. It's a button you have to press EVERY time, not an "automatic share my thoughts" sort of thing.

We'll also be adding a cloud-sync feature at some point so you never have to lose your thoughts when you switch phones. When we do this, all the thoughts will be end-to-end encrypted; so no Quirk employee will ever be able to see your raw thoughts.

Because we're a paid product, we have little incentive to read your thoughts or strip-mine your data. We're in the business of making you feel better; we're not reading your deepest darkest secrets to figure out how to sell you shoes or something. We don't want to know your thoughts. Unless someone explicitly tells us otherwise, we treat them like passwords.

> Because we're a paid product, we have little incentive to read your thoughts or strip-mine your data. We're in the business of making you feel better; we're not reading your deepest darkest secrets to figure out how to sell you shoes or something. We don't want to know your thoughts. Unless someone explicitly tells us otherwise, we treat them like passwords.

I trust you when you say this, and I trust that you are being sincere. I do not trust the incentive structures of Silicon Valley to permit you to adhere to this principle.

> I do not trust the incentive structures of Silicon Valley to permit you to adhere to this principle.

Neither do I! My point is more so that we try to take steps not just to say "hey we don't read your thoughts" but also to align the incentive systems so that 10 years down the line we're at least pointed in the right direction.

Why not switch to a Patreon model and give people simple “export all data as markdown/json” inside every version of the app? In that way there’s less incentive for the business to become potentially untrustworthy, but if it does, people’s emotional investments are still protected by being firmly in their own hands.

My therapist can't read my mind either but I bet someone could craft some laser-targeted ads directly at my nervous system with her notes in aggregate :P

I'm glad you're thinking about this. Good luck!

I still don't like it. I've dealt with a lifetime of stigma, to the point where I don't even want to write an anonymous comment about it.

> Because we're a paid product, we have little incentive to read your thoughts or strip-mine your data

That may be who you are now, but what if later you want to take on investor money? Then, the incentive to mine data becomes this: even more money. As Batman once told me, "power is never innocent."

These kinds of apps should ideally be open source and with an option to completely turn off the telemetry. Also fully desktop and/or mobile native, without web backend of any kind.

Thank you for recommending it! I really hope it's helping.

Re:pushed-to-be-general -->

1000%, this is definitely a concern. We're definitely new to this world, so we're talking a lot with more experienced founders who understand how to navigate it better than we do. We also do a lot to setup systems that align the incentives of the business with the incentives of the user; for example being open source is one strong system.

What user data do you store? Do you sell user data? How do you protect user data?

I trust your founding story, and I truly believe you are acting in good faith, with positive intentions. I think your rationale for a subscription model is wise & insightful.

However, recent history has taught me to be cynical. I have to think there's a market for more and higher-resolution emotional profiles to open up new targeted advertising delivery channels. Especially information that can be used to feed algorithms for building higher resolution models of my inner life.

To answer the obvious question: we don't store your thoughts.

We store anonymous behavioral info like strings of "USER-a23o13jaf signed on to a free trial." Then our bug-checking system also stores some anonymous device info like what phone you're using, how big it is, the language you're on, that sort of thing.

We don't sell your data. That's pretty scummy.

We protect user data by collecting as little as possible and not storing it. We don't collect your email or your phone number or anything really. Signup is handled entirely by the stores.

> recent history has taught me to be cynical.

You should be cynical. I'm cynical. There's no real way for us to convince you to trust Quirk, since people at companies change and because people change.

That's why we're so focused on systems thinking rather than just telling you to trust us. We'd rather set the company up so that it's always headed in a direction that benefits customers. This isn't a perfect solution, but it's the best one I can think of.

To improve trust: allow each user to see the raw data that’s sent with every request and when. Default telemetry to on as you’re open about collecting it (and it necessary for debugging), but allow people to turn it off completely. You could even store it on-device just in case a user has a problem later and decided to help by sending past telemetry.

Don’t take VC money? They are a YC company. That ship already sailed.

Got discouraged by the fact that I have to introduce my credit card details to try the app. This is a no no for me and I'm sure that it will be a no no for a lot others.

Literally their first sentence is bolded and calls out the subscription fee.

I wasn't able to make it to WaffleJS yesterday but I hope your talk went well! As someone who's dealing with anxiety about the job hunt/interviewing for my first tech job right now, I can definitely see myself using Quirk.

Best of luck.

Thank you! Good luck in the job hunt/interviewing. It took me a LONG time and a lot of tries before I stopped freezing up in interviews.

CBT (with or without Quirk) is definitely something that can help you here and definitely recommend giving it a shot!

I never post here but I'm a huge fan of this and what you're doing. I am going to share this around with folks I know will benefit from this.

Awesome product and idea, I can definitely see this solution (or similar ones) be a crucial "culture-boost" for companies. And, of course, +1 for open sourcing it.

With that said, any reason you guys decided to GPL it? I actively try to avoid both contributing as well as using GPL-licensed software (e.g. non-permissive libraries). All my stuff, for example, is MIT'd.

To be honest, I just kind of picked it and went with it. I think at the time I had been talking a lot with my dad (an old copy-left enthusiast). I never really thought it would really be a thing.

Fair enough, I don't think it's a huge deal, it just makes it very inconvenient to use any cool parts of your code commercially or in other projects. Sometimes I'll work on a project without knowing if I'll ever commercialize it, and I try to err on the side of using MIT (or Apache) libraries/code as opposed to (L)GPL.

True, it makes sense for software packages, but less so for real products. If we build out any packages for public use in the future, they'll likely be MIT not GPL.

What do you mean with culture boost?

How do you see CBT could help teams?

I think they're referring to the various Mental Health as an employee benefit. Quirk isn't really that; we don't have any plans to do b2b stuff. But I admire the folks that do!

That said, CBT is extremely helpful in personal interactions and generally getting along better with other people. It teaches you to not mind-read and generally assume the best of other people.

For example, let's say you're a new engineer and you just made a big mistake like leaking a secret key online. In your inexperience, you don't realize that this is something that happens all the time.

Instead, you think "oh no, what if my manager finds out." So you avoid conversations and act defensively. Eventually your manager DOES find out and the whole situation goes from an innocuous key reset to a much bigger deal.

But the initial problem was caused by the catastrophizing that happened when the new engineer thought "oh no, what if my manager finds out." They likely assumed they might be fired or disciplined; which may be closer to the truth now that they've kept it a secret.

This happens all the time; it's a self-fulfilling prophecy that happens when we don't check-in with our thoughts. It's really easy to believe the "distorted" view of what we're thinking rather than a more logical perspective.

Quirk's advantage here is you can immediately pull out your phone and check this thought. You know you're feeling stressed or scared, so that's your cue to pull out Quirk. Then, it can guide you through the process of logically challenging this. It's likely that by the end of it, you'll have given yourself enough perspective that you don't feel so afraid.

So you don't keep it a secret, you tell your manager, it's an immediate easy fix and you reinforce a healthy trust for your team and the folks you work with.

The potential for this kind of application is great; it allows people who would otherwise decline, postpone, or forget to get this kind of therapy. All while delivering it through a much more cost effective medium.

Just wanna say that "Feeling Good" by David Burns changed my life.

Best of luck to you :)

Mine too! Thank you!

1) How will migration to new iOS devices work if the thoughts are only stored locally?

2) I popped the app open when I found it, but haven't used it yet. One thought I had then, I'll share here: if it doesn't have a mode similar to the Lotus Bud app, it might be nice to have something like that. Basically, you can set up the Lotus Bud app to check in randomly during the day (window of time, and frequency are both configurable) and invite you to "clear" for a moment. Quirk could ask if there's a thought stream you'l like to address.

1 - At the moment, we don't. :( One of the next things we'll be working on is an end-to-end encrypted cloud sync. So we won't storing the "raw" thoughts, but you'll still be able to get access to them.

2 - Ah, our reminder notifications kind of work like that, but we'll probably try to make that more configurable as the days go on.

1 - DayOne (and to a lesser extent 1Password) had problems from this. Both supported Dropbox initially which was great for privacy-conscious users as it was user-controlled sync, and easier on the company as they didn’t have a server infrastructure (or much of the accountability). They’re both now pushing for sync to their own cloud (helps drive the subscription model) and it’s starting to show if you look at the behaviour of those same privacy-conscious folks. I think it may be a mistake to move to “trust our black box” cloud sync in your case as you’re asking trust of the very people who in some cases have feared trust backfiring for their whole lives.

If anyone is facing mental health problems my advice would be to minimize screen time and avoid using the internet unless absolutely necessary. More often than not screen time contributes to the problem.

I downloaded it yesterday and plan to try it out this weekend. I think this is great because this seems to be an empowering way for someone to do some CBT with a low barrier to entry and very accessible.

Thank you! Please let us know how it goes for you. It definitely takes a bit of practice; it took me a bit to get it at least.

Our biggest challenge at the moment is helping folks through that "initial" barrier.

Hi Evan! It's Brian - I helped out on the internationalization issue. Just here to say I'm glad to see this showhn and good luck on everything!

Thanks Brian! :D

I've actually been building something similar myself. Although mine is mostly just for tracking stuff in your life and forming statistical analysis from the data. And I do nearly everything server-side.

I think you guys should set up a website to explain and promote this a bit better. Basically how you introduced it here but just on a website.

Check it out! :) www.quirk.fyi

Any reason there seem to be two domains? One is quirkapp or something. Then there’s the one you listed.

Cool thanks! Now this explains a lot more of what I'd want to hear as a potential user

I started working through the exercises and saw a dramatic drop in panic attacks. (https://i.imgur.com/P29BSRC.jpg)

That's an impressive, likely clinically significant effect, congratulations! May I ask what is the x axis in the graph?

The x axis is a labeling bug. I THINK it was weeks? But the code that generated this was a small python script that was on a laptop that I wiped.

I wouldn't assign any clinical significance though to it (or to Quirk quite yet), especially since it's an N=1. But even if it's buggy, that graph shows one of the most important things that's every happened to me.

Looks like epoch time

If curious about the history of this type of treatment, check out a very interesting character named Albert Ellis: https://en.wikipedia.org/wiki/Albert_Ellis

Great app and execution!

I have been considering an open source + subscription model for an app I'm working on. Do you have any concerns about "piracy" (someone forking it and removing the subscription stuff)? Any lessons learned along the way?

So far no one has done it. Quirk's got a closed source api that it interacts with, which makes it a bit trickier to straight rip off. Plus, we own the brand, which is a pretty big reason why people try Quirk in the first place.

How are you handling operating in states which require a license for conducting therapy?

Not a lawyer, but my guess is that since people are self-reporting into a text capture tool, quirk is no more liable than Microsoft would be if users were to do this into Word.

Mostly this yeah. We're not talk therapy, we don't claim to connect you with a therapist, and we don't take health insurance.

How do you guarantee privacy?

I know it's open source, but I can't see how I can check that the app store version is an exact copy of the OSS version.

Obviously, I really would not like to see my most intimate thoughts end up for sale at some information broker.

This is a great question. Is there a technical way to prove that an iOS or Android app is built from a specific git commit revision? And what about a web app? I know that these apps are signed but can a signed app be tagged with a transparent revision number which cannot be tampered with. Kind of like a sha2 hash that can be reproduced by building the same source code.

This is something I've wished for for a while. There are many developers out there whom I'm sure would love people to trust them and that are trustworthy. But I feel we now live in an age of "trust, but verify" and there's no great way to verify privacy claims even if we really want to believe something is benign.

In some cases it may even be unintentional leaks if someone isn't aware of an issue.

Unfortunately, there is little incentive for Apple, Google to fix this situation because the majority of people simply don't care.

You can't.

I think we're a reasonable bet because you know where the money is coming from. We have no incentive to sell your thoughts to anyone, because you're literally the one paying us.

But you'll really just have to trust us. I'm a real person behind the scenes here and I don't have any desire to read your thoughts. When I show people Quirk, I wipe everything off my phone and hand them a fresh copy. I don't want anyone reading mine either.

So we don't store them off the phone. And if we ever do, they'll be end-to-end encrypted.

The little circle guys are hilarious. I really like how instructional the landing page is, with a sort of old-world informational ad copy style instead of just a bunch of flashy screenshots.

Taking it for a spin, and wishing you the best!

Thank you! It probably needs a refresher at some point, but I'd like to keep the explanation focused style.

Plus, we'd really like to make the site have lots of free explanations about CBT and it's concepts. That's why we picked the "fyi" domain originally.

Your app charges for subscription but it is also available as open source. Does that mean somebody else can potentially publish it on App store and offer it for free?

They could but a lot of features would break. The app communicates with a closed source API. Plus, we still own the name and the brand and would likely send a take down if someone didn't remove them.

I LOVE that the app is so approachable and easy to use. Great job!!

Cool idea, but what advantage does open sourcing the app give you?

I know folks are supposed to have reasons for open sourcing things and I might a bit naive here, but the best answer I can give you is: ¯\_(ツ)_/¯

I like open source. I couldn't find any strong reasons why it couldn't be open source.

But what I think happens to a lot of open source consumer products is that they're primary feature is being open source. Outside of Hacker News, we don't tell advertise that we're open source. It's not important.

For most people, we're a really good CBT app, not an open source CBT app. Our competitive advantage to the average person is that we're good, not that we're open source.

That said, I think there's advantages for customers. You can be reasonably certain we're protecting your data well. If you don't like something, you can suggest a change. If you want to use Quirk in your research, we're, in general, the most open about things.

To be clear, we're not 100% open source, there's a small closed-source API that quirk uses to do some of it's features.

This is so cool! I love how simple the UI is.

What's to stop someone making a free clone from the open source version if the paid version really takes off?

Purely my opinion:

Like many hip big startups today, one of their strongest competitive advantage is having users. Most users come because product A offers a better UX than product B.

At least I feel that most big FAANG companies are mostly UX plays. In some cases you have to get very technical and innovative to give a good UX (Google), but in other cases not (Facebook).

Hmm, there's a closed source API that Quirk interacts with. Plus, we still retain ownership over the name and branding.

This looks amazing, thank you

I was dissapointed to open the app again - I had been using it frequently when I switched to paper thought-records. When I reopened it I was greeted with a paywall.

Alternatives on iOS exist such as Thought Record that are free. Until there is an obvious value proposition that differs from free alternatives, I see no need to pay merely for a 'quirkier' interface.

This is due to a bug where some folks didn't get caught up as "free forever."

Send me an email and I'll fix this for you and give you the app for free: ejc@quirk.fyi

Hopefully you'll like the new version as well!

Thanks so much for this offer! I ran into the problem too on my iPad.

And Quirk is great for such a new product!

I've used a couple of principles regarding CBT to become more social and to get a better dating life. The first book I read was Michael Hall's Mindlines, which was an NLP book (a nonsense pseudoscience acronym). A couple of years later I learned about CBT and saw it has a lot in common with NLP. The biggest difference: CBT has scientific research into it and NLP doesn't, a couple of people simply made it up. When I tackled these problems I was 16, around 18 I was a lot more scientific about this.

The key insights regarding CBT for me were the following:

1. Start with small goals and celebrate them.

2. Log the following questions:

A. To what extent do I think event foo or thought bar will happen? E.g. to what extent do I think I'll get ridiculed while introducing myself to this new group of people?

B. (after it happened) what actually happened and to what extent was this in line with my expectations? After doing it for 10 times, 5 groups were cold and generally non-responsive, 1 ridiculed me and 4 were very friendly and happy to have met me (from experience this would be a particularly worst case realistic example for me).

3. I've used this so many times: if you feel terrible, ask yourself whether you're having the right perspective. If you feel like you do, then ask yourself it's the only right perspective. If you feel you do, see if you can find other perspectives anyway and ask yourself whether those perspectives gave you new information. This did wonders on how I viewed my physical attractiveness. My opinion used to be: I'm ugly. My opinion now is quite nuanced and grouped into segments by age and personality type. Conclusion: the people I like either don't care (too much) or they find me attractive. It took me a long time to this conclusion. For a long time I had another one which was: I'm not that good looking but looks don't matter too much in dating, or at least not when people meet me.

Those 3 insights really helped.

It's funny, I also studied psychology but that never really helped as much in this area as CBT before it.


If you guys feel up for it, then send me an email (see my profile) to reach out to brainstorm on things regarding CBT/psychology, dev or UX. I thought this was completely solved but apparently it isn't.

I wish there were more positive psychology[1]/tech companies on the whoishiring thread, if I weren't applying for McKinsey right now, I'd have publicly asked to see whether I could help you guys out.

[1] While CBT isn't part of positive psychology, I feel they're quite close to the spirit of it which is why I include it.

1000% this: > if you feel terrible, ask yourself whether you're having the right perspective.

One of the most common bits of "debugging" we do with folks is that they know they feel like crap, but they're not sure if there's a thought associated with it. Almost every time, the thought that they should be challenging IS "I feel like crap."

I find it one of the most intriguing parts about CBT, because I don't know of any research on one of the fundamental assumptions it has (from my interpretation), which is: your feeling/emotion is caused in a big part by a thought you have.

For me this seems to be at least mostly true, which is why I can use the assumption. It was then that I realized that I'm a believer of the appraisal theory [1]. But I still didn't find any convincing paper on it to what extent this is true.

The danger this can have is that it may be that some feelings/emotions are not caused by your thoughts, but because of something else, but yet you try to CBT it away. One example is: you simply have a chemical imbalance. The first question I asked in this case is: do I exercise? If not, then let's hit up that dopamine button first in a healthy way. If I do exercise, the next question is: am I eating healthy? I also put social contact in as a 3rd possible factor. Whenever I went to meditation retreats (they double as a silent retreat), I noticed that my thoughts went in loops and other people break you out of your thought loops. It was then how I viscerally experienced how one might become crazy when they are left alone. I also found ways to battle it (e.g. try to do math in your head and come to newfound knowledge).

My full (mental) checklist is:

1. Am I exercising? If not then there might be a dopamine issue.

2. Am I eating/drinking in a healthy way? If not then your body might not be working properly.

3. Did I sleep in a healthy way? If not then the same as point 2.

4. Am I experiencing constant stress for more than 2 weeks? If so, then cortisol might be screwing up my HPA axis [2]. I should check for burnout/depression symptoms to verify.

5. Have I been a hermit for more than 2 weeks? If so, have I been focused or mind wandering? If I have been mind wandering mostly, then it's time to visit some friends.

[1] https://en.wikipedia.org/wiki/Appraisal_theory

[2] https://en.wikipedia.org/wiki/Hypothalamic%E2%80%93pituitary...

Are you hiring? :joy:

Probably not for a bit; but shoot me an email with your info! ejc@quirk.fyi

Awesome! Sorry, I was only semi-serious, because my last name is Quirk and it seemed like a good fit :D

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