Post novel startup ideas as comments and see how many karma points it receives. Who knows, one might be developed and you can take credit for thinking of the idea first.
Oh, one more! A web consultancy that advises start-ups about i18n, translation, local laws, cultural differences, etc, so they wouldn't think: "Foreign countries are scary, let's conquer the domestic market first." Then we, the Dutch, could actually have used great sites like Amazon, Facebook, Ebay, and Craigslist instead of the terrible knock-offs that won here, solely because they were available earlier.
Seriously, Europe is a huge market, and if you wait until you're big in the U.S., you'll be too late. It's the internet - the rest of the world is just one hop away!
Here's one I wish someone would take from me and spend lots of time perfecting. I've been trying to get around to it, but have other stuff to do. It's on my "fun weekend hacks" list.
I have a commute that's about 30 minutes each way and 1 hour lunch every day. That's two hours I want to fill by reading printouts of interesting text and listening to interesting audio on my iPod.
For me PG essays would be the ultimate example of texts I want to read and Venture Voice (podcast) would be the ultimate example of what I want to listen to (others would obviously have different taste). I've gone through that stuff repeatedly and tons of other similar content, making it difficult to find enough new stuff to keep me entertained.
A site that let me assemble text-printout "playlists" and iPod playlists for a week or more in advance would be great. It requires measuring or guessing how long a given text (or book) is to see how long it would last and organizing the audio into chunks that are optimal for my given time allotments.
It seems kind of a weird maybe or niche, but it arose from a real problem I have and I think it could be pretty big. There has to be more people like me that want really targeted text and audio content to consume on their commute.
This is exactly what I am building - a website to create On-Demand Magazines from Internet content, like PG essays.
The splash page will be up in a day or 2 and I'll be collecting email addresses and feedback for the beta test.
The name of the site is ShelfMade.net. As soon as I have ANYTHING up I will get the link to YC. BTW if you can do this in a weekend, please get in touch.
WeChangedItToZ.com: a site that translates English-language text written by Brits into more modern and acceptable spelling. "Organise" (which appears to be pronounced "organ-nice") becomes "organize" (so much more modern and cool!). "Colour" (pronounced "col-ooor"?) becomes the trim and becoming "color".
Simple idea, and it's worth $6 billion! W000t! I win.
Macro news. I get the news in my RSS feeds and its damn near useless.
These events tell me almost nothing about the world. The signal to noise ratio is not good. I don't want to know that another 6 people in iraq have been killed in some incident, I want to know whether the situation there is getting better or worse, is it happening in a new area than before, does it involve a new group of people...big picture stuff.
This is the kind of info you typically get from reading feature length editorials but I don't have the time.
I want big picture news and global trends, complete with graphs and visualisations to give me perspective on an issue, delivered to me in an RSS feed.
Also I want to be able to subscribe and unsubscribe to different issues that I wish to follow from start to end.
For big picture news I find nothing beats The Economist. No more kidnapped babies or small plane crashes! I've stopped reading (online) newspapers, and I don't miss them at all.
For subscribing to different issues - interesting! I'd like that too. In theory, you could subscribe to the RSS of the history page of wikinews/wikipedia. You'd get all the updates, but I'm not sure how useful or reader-friendly that is in practice.
To some degree I'm wishing for a new style of journalism. Or at the least journalists actually learning to write headlines that make any sense and are useful.
The startup ideas there are:
- Subscribing to specific issues that you want to keep track of over time is something I want to see way more of.
- Sites that use data visualisations to increase understanding and perspective of...anything really, is something I've been thinking about. Think the world would be a better place if people had perspective and we have access to a lot of data now to play with.
One of the best thing about the Economist is that they create podcasts for certain articles you can put on your mp3 player. Works great for downtime you never have.
Something that is close but it is a print magazine. Is called "The Weeek" magazine. Very Very high signal to noise ratio.
They do exective summaries of the news, one page or less covering issues from all sides.
They just launched a new website www.theweekdaily.com no rss feeds that I can see. It is the one drop everything and read it magazine when the postman shows that I subscribe to.
Well worth the $50 dollar subscription.
Location based gaming: playing games while moving around in your town with an iphone-like device (and having your location matter in-game.) You could port any kind of game to this concept: counterstrike, racing, even monopoly. Or pac-man, chasing your friends around the block. Or Sokoban, pushing giant boulders through Fifth Avenue.
I had this idea six years ago. I'm appalled that it still hasn't happened. Of course, very few devices have GPS chips, and operators are wary of sharing the location of their customers with random companies. But still!
there's one not listed, that's a variant of pokemon; pretty fun
I forget the name (when I saw it, it was only available in japan - though the company is French)
there are already location based hit games... just not in the US.
This will probably stay the same for at least a few more years. This is due to a dif in culture. In most places in the US (aside from NY), most people have room and time to relax at home ( = using wii, 360, or PC for games). Until iphone and treo (and even now) most people don't really use their mobile phones for much... the reasons go on and on...
A better email client - automatic classification of incoming mail, understanding mailing list headers, reminding me to add attachments if I've said "see attached", understanding bounce messages and tracking them against that address in my contacts list ("You're trying to email <foo>. I see that didn't work last time. Please check the address is correct."). etc.
If you make a better webmail-client, make switching easy! Allow me to give your site a try with my current archive of (gmail) messages and contacts, and keep gmail up-to-date, so I can switch back if I decide (and I'm a pessimist) that your service isn't good enough.
I'm a hacker. That's just how I roll. Last night I built a "digg/reddit for pictures" type of site. No sweat. But now I need someone to maintain/market/add content / etc. It's a huge pain, and I'd much rather stick to coding.
What would be nice:
Some kind of site/forum/network/etc that connects hacktrepreneurs with talented and/or cheap marketing people. (sometimes you get what you pay for; sometimes the work is just not that hard)
I don't mean the stylesheets - no, those come later - no, I mean a really good, one size genuinely does fit all, semantic HTML layout. Does blogs, does galleries, does comments, does discussions, in a single well thought out semantic HTML format.
Then you have a library of CSS templates that people can use. And people can upload new templates. And each template gets checked, to make sure that it represents each element properly, and works across browsers (yep, that's a validator).
The business model is this: advertising on the template inspection pages, so that the "greek" text on the layout pages is all advertising.
Advertising is split with the authors of the pages which are downloaded - you copy the CSS to yer drive (downloading the zip file of the CSS + images) and the author of the template gets a cut.
How do you prevent fake downloads distorting the figures? Up to you, but I think you can filter based on IPs.
But the real key is that semantic HTML on the front end, because once you have that, it's easy to do the rest. But that part is a genuine hardness.
Or you could, you know, write the software for this:
This may be more of a pipe dream than a viable idea, but I'd like to see someone come up with a new model for an insurance company where everyone's interests are more aligned. For example, a health insurance company that paid doctors a dividend on the long term health of their patients minus costs instead of per procedure. If it worked, doctors and patients would be the ones balancing the effectiveness of treatments vs costs rather than an insurance bureaucracy.
Note that to prevent doctors from working with only healthy patients, they'd be paid out for improvements in health rather than absolute health.
If you could commit to one insurer for life, your interests would be fairly aligned, (assuming insurers care about their reputation and don't want to pull the plug as soon as cost > premium.). Suddenly prevention programs would be worth the costs. The problem is that you couldn't switch if the insurer turned out to be really bad, or started to jack up prices.
By the way, this is basically how health care works in Europe - single payer 'socialist' health care: More cost-conscious, more prevention, everybody covered, less choice, lower quality for those who could afford it.
I think it's both Joe Krause and Max Levin that both quote the HP founder that, "You make what you measure"
The problem is that sometimes it's not obvious what you measure, since what you might be interested is something vague like "effectiveness" or "progress". And, like all pattern classification problems, if you pick features that don't give a good signal to noise ratio, your assessment could be way off.
Even worse when that being measured knows what's being measured. A feedback loop, if you will. And thus, the fact that you observed something changes it, which sways the results from if you didn't measure it. Sounds a little bit like photons in physics, but I'm guessing photons work off of different principles.
Joel is merely saying that you should be careful of what you measure, because some metrics (lines of code, hours work) aren't indicative of what you're really trying to figure out (progress, effectiveness).
Joel is basing his argument off the book "Measuring and Managing Performance in Organizations." In points out that in situations where the employee gains natural satisfaction from serving the customer, measurement often makes performance worse. The additional effort spurred by the measurement is offset by the distortion to the effort based on what is being measured. However, in the case of doctors, measurement is already taking place, and because it is highly distorted. Thus switching the measurement from number of procedures to increase in health would be a substantial improvement. It may, however, be even more of an improvement to pay a flat fee, regardless of the number of procedures.
Joel talks about software development, where the real impact of individual contributions is hard to measure. Any metric you set will be inaccurate, and thus harmful, because people will be motivated to maximize the metric, rather than real value. You just have to find people you can trust and empower, then trust them and empower them.
Now, healthy days is quite a good heuristic to evaluate a doctor's performance, don't you think? How do you think it can be gamed; i.e., what exploits can you think of?
A social music site / including music client and music store for classical music. Both last.fm and iTunes suck for classical music, which has a completely different set of concepts from pop music. Composers, conductors, orchestras, choirs, movements, etc. It's a smaller market, and the people are less technically inclined, but they're richer and less likely to pirate music.
I was thinking of something like that, but I am not familiar with the copyright laws etc even for test listening music. How does last.fm deal with that? Loyalty fees to the big four?
Almost non-existent. The problem isn't licensing (any more than the current licensing issues they're having). The problem is access to all the CDs.
You read that right: I know for a fact that Pandora gets its music by ripping CDs. No, really. I've been there and seen it happen. Crazy. You'd assume the record companies would send over the digitized files -- no such luck. They have one person dedicated to ripping CDs, full-time.
Anyway, last I knew, Pandora was trying to work out a deal with some guy who owned thousands of classical CDs. So as long as the company can manage to stay in business, Classical is probably coming.
I'm writing a practice app, which was my second idea but I think I have a better chance of getting a version 1 done quickly. (Not planning on applying to YC this round. I may be ready by next summer though)
Basically, it's something about MMORPGs that's never been done quite right. I'm sick of trying to explain how it should work, and want to show people instead.
'Tradeskill' engines don't really reward creativity. With some minor exceptions here and there, they all follow the model of:
(combine X Y Z) -> ItemXYZ, where X, Y, and Z are constants and ItemXYZ is an element in a (relatively small) finite set of predetermined items.
Instead of adventuring to obtain ItemXYZ, you adventure to obtain X, Y, and Z component pieces that you combine to get ItemXYZ, assuming you have enough skill. It feels sort of silly that way. The actual artisan, the guy who combines X, Y, and Z to get ItemXYZ, is not really creative. All he's doing is following a predetermined recipe and applying the skill he got by doing hundreds of mindless combines.
Blizzard's answer to this problem was to shift the focus from item creation to collecting ingredients. They made sure it was fun to gather X, Y, and Z, without just making X, Y, and Z exclusively monster loot. Compared to some other games, WoW's engine was also simpler. Blizzard's solution worked, but only because they sidestepped the core problem.
Tradeskilling appeals to people because it's a creative activity. Players want to be able to feel pride in the stuff they can make. The result is disappointing. Imagination picks up the slack, the novelty wears off after a few hundred clicks, and unless you're addicted to gaining the next level; you aren't going to bother. The way it works now, it winds up only really appealing to people who enjoy organizing things.
From what I've read, Second Life is the opposite. Gamers don't like Second Life, because there's no consistent, professionally-crafted fantasy world to give meaning to anything. It's completely open-ended, allowing for lots of creativity, but what's the point? Obviously plenty of people find a point to SL, I'm just saying this what the typical WoW or EQ or FFXI or DAoC player thinks when they see second life.
I've only heard of two other games with interesting tradeskill engines. I wasn't able to play either of them, so I don't know how similar they are to what I'm going for. One was Horizons, which tanked almost immediately; the other was Star Wars Galaxies, which suffered from being inappropriate for the mainstream Star Wars audience. I say 'suffered' because even though the game was actually doing quite well, it never came near its potential given the franchise name. But whatever the reason, the game was overhauled and the original SWG is no longer available.
That sounds pretty close to what I'm going for, actually. The solution is, of course, to define a set of building blocks that is orthogonal but easy to use, so that clever artisans will be able to combine in many different ways. (like the 25 different blades, with 25 different handles, etc from the article). My ideas are a little different, but mostly in the details and what sort of controls I'm hoping to give the artisans.
A social fashion site. Categorize your entire wardrobe with photos, favorite brand, sizes, date of purchase, photos of you wearing combinations, links to friends, that kind of stuff. If girls would program, such a site would have been invented around 1997 - instead we got slashdot.
I think there is a YC funded company doing this already, I'm sure someone here will remember the name.
My fashion idea involves you being presented with a bunch of different outfits and picking which ones you like to train a recommendation engine. Then, the site recommends complete in-season outfits (with affiliate links)
Dammit, hook that crap up to my closet, so it prepares an outfit for me to wear every morning, instead of having to fumble around for that crap when I just woke up. If I don't like it, I can say, "another suggestion!"
And allow it to order one new outfit every 4-6 months based on which outfits I've been wearing, so I don't have to go to the mall to buy clothes.
i actually suggested this idea 2 years ago to the devs at delicious monster as a product similar to their delicious library (since their product already supporting lending, isight, cool cataloging, ...)... no bites
By the way, why I think this would work is that many people care about clothes, people love organising/collecting/classifying, and extremely targeted ads would be almost trivial.
A program that scours the obituaries to reverse-engineer email addresses and online identities so that people that you've met online can find out you've croaked.
Seriously. That what makes this exercise a little dubious. It's never version 1.0 of the idea. Sometimes it might be version 23
For example, here's v2.0 of my idea.
Identity Life Insurance. For a fee, either use the internet and/or a physical inspection of people's computers after death to shut down online accounts and notify correspondents that you are deceased (and perhaps what to do about sending flowers, etc) This would give the insured privacy in their final online affairs and also make sure the people you chat with daily know what happened to you. Sales would be to traditional insurance companies (which could offer it as a rider) and funeral homes.
Now does that sound better? The fun of this game isn't necessarily in the idea, it's in the evolution of the idea in response to market forces. Cool stuff!
So this would work for people with no relatives, close friends or at least a lawyer? I imagine for most folks someone must know when they died and having some instructions for what to do when you pass away shouldn't be too difficult.
But there are many problems with your assertion that it's just knowing someone who will follow directions:
1) Many people do not know enough technology to be able to track all the accounts a user might have.
2) Many users don't want their families going through their accounts when they die (ie, cybersex and online fantasy issues)
3) Some people may die unexpectedly, with insurance but no directions as far as online presence
4) The average family memeber when left with an unexpected death, even with a list of online associates, will not know how to approach these people (what were their roles in the deceased life?) or what to say
5) It's just a hassle and is easier to have someone else do it
and 6) It's not something people think about a lot, but with the wired-up X Generation and tail-end-of-the-boomers getting older, it will become a more pronounced issue.
Could work for singles too. Instead of a dating site that matches people to each other, it matches people to fun activities and events. Send a bunch of singles or couples to Cocktail-making or salsa dancing class and they've got a common ground to talk about and can still enjoy even if they don't find someone.
though part of what i like about it is the idea of two people using a computer together. the internet is so bizarre in that it connects you to others who are far away, but simultaneously isolates you from folks who are nearby.
Try looking for places where there is quality content and people who would enjoy that content if only they could find it. Where this occurs, there should be lots of potential to generate value, by matching people with content they might like.
For example, last.fm does this for music, del.icio.us/digg/reddit/news.yc for content, and I'm sure there are lots of other examples of services like this.
Here are two areas where such services, to my knowledge, don't exist: web comics and blogs.
No. I just like ideas. I have my own ideas, I just think it would be cool to get a list of ideas that people have but know they won't be able to work on. To prove this I'll start by listing some of my own. PG says one point of this site is to make you look like a desirable candidate for a start up so if you can list some good ideas or critique others it makes you look good. Ideas alone have very little intrinsic value on their own. The best ones are ones that can make things just not suck.
1. An online radio site that lets you select station by what mood you want to feel.
2. A website like 43things but with a timeline for each goals that tries to hold you accountable for them through social pressure.
3. An online video website that are like demo conferences where people can share their ideas.
4. A social network (or app) where users could do market research and product development for companies and get paid for their work and ideas and it would turn it into a competition.
Agreed, an easier to use operating system would be great for most power-users too. I'd start by eliminating the desktop and completely abstracting away the file system, instead letting you browse all of your media files at once, all of your documents, etc.
I'd do it with a dynamic start page that would let you accomplish everything from within Ubuntu/Firefox in kiosk mode. The design's the hard part, but I have the beginnings of a workable design.
That's just cliche. What is intuitive? I'm using a Mac right now. I like it, but I've had endless troubles explaining it to others. It's things like the difference between installed programs and running programs. The difference between task based and object/type based actions. People don't grasp how file systems work. People don't grasp the difference between main memory and harddisk and everything connected to it like saving documents. And don't get me started about networks and login identities...
The desktop metaphor is useless and inconsistent with other things like windows. The desk I'm sitting at physically does NOT have windows cluttering it up! Windows are in the wall, not on my desk. I don't know who had the idea that this is in any way intuitive. My desk also doesn't have programs that are connected in some magic way to my documents. A menu is something I use in a restaurant, not at work or when I watch a movie.
These are not Mac specific problems, but the Mac doesn't solve them either. We need a radically different idea of how a system's complexity scales with its growing capabilities. We need to stop confronting people with that old Von Neumann architeture and we have to stop the attempt to mitigate it with with grotesque inconsistent metaphors.
My grandfather can't understand how his DVD player has more than one disc. He just doesn't have any the technical ability nor does he really care about anything but how to get email. I've tried multiple operating systems and instructed him on computer use. He's just from another generation (he's really old), even though he has an engineering degree from Notre Dame.
It would be nice if there was some OS that would boot up and nicely display like 4 or 5 choices to do; flickr, email, internet, news aggregation, and storage. Have the system handle everything else and they could actually enjoy it.
Yes. I applied and got an interview with YC for what was more or less this idea... among other reasons for our rejection, one was that we failed the flexibility test by strongly disagreeing with Paul Graham's suggestion that the simplified subset of our problem to focus on was an "internet picture frame with email" which conflicted with our stated goals of not trivializing the internet appliance.
I would go so far as to not give the user access to low level storage, but only give them access to their files in the context of specific tasks that used them (unless they were in advanced mode). Email me if you'd like to talk about this more.
It worked out pretty well for my father, who's 80. It probably ruins the experiment that he had been using (well, failing to use) Windows for years previously, however.
How does he do with the mouse? From my experience, people over a certain age just never quite get the mouse. Touch screens are the ideal solution, touch pads would be fine except the ones on laptops are too small, and trackballs are a whole lot better than mice because they let you seperate the action of clicking and moving... too often I see older computer users accidentally moving the pointer when they want to click.
He understands and uses the mouse; his problem with that is not the concept, but tremor. The first mouse he ever used at home was a trackball, but he hasn't had much problem switching to the scroll-ball mouse that came with his Mac Mini.
On Windows, he never seemed to get the difference between situations calling for left click and those calling for right click, so the single-button mouse was a relief.
Touch screens can be good for a lot of things but a desktop just isn't one of them unless the monitor is laid flat on the desk. Having to reach out your arm to the monitor constantly is not comfortable.
I agree. I built my grandparents a computer from scraps and it ran swimmingly, but they never could manage the mouse and the home country just doesn't look the same on a cathode ray tube. They were politely disinterested.
By focusing on simple yet powerful tasks, like photos, news, and email and making the entire GUI these tasks in blindingly large fonts, you can make the experience obvious to uninitiated users.
It's not intuitive. I had to google for stuff like "how to rename a file in OS X", because Apple felt too good for putting such things into the menu for the right mouse button. Windows might be messy, but there is essentially just one rule to remember: when stuck, try the right mouse button. I don't see an equivalent for that in OS X.
I only just started using it. It is definitely an interesting community. There are other projects like it, but CambrianHouse seems the nicest from what I have remember. It seems if nobody else cares to comment on your ideas, they might even throw in their own employees to do so.
Paul Graham argues the fear of looking bad is what motivates people. Based off of that concept I thought of the idea of a social network kind of like 43things with the major differences being:
- you can set a timeline of when you want to accomplish the goals
-people could rate the difficulty of the goals and you have a reputation based on how many goals you complete and the difficulty of them
- could be public or private
-could be more individualized, ie finish essay for English before Friday
What would be great is a way to subscribe to torrents as you do on tivo. So that my shows can be downloaded with out me having to search, Like a virtual tivo, but better. yes, that would be sweet.
TVRss + Miro (formerly democracy) or utorrent, tutorial here http://lifehacker.com/software/bittorrent/hack-attack-get-yo.... You might also want to check out TV-links.co.uk if you prefer convenience over quality. As for a legal option, no idea, joost?
TV-links is far from horrible, click the show you want and start watching.
Torrents are a non-starter for regular users. Download a .torrent file from one of many tracker sites and open it with the client on your computer is simply too many steps and it's too different from the experience on limewire etc.
A temporary phone forwarding service where you can give a number out freely but say "It will only work for seven days." Same idea as the onetime email address but for phones.
-a real estate aggregator to provide people with historical price trends on given property. Eliminate the information asymmetry between the real estate agent and the buyer.
-public and uploadable data warehouse, a la Many Eyes and Swivel, but with better graphic capability, something along the lines of trendalyzer. Design it in a way that rewards contributors with points to build resident 'data experts' and to enable referencing and storing of different people's analysis.
-web-based statistical software like SAS. allow sharing of programs through repository as well as collaborative editing.
Solve whether P=NP, but keep it a secret. Slowly mathematicians and computer scientists will go nuts wanting to know, so one day you hold a conference with an astronomical entry fee. Profit. Easy as pie.
Generic combiner of web 2.0 apps. Not the same as a personalized homepage, e.g. google's homepage. The system would work with any web 2.0 app that provided an API, and would either give users an easy plug and play method of combining the apps to create something new, with easy app discovery; or use AI to resolve general goals. Basically, SOA on a global scale.
this isn't a startup, but i want it: a flashlight on my cell phone. i'm always using my phone's screen as a flashlight in the dark, but it's not very bright and drains the battery.
they've got those tiny, efficient, mega-powerful LED flashlight keychains -- can't be THAT hard to integrate one into a phone.
I know this topic is really already past, but what about a fully up to date web browser. Full CSS standards compliance, JIT javascript, nice developer tools, memory models, full garbage collection, just the web browser we all want right. Pay attention because here is the interesting part. Sell it to Microsoft. This is not a web browser where you are trying to capture the market. No just sell it to Microsoft because they aren't going to fix their browser anytime soon. I imagine they can't. It's too old, and maintenance has become a nightmare. That's why their browser is so behind the times. So, they got a problem, a big one, so someone should fix it for them.
I'd also like to see visual voicemail integrated into gmail or yahoo mail or aol or any client, but real visual voicemail...with voice-to-text technology for the message to be right in your inbox and a click to call back link or button.
i don't know. find a way to take away some of the pain I feel as a computer user:
having to have like 30 different logins for all the sites and communities that interest me where i inevitably wind up using the same password for all of them and a similar username for most of them
feeling paranoid when i browse the internet on a windows machine even when i use firefox, run windows update, and avoid leet juarez d00d sites
being frustrated that the extension language in excel is fine for quick, minor things but doesn't scale up to real software development efforts while the users i develop solutions for have no appreciation of how craptacular VBA is
Hijack that google image labelling game idea and apply it to something like WoW. Basically, make an open platform that encodes computer AI hard problems humans can solve easily into a MMOPRG format.
The google game was kind of fun. Another unoriginal example would be using RTS games for real wargaming. I haven't come up with a good characterization for the general case though.
A social network for all the people building social networks. It's perfect--now they can talk about how they are building the next Facebook to people that actually believe it can be done! It will also allow you to import all of your existing social networking profiles into one place. Widgets. Video. Wikis.
Sure there is room. It's just that if "social networking" appears in the project title there's a high probability that they haven't figured out what that room is. I'm sure there are a lot of great new ideas that include some kind of social network, but have something more interesting to say about it than that it is a social network.
See for example this discussion on using lifecasting to encourage good behavior in a world where few people are genuinely religious (you could also combine this method with a traditional religion):
Well the problem with the good behaviour idea is that governments are not waiting for volunteers to start lifecasting. They simply put their surveillance cameras everywhere. This is already happening in Europe, don't know about the US. Big Brother is growing...
Another interesting question is what kind of good behavior would social pressure evoke? Would people use the same yardsticks for themselves as for other people? Or would they just try to hold society in check, so that they themselves could roam free?
Seriously, Europe is a huge market, and if you wait until you're big in the U.S., you'll be too late. It's the internet - the rest of the world is just one hop away!