Ask YC: my just launched startup needs your comments | Hacker News Reader

Ask YC: my just launched startup needs your comments

20 pointsbtw033 comments

There are so many times when I wake up in the morning after a fantastic dream during the night I feel so deep a need to write the dream down, it's just a pity to let it pass away, it can get forgotten in less than an hour. So I code this tool just for that, keep your dream journal.

http://keepdream.com

I am a one-man team wearing so many hats, currently a graduate student in Shanghai.

Give my startup some comments or suggestions. Thank you.

View on HN

That's cool, but the YC app question about "What do you know that your competitors just don't get" applies here. The core functionality seems twitter-like (and with a twitter API, you could get everyone's twitter dreams displayed on this site too).

You need something that is unique and distinct about this dream journal compared to the world of microblogging. I do, however, like being able to read about other's dreams in Chinese though.

I disagree. Every startup, except a few, can only hope to target maybe 5% of anyone who ever stumbles on the site. So, the only opinions that matter are those of the 5% who sign up and use the service. Everybody else is just thinking up reasons that this is either not needed, takes too much trouble, already has a better alternative, too busy to look into the site further and then forgets about it, doesn't feel the site is safe or secure, doesn't feel they know the author, think it costs too much money, thinks it costs too little money, or thinks the idea is too simple and therefore must be put down because, who do these people think they are, writing free web apps for the public--Mother Teresa? They can do the same in three seconds!

So, let's say someone creates a service that creates an online store application only for shoes (let's call this shoestring) compared to an online store application (lets call this Viaweb).

Although, the 5% who sign up do matter in determining Shoestring, it still matters that this application have something that bests Viaweb, whether it is a better interface for shoes, or whatever the one key ingredient is in differentiation. All I'm saying is that this app should have that key ingredient. It may be better as a FB app that can be sent to your friend ("hey, you wuz in my dreamz thinking of foodz").

Isn't there a difference between startup and random web-based tool?

I think it depends on whether some "bubble" has popped or not.

One Google ad.

In all seriousness, even those of us who are veteran native speakers of English (which our Shanghai-based contributor may not be) sometimes have trouble with the distinction between a "startup company" and "a fun new website". Entire issues of Fast Company and Wired have been known to miss this distinction.

This is a charmingly simple idea and the tool is correspondingly charming and simple. And, for the moment, blissfully free from ads. Well done.

Thank you for your praise. :)

I actually think it would make a viable business if there was some integrated astrology/fortune telling aside of the dream. Horoscopes are huge businesses in Asia. (Of course, you'd have to know how to produce horoscopes)

>Of course, you'd have to know how to produce horoscopes

It's easy. Pull a bunch of random, vague shit out of your ass. I bet someone could write a wicked horoscope generator, kinda like the postmodern essay generators: http://www.elsewhere.org/pomo/

Thank you for your interesting idea.

Let's leave the hating to Techcrunch's commenters.

It's not "hating", it's called productive criticism. When I ask for a feedback on our product the thing I fear the most is a bunch of OMGs. I've seen that before: people say nice things because they want to appear nice, gain karma points, get rid of you with your questions, whatever.

What I value the most is some critical thinking, harsh even, because nothing is "great!", everything we do typically sucks until a few iterations.

"Isn't there a difference between startup and random web-based tool?" smacks of sarcasm to me. Of course I value constructive criticism, even complete dismissals of an weak idea. I'm just turned off by the schadenfreude seen on web-industry sites.

Too Twitter-y. You need clever fresh terminology (i.e. not "Following") and a slightly different design. It looks good, but I get an "I've seen this before" feeling. If you're targeting English and Chinese speakers, it could be cool to incorporate that into the design, probably to the extent that non-Chinese speakers can safely parse the characters as "icons" though.

You definitely need something that makes it standout. Why shouldn't I just write down my dreams or post them on my blog? I could see some cool text pattern recognition features encouraging me to post a dream; that is, who else had a dream like this last night/ever? You mention natural language processing, so perhaps you've already got something like that in mind...

Being able to reply to individual dreams might be annoying, though maybe we could post "interpretations" (in the Freudian sense).

Perfectly twittery. Excellent user design. I love it. All you need to do is 1) spread the word to non-developers and 2) solicit feedback from registered users (most important source) as well as those who aren't ready to register yet (a feedback box.) I wouldn't use it because I don't remember my dreams. But I would have a starry night background, with some purple in it, like the background of the Mac OS X Leopard desktop.

>Too Twitter-y.

I like the twitter design, can't design a better one. So I mimicked the interface, it's clean and straight.

>Why shouldn't I just write down my dreams or post them on my blog?

Had this question in my mind from the day 1. I am working on a dictionary like database, the only difference being the explanation would be interpretation. Then it should work like this, for every word when there is a match in the database, the word become a tag and the word/interpretation pair is displayed below the dream text. The only difficulty is building such a database.

>Being able to reply to individual dreams might be annoying, though maybe we could post "interpretations"

I agree, interpretations make more sense than comments here.

Indeed, I was thinking almost the same thing--I'm reminded of [that music app I can't think of the name of at the moment] where you sort of explore a "web" of songs.

This could be quite similar, but with a "cloud" of dreams (heh) with a sort of haunting, serene element, perhaps with vector animations of recurring nouns floating about in silhouette... for convenience and impact, it would ideally be a screensaver. At any point, you could hit a certain key (probably Enter or Space) and instead of dumping you back to the desktop, it would pop up a text editor for entering your new dream, which would then float away into the "cloud."

I could even see adding a sort of MMO element where each person gets a little avatar that slowly explores the "dream world" for them, being moved to a new dream's location on the cloud whenever one is entered. The little avatars could bump into one another and tell you in the morning of who you "met" in your dreams.

This is a pretty cool idea. I guess that your success will depend on whether or not it's an interesting place to go to read about other people's dreams, though: in most communities there is a real gap between personal use and people who contribute.

So you should allow people to sort by language, for example.

It is much more in the nature of a social experiment than a tech startup, but I think it's pretty cool.

Twitter, delicious etc all haven't got this feature, but it's great idea to try.

I'd recommend adding a simple client, that syncs with the site. And think about the situations under which someone wakes up:

Are they definitely going to have internet access? Nope. Better have an offline mode.

Is their computer going to be on already? Maybe not. Better give it an automatic start-on-startup option, and make sure it's lightweight enough to open in a flash.

Are they going to wake up with their computer around them? Probably so, really, or at least the majority of the time. So a desktop client's a good call, because it doesn't have to be portable, and you can store detailed settings, because they're not going to be changing them over multiple computers.

Also, you're a grad student in Shanghai? Wouldn't an SMS interface be well-received? I sense potential twitter partnerships!

If you want to talk about this anymore, I hang out in #startups on irc.freenode.net a lot. I like your idea, I'd love to help you brainstorm more.

>Think about the situations under which someone wakes up

I haven't think about it, accessibility is essential here. Thank you.

>Also, you're a grad student in Shanghai? Wouldn't an SMS interface be well-received? I sense potential twitter partnerships!

Yeah, I am in grad school in Shanghai, everybody gets a mobile phone here, but I wonder if SMS is too cumbersome for dream input, after all it's a long text to type with the tiny keyboard.

>If you want to talk about this anymore, I hang out in #startups on irc.freenode.net a lot. I like your idea, I'd love to help you brainstorm more.

Thank you for your interest, I definitely would contact you when finish my paper reading homework. :)

Kind of cool... you probably want to add some things like tags so that people can look through them, and maybe also filter on languages.

It's a good idea, but I plan to add some natural language processing to auto tag the meaningful nouns/verbs in the text, I wonder if this is a better approach than adding tags manually in a dream describing text.

Tags would be much too unwieldy for this. Keep it very simple. How about a short-list of categories selected from a drop-down? This would need more thought but a very short-list like: 'falling', 'lucid', 'nightmare', 'premonition', 'embarassing', 'funny' etc.

Nothing more is needed and it makes browsing by category fun for the random surfer.

I agree, a drop-down is great in the post page. Thank you for your suggestion.

Kind of cool. I could see it being a bit hard for the casual browser to find dreams of interest to them. I've seen others suggest tags or categories, either of which could prove useful. But you might also want to include a simple reddit-esque up/down vote for dreams people liked reading. That way people can just view the most interesting dreams.

Great idea, a karma/vote system and a popular link are necessary in such a site.

If you're going to be displaying Chinese content for an English audience, you should look at integrating some sort of translation support. I'd recommend the Adso plugin:

http://adsotrans.com/blog/adsotrans%e7%9a%84%e7%bd%91%e4%b8%...

Once users click on text the content gets semantically analysed and annotated with pinyin/english popups.

I like it.

Don't expect to get huge, but this is a cool app for your portfolio and if you start pulling huge traffic you could make a couple hundred a month in ads.

Try making a facebook app with the same functionality? Sometimes those get lucky and get huge and making facebook apps is a decent skill to learn.

The revenue model is to tell people what the dreams mean :)

I will have a carefully thought on this revenue model. ; )

Have you checked this one out? http://www.dreamjournal.net

probably doesn't hurt if you try integrating this with facebook/open social

maybe some icons that quickly classify (like right next to the title) dreams would be good too (nightmares, happy, sexual, weird, mundane, ....) (in addition to tags)

Reply on news.ycombinator.com