_______ __ _______
| | |.---.-..----.| |--..-----..----. | | |.-----..--.--.--..-----.
| || _ || __|| < | -__|| _| | || -__|| | | ||__ --|
|___|___||___._||____||__|__||_____||__| |__|____||_____||________||_____|
on Gopher (inofficial)
URI Visit Hacker News on the Web
COMMENT PAGE FOR:
URI Show HN: Dlog â Journaling and AI coach that learns what drives wellbeing (Mac)
dr-j wrote 1 hour 56 min ago:
The responses here so far have been amazing; working hard on your
suggestions. I wanted to also share that Dlog received its first sales
and founding supporters from Show HN in the last 24 hours; I can't
thank you enough for the encouragement and motivation.
huem0n wrote 4 hours 11 min ago:
Please, both hackernews readers and the author take a closer look to
see the cracks on the site. I'm fine with using AI generation, but it
needs human review especially for legally binding stuff like the
privacy policy.
1. Look at the concepts pie on the home page. The text in the pie is
unreadable. Its overlapping and overflowing, white text clipping onto a
white background with terms like "topic tagging" that are not an actual
example. Its like no human looked at the image before putting it on the
website. Maybe just a slip up, we all make mistakes, let's keep
looking.
2. I didn't understand the data storage/privacy from the video, so
let's look at the privacy policy. At one point the policy says "Do we
receive any information from third parties?
No, we do not receive any information from third parties."
Right before later saying:
"journal entries or project-related text that you select are sent to
the ChatGPT-5 thinking nano API operated by OpenAI."
Open AI *is a third party*! The answer is "Yes we send data to Open AI
under these conditions". That's bad.
3. Lets look deeper. The privacy policy says they store 3 things with
the first bullet point (in full) being "A unique user ID number that
cannot be used to identify you." You're telling me a literal
Identification (ID) Number can't identify me? Why does it exist? That
is borderline nonsensical.
4. The video has similar vague stuff saying the data is processed
locally after saying its going to chatGPT 5.
I'm giving harsh feedback because I want a project like this to exist,
be done right, and succeed. I understand "ship fast and iterate".
You're going too fast and you're not shipping an MVP, there is lots of
feature creep.
Even when everything looks good, people should be hella skeptical
about an app that wants to (potentially) harvest extremely personal
daily journal logs. When every page smells like "I generated this and
didn't fully check it" it makes me imagine how many hidden problems
there are in the codebase.
- The kinda-rough AI video tells everyone "I don't have time to record
a 5 min video of my own project". If you want me to believe you care,
at least hire a narrator on fiver for $20 if you don't like speaking
and/or showing your face. Why should I trust what you say you'll do
with my most personal data when you don't even show yourself/show a
human?
- There's only three important things: pricing, privacy, and the data
analysis / coach. Leading with price is good/solved. What's missing is
clarity about privacy. The hackernews post is much more clear, the
website is not. I don't need more words, I need to know when the data
is and is not shared and I need to be convinced you're responsible.
Right now stuff like "Dlogâs private AI model" makes it confusing
what's local and what's shipped to OpenAI.
- Even when explained clearly, privacy is going to be a problem. Let me
use me use my own model/token/url. It's easy to point to a local URL
that responds with data in the exact same format as GPT 5. That kind of
feature is 10x more important than changing the color of the
background.
- I'm not getting a coaching app because it has a good theme engine.
Finish talking about coaching/analysis before going into themes and
calndars etc. I don't even care how data is entered into the app, until
after I know the useful things its doing. Give a real example of
insight that changed your daily choices.
- I think you can do it, and I'm glad to see someone trying to meet
this usecase.
dr-j wrote 2 hours 0 min ago:
Appreciate the detailed look. A few clarifications and immediate
fixes:
⢠Concepts pie: noted. Itâs a minor visualizer and not a current
priority; there are dozens of other charts in the Dlog Lab tab.
Iâll queue a fix, but Iâm focusing elsewhere first.
⢠Privacy policy: youâre rightâOpenAI is a third-party
processor. Iâm correcting the policy to say exactly when data is
sent, what is sent, and under what controls (Enterprise API with no
training/retention). Iâll also add a simple data-flow diagram.
⢠Local vs cloud: journals live on-device; the SEM runs locally.
Scoring only happens when you explicitly choose to scoreâthereâs
no background upload. Iâm adding a per-journal âInclude in Coach
analysesâ toggle, and a simple âRemove namesâ anonymizer (ships
in the next few days) so names are stripped before any scoring call.
⢠BYO endpoint: not on the near-term roadmap. Iâm prioritizing
clear privacy controls and product focus over supporting custom model
URLs right now.
⢠Copy/video: Iâll tighten the site copy to lead with privacy
and analysis, and make the local vs cloud boundary crisp.
Thanks for pushing on clarityâfixes are in motion.
behole wrote 4 hours 23 min ago:
This looks really cool! I would love to take you up on your offer to
show hn viewers! Dm you my Dlog signup email?
dr-j wrote 2 hours 17 min ago:
Awesome. Yes, please do; or email me at johan@dlog.pro
inavida wrote 8 hours 44 min ago:
Hi I'm nobody and the stories on here constantly remind me of this old
parable. There once was a man who was afraid of his own shadow and who
hated his own footprints. Somehow he got it into his head that if he
could just run fast enough he could outrun them, so he ran and he ran,
but no matter how fast he ran his shadow stayed by his side, and the
faster he ran the more footprints he made. Somehow the man got it into
his head that he just wasn't running fast enough, so he ran and he ran
and he didn't stop or rest and he died. Not knowing that standing still
is how to stop making footprints, and that resting under a tree is how
to stop making shadows, is just so tragic.
dr-j wrote 1 hour 46 min ago:
Thanks for this; that parable lands. Dlogâs goal is the opposite:
surface your personal patterns so you can deliberately stop, rest,
and protect downtime.
nathan_compton wrote 2 hours 31 min ago:
The only thing I hate more than the idea of a life coach is an AI
Life Coach.
It takes all kinds to make the world go round, of course, but the
entire vibe of this part of contemporary culture grosses me out and
depresses me.
dr-j wrote 1 hour 46 min ago:
Fair point, AI coaches arenât for everyone. The Coach in Dlog is
optional; nothing is scored or sent unless you explicitly choose
to.
desijays wrote 10 hours 46 min ago:
I don't do subscription based apps as a matter of principle if it has
no business being a subscription based app
dr-j wrote 5 hours 41 min ago:
Fair enough. But, you may not have noticed this edit that was added
to the original post which explains I am giving a perpetual license
and 1 million tokens to all Show HN readers; i.e,. so no
subscription.
For general users it's free for 14 days with 10K free tokens; then
its 1.99 per month at the moment. However, for HN readers that DM me
or email me (johan@dlog.pro) with the email they register with, I'll
give a free perpetual license so there's no monthly fee; and add 1
million tokens.
minihat wrote 13 hours 47 min ago:
You may enjoy the 2024 Hugo winning short story, "Better Living Through
Algorithms":
URI [1]: https://clarkesworldmagazine.com/kritzer_05_23/
dr-j wrote 1 hour 51 min ago:
Thanks for sharing that story, great cautionary tale about apps that
drift from âhelpfulâ into prescriptive/social enforcement.
Dlog takes the opposite approach:
⢠Agency first: The Coach proposes; you decide. No phone trees, no
task assignments, no social mechanics. You can ignore, not use or set
a low-cadence to the guidance.
⢠Explainability over prescriptions: Suggestions come with a brief
âwhyâ based on your SEM reports (which factors moved and by how
much) plus charts so you can sanity-check against lived experience.
⢠Local-first privacy: Journals live on-device (EventKit). Scoring
+ SEM run locally. By default no raw text leaves the device unless
you use the coach, this is optional, but there is a bit of a leap of
faith here with OpenAI; until I enable on device LLMs in due course.
⢠No hidden incentives: No affiliate nudges, upsells, or growth
hacks. It doesnât decide what you wear/eat or route calls to you;
it surfaces patterns (e.g., âenergy dips after external callsâ)
so you can choose actions that fit your constraints.
If that story raised a specific worry: loss of autonomy, privacy
creep, or community spam, then does the above address it? Iâm
especially interested in whether the âwhyâ behind recommendations
is clear enough or needs to be tighter.
fernly wrote 14 hours 20 min ago:
Questions while watching the video.
Calendar is central, but I use a Google calendar which is important to
me. Connect it?
Seems like a "dlog" is a calendar entry. So is my "journal" broken up
into separate pages, not a sequential document or blog?
2:30 ff, strongly suggest that for your next video you pre-script it to
avoid fumbling and mumbling.
5:10 side note, interesting that your personality(?) model was from
2018, well before LLMs.
7:50 for an app to produce such output (impact of a friend on mood) you
surely must do a copious amount of extremely frank journaling. When,
and in what format? As scattered calendar entries? I'm confused how I
fuel the app.
10:40 relating diary entries (reported activities and attitudes) to
one's stated goals -- this is what I would expect an AI to do, and tell
me about them rather than the reverse.
I'm sorry, I just don't see how I could use or adapt to something like
this when I have a well-established diary/blog and calendar, it would
mean changing many daily habits and adding what looks like a lot of
detail work.
dr-j wrote 5 hours 23 min ago:
Thanks so much for your detailed feedback and for watching my rather
unvarnished intro video with too many ums and ahs!
- Calendar is central, but I use a Google calendar which is important
to me. Connect it?
Dlog will use the default apple calendar which, if is your Google
calendar, will display automatically.
- Seems like a "dlog" is a calendar entry. So is my "journal" broken
up into separate pages, not a sequential document or blog?
Not quite, a calendar event has a title and notes. The title of the
dlog will be whatever you call it, the default is (if your name is
anon) Anonâs Dlog. The notes of the event are where the journal
entry is stored; along with Dlog tags such as goals, journal type,
and sentiment scoring.
- 2:30 ff, strongly suggest that for your next video you pre-script
it to avoid fumbling and mumbling.
Yes, agreed, I made that very quickly yesterday. Iâll re-record it
today. I still want to keep it quite unvarnished though as the HN
mods told me that this is what Show HN community prefers
- 5:10 side note, interesting that your personality(?) model was from
2018, well before LLMs.
Yes, I never dreamed this would be possible until the introduction of
ChatGPT.
- 7:50 for an app to produce such output (impact of a friend on mood)
you surely must do a copious amount of extremely frank journaling.
When, and in what format? As scattered calendar entries? I'm confused
how I fuel the app.
Yes, youâd just journal normally, as you have various important
experiences you can journal about this in free form, stream of
consciousness etc.; or use the guided four rings prompts in the
Journal Coach at the top left of the Dlog entry area. It doesnât
have to be copious, or systematic, because the model is time series,
it doesnât require fixed repeated entries. If you use Dlog for a
few days or weeks Iâd be very interested to see if you found the
responses useful. And again, if you send me a DM Iâll provide a
free perpetual license so Dlog is always free for you to use.
- 10:40 relating diary entries (reported activities and attitudes) to
one's stated goals -- this is what I would expect an AI to do, and
tell me about them rather than the reverse.
The feature at 10:40 relates to just summarizing the entries that
have been added to that goal; it is not related to the AI coach
(which does what youâve stated that you expect i.e. relates diary
entries to activities and attitudes to ones goals)
- I'm sorry, I just don't see how I could use or adapt to something
like this when I have a well-established diary/blog and calendar, it
would mean changing many daily habits and adding what looks like a
lot of detail work.
Apologies, I do think the technical intro video is giving a lot of
behind the scenes background information which may be overwhelming if
youâre simply looking to journal and improve well-being.
I would gently recommend that you try it out for a few days and see
that itâs fairly intuitive to use and well worth the process once
you start seeing the insights from the coach (which get better over
time).
Warm regards, Dr J.
canadaduane wrote 14 hours 54 min ago:
I like the concept, but I bailed at "GPT 5". The only thing that has
given me peace of mind and the ability to journal honestly and
successfully is Obsidian, because it lets me own my data (as text
files).
dr-j wrote 5 hours 50 min ago:
Thanks for the feedback. I love Obsidian. All your data is stored on
device i.e., in calendar entries (a Dlog is saved in your default
calendar, that's why you don't need to sign in to use Dlog; the title
and text of a calendar event are repurposed in Dlog to be a journal;
much like the reminders have been repurposed to comprise projects)
and on the in app on device Rag database, it is not stored in the
cloud etc., so you do own all the data; and it is not shared to the
Dlog server which is only used to track token usage. GPT 5 is
necessary for the moment; Dlog uses the enterprise API, so it doesn't
train on your data. However, I know this is a major concern for many,
so, in the short term I will be adding an anonymiser; and the ability
to approve and edit prompts sent. If you do decide to sign up to use
Dlog send me a DM with the email used to register and I'll send you a
free perpetual license so Dlog will always be free (excluding tokens
for AI use) and 1 million free tokens. Thanks again!
thusjustin wrote 15 hours 52 min ago:
I love the vision. From what I can tell, you're building something that
I think should exist and we have the technology for now. I think we
need a place to put our 3, 5, 10 year goals, and some kind of process
to keep us on track for that. And it's so personal, of course the LLM
aspect needs to be local-only.
One concern I have is that I think I will need more than an empty "add
Journal entry" nudge or prompt. I think I would want what a real coach
would do/say. Something like, "How's the meditation/exercise/calling
friends/making stuff going?"
dr-j wrote 5 hours 56 min ago:
Thanks for the positive feedback! The LLM aspect is using enterprise
API from openAI, their policy is to not train on user data; I've used
it extensively for the last year with my own personal diary Dlog
data; pretty dry stuff to be sure. However, I will be working on
allowing the user to use their own on device models; at the moment
running local models is too memory intensive for most users so I need
to do this once the models are more compact. Apple have recently
allowed developers to integrate AI through their foundation model LLM
but the token window is too short, 8K only if I recall correctly; and
the responses just aren't that useful. The technology will improve
and trust in openAI's systems should increase; we're early days here,
but for now the main mechanisms for LLM we are has to be the
enterprise openAI; there's nothing better IMO. Another thing I am
working on in the short term is mechanisms to anonymise content and
approve the prompt before it is sent; as well as a simple toggle that
when switched on removes that journal from being included in Coach
analysis. Thanks again. If you DM or email me and let me know the
email you used to sign up with Dlog then I'll send you a free
perpetual license so DLog is free to use forever, excluding the AI
tokens; and I'll give you 1 million free tokens. Have a wonderful
day. Dr J.
danpalmer wrote 16 hours 55 min ago:
I'm a bit intimidated by the long list of things this app is trying to
do.
Is it a project management tool? If so, how do I share everything with
my team? Project management tools are defined by their collaboration
and workflow features.
Is it a journaling tool? If so I absolutely don't want my team in the
tool, and so can't use it for project management. How does it encourage
me to do better journaling and build the habit?
Is it a wellbeing tool? How well does that work if I don't put my
project management in there? If I can't use it for half the stuff it's
intended that I will, then it might be of limited use.
Is it a coaching tool? Why would I want to use an AI coach over a
mentor or human coach?
Is the AI required? I have no idea how many tokens I'd need to use
something like this? Do I need a million a day or a million a year?
(When coding I tend to use ~10-50m input tokens per session, will this
cost $500 per day to use?) If the AI features are optional, what is the
product without it?
Overall my feedback is that there's a lot here, and I think the product
needs a much clearer story. The copy on the site is long and rambling
and needs a lot of tightening up. Personality is good, but in
moderation.
dr-j wrote 6 hours 15 min ago:
Thanks so much for the feedback! Below are my responses.
- Is it a project management tool? If so, how do I share everything
with my team? Project management tools are defined by their
collaboration and workflow features.
Dlog is not meant to be shared with a team; I will look at enterprise
versions later on; but for now, a Blog is public, a Dlog is private.
So, the project management definition here has been relaxed to mean a
journal with a reminder list(or lists) organised by goals; the
timeline view allows you to track these.
- Is it a journaling tool? If so I absolutely don't want my team in
the tool, and so can't use it for project management. How does it
encourage me to do better journaling and build the habit?
It absolutely is a journaling tool, and so your team can never view
these; I am working on a feature for you to tag whether a journal is
for work; this will impact the coach responses; and could be used
later on for the enterprise version (targeting this for next year at
some point). In terms of encouraging you to do better journaling,
when you start a New Dlog there at the top left is a toggle called
âJournal Coachâ, there you can get prompts on the 4-rings (i.e.
the Dlog Model at the heart of the app that trains your Dlog Coach is
based around the 4 main constructs, Personality, Character, Resources
and Well-Being); or ask it for feedback on whatever you are
journaling about.
- Is it a wellbeing tool? How well does that work if I don't put my
project management in there? If I can't use it for half the stuff
it's intended that I will, then it might be of limited use.
Dlog is based on the 4 constructs of Personality, Character,
Resources and Well-Being. The Coach will give advice from day 1 after
you have taken the baseline survey. It doesnât matter if you do not
use the project tool and just use it for journaling. Projects work in
tandem with goals as well. I am keen to hear what features youâd
like in the project management side or any other area you would like
to see improvements in. I do recommend that you watch the youtube
video (if you havenât already) I made for Show HN in the original
post. That should (I hope) explain how this works, but it is
unvarnished and 15 minutes long, so I need to get a clearer explainer
about the main story here.
- Is it a coaching tool? Why would I want to use an AI coach over a
mentor or human coach?
The AI coach is deeply informed by the in-built SEM model; the model
I developed during my PhD; based on sentiment analysis scoring of
your journals (including projects, but again, itâs not required).
There are many buttons available in the coach, check out the video
for an example of the output, for example, it can say how to
âStrengthen Relationshipsâ (thatâs the example given in the
video), by looking at which important people in your life over time
have impacted your well-being positively or negatively over time, and
provide a quantitative and qualitative analysis of this; giving
advice in terms of how to improve these relationships and protect
your mood and affect. Any responses you ask it the Coach will always
try to improve your resources and well-being, focusing on the 4 rings
and the Dlog model.
- Is the AI required? I have no idea how many tokens I'd need to use
something like this? Do I need a million a day or a million a year?
(When coding I tend to use ~10-50m input tokens per session, will
this cost $500 per day to use?)
Unless youâve written over 10 million words its highly unlikely
that youâll use a million tokens in one day. 1 million tokens costs
5.99. But if you dm me your email address that youâve used to sign
up with Dlog Iâll give you 1 million and a free perpetual license
so you can use Dlog for free forever (excluding the tokens). Iâm
doing this for all Show HN readers who write me directly.
- If the AI features are optional, what is the product without it?
Without the AI, Dlog is still a useful journaling and projects too.
You can actually use another website I made called dlog.site that has
no AI features, meant for PC users to use; I'm not really promoting
that at the moment as I'm trying to get feedback on the primary Mac
OS.
- Overall my feedback is that there's a lot here, and I think the
product needs a much clearer story. The copy on the site is long and
rambling and needs a lot of tightening up. Personality is good, but
in moderation.
Agreed. Working on it!
Again, many thanks for the very useful feedback.
Dr J.
trogdor wrote 2 hours 51 min ago:
I suggest you practice writing more concisely.
The comment you replied to said that ââ¦there's a lot here, and
I think the product needs a much clearer story. The copy on the
site is long and rambling and needs a lot of tightening up.â
In response, you wrote more text than I am willing to read â even
though I was interested in your answer.
dr-j wrote 2 hours 14 min ago:
Fair pointâmy last reply was too long. Hereâs the concise
version:
⢠What it is: Personal journaling + goals; not a team PM tool
(no sharing yet).
⢠Habit & wellbeing: Baseline + weekly model â small,
specific nudges. Projects are optional.
⢠Coach: Optional AI that summarizes your patterns;
complements (doesnât replace) a human mentor.
⢠Privacy & cost: Journals stay on-device; by default no raw
text leaves the Mac. AI is per-prompt/opt-in; typical use is
modest. (HN readers can DM/email for a perpetual license + tokens
to test.)
⢠Action: Iâm tightening the site copy and adding a clear
âwhyâ explainer.
Thanks for the nudge.
kstrauser wrote 18 hours 40 min ago:
First comment: I freaking love your privacy policy. Seriously. Great
job!
Second: I haven't downloaded it yet because my itsatrap.gif warning
bells are going off about pricing. On a scale of free to kidney, what
are we looking at here? Is this going to be priced for end users, or
will it look closer to an enterprisey kind of plan?
dr-j wrote 18 hours 30 min ago:
Hey kstrauser, thanks for the first comment!
Can you let me know what would reduce the warning bells regarding the
itsatrap.gif? Like, what gave you that impression? Really need to get
this right.
For general users it's free for 14 days with 10K free tokens; then
its 1.99 per month at the moment.
However, if you or HN readers that DM me or email me with the email
they register with, I'll give a free perpetual license so there's no
monthly fee; and add 1 million tokens.
Thanks again for the feedback, I'm glad you liked the privacy policy!
:))
kstrauser wrote 18 hours 15 min ago:
Oh! That's pretty reasonable. It's just that I've seen so many
slick looking tools that go for waaaaaaayyyyy more than I'd ever
personally consider paying for them, like "use AI to sort your CD
collection for only $29.95 per month! For only a cup of coffee a
day, you could have perfect sorting!"
Do you support Bring Your Own Key (BYOK)? If I'm the first to ask,
I'm 100% certain I won't be the last. That a standard question we
get from our own customers.
dr-j wrote 17 hours 51 min ago:
Also, I added the pricing to the home page; hopefully this
reduces those warning bells! Thanks so much, I hadn't considered
that that would be a barrier.
anigbrowl wrote 16 hours 19 min ago:
Should include systems requirements on the download - I got a
grumpy MacOS message about needing to do an OS update to run
it. It's not a big deal but little frictional hiccups like that
results in invisible abandonment.
dr-j wrote 6 hours 8 min ago:
Will do. If you do update your OS and try out Dlog do send me
a dm and I'll give you a perpetual license and 1 million
tokens to test out Dlog.
dr-j wrote 18 hours 9 min ago:
OMG that's ridiculous. I will definitely look closely at
integrating BYOK. Do you have an app that uses BYOK?
kstrauser wrote 17 hours 14 min ago:
That might've been an exaggerated example, but only a little!
I've seen lots of tools add AI as a feature and immediately
jack their prices.
Sure, lots of apps do that. For example, Zed ( [1] ) has a
subscription plan but alternatively will let you plug in your
own key and then provide the same features free of charge.
URI [1]: https://zed.dev/docs/ai/llm-providers
dr-j wrote 1 hour 59 min ago:
Awesome, I'll check that out! Thanks!!
DIR <- back to front page