- Buy some recent high-res data, maybe from DMCII (http://catalogue.dmcii.com/). You can get relatively recent (month old) data for very cheap. Only issue is to find high-res enough data. - Link the image to a gg map / OSM equivalent for geocoding - Create mechanical turk or taskrabbit to search for pools / houses without pools
Do that for one neighborhood and see if you can make a sell. No automation needed, or at least no complex CV involved.
I have used their xml files previously to automatically tile roofs with solar panels.
[0] http://googleenterprise.blogspot.ca/2014/07/bringing-google-...
This would make "budgeting" very easy. Have a "food" account, an "entertainment" account, etc. Do weekly or monthly budgets by transferring money into your mini-accounts, and denying transactions for each account when it goes over budget. (Or let the transaction go through from a backup account, but notify you that you went over budget.)
Also, have an API that anyone can write apps for.
Of course, I'll never do this because starting a bank is really hard.
Anecdotally, I do much of what you describe and such a system is nearly my most valuable financial asset.
For that and other reasons, I think the proposed system is worth exploring.
I will say though, I'm a very happy 360 customer.
Many of them around today pre-date the bigger banks by some years.
Still seems like a massive opportunity. $24 billion market in the US. Inconvenient locations (for many people). People have (collectively) a massive amount of under-utilized space. Not without its challenges but neither was AirBnB when it started.
I've also had the idea of an AirBnB for 'toys' (i.e. sports equipment, motorcycles, bikes, instruments, etc).
There is already RelayRides for cars, and qraft is probably the closest existing site to my idea, but I think it could be executed much better than qraft.
- Tablets for seniors: when the elderly population sees an iPad ad, they're not captivated or entranced; they're intimidated and disappointed that they're left behind by technology. I envision the "jitterbug for tablets" -- built on Android with big, tactile buttons; a 'never get lost, take me home' feature; remotely controlled functionality (IE turn on/off apps); etc. They wouldn't use much bandwidth, so you could build 3G right into the device and charge a significant monthly premium -- after all, it's a dramatic quality-of-life improvement for someone sitting in a retirement home.
GE and a few other companies are doing similar projects, but no one is really executing all that well IMO. Problems: would be super hard to get off the ground / defend, and the market is becoming increasingly obsolete.
* based on IE7 so a lot of websites don't work/render properly. This does not seem important when the core of your users go on google/wikipedia and that's it. But my grandmother loves to order online but is limited to amazon because most of the other websites don't work properly. She would gladly pay more to be able to simplify most of the main websites UI for her and ease her order flow.
* Impossible to have more than one tab. This seems logical at first because it simplifies everything, but when she is ordering something as a gift and wants to look for an address she has to close the whole application and go in "notebook" and start over again. Same with the idea of having multiple tabs in IE.
* Email services really limited. Their idea is to avoid spam you have to whitelist every email address. But this is not negotiable, so when you order for the first time online you can't use your system's email because you'll never get the "confirm your email address". Or you use a gmail account for this which immediately make things too complicated for her.
* Most of the other features (games especially) are really limited/dumb such as having to recognize a banana. Creating games for her is something I'm planning to do but with the old browser inside the machine it makes things harder.
* Pictures management! We all send pictures to her by email but she can't organize them at all. I've been thinking of a UI for this but when I draw it/explain it I immediately see she would be completely lost. She needs a simple UI on top of explorer of something that would do "delete, crop, move, rename, add comments". And on top of that the main feature she needs is "print it for me and mail it to me". Websites already exist for this feature but she would have to understand how to upload photos and then use them which is clearly impossible.
* Keyboard! Another big issue for her is being able to raise her arms high enough for a long period of time to type an email on the screen with the tactile keyboard. At the same time, the keyboard delivered with the computer is not in alphabetical order. I've been trying to teach her but it takes her a long time to write so she mostly gives up and limit herself to a few words!
Hope this helps!
Some awesome friends of mine at Chapman University have been building exactly that: a tablet for seniors.
Tangentially relevant: watching a nontechnical senior citizen use an app I worked on was a humbling experience. Things that seemed obvious to me were disorienting and discouraging, even after patient explanation that could never be offered in production.
It doesn't let you check out from within the cart itself (is that really necessary? If that is what you need, why not just create an app like the Apple Store has?), but it does let you use the same cart both in and outside the store. I bought mine since it makes it a lot easier to walk my groceries home.
Was thinking more along the lines of direct to fridge and freezer options. Ideally with cooling or refrigeration properties so cold food can stay cool on the ride home, and the whole thing can pop into the fridge.
1. http://raintown.org/lava/ 2. http://www.cs.bham.ac.uk/~singhsu/
There's also PSHDL from Karsten Becker[2] at TUHH. It's immature and seems to be focused on teaching at the moment.
[1] https://chisel.eecs.berkeley.edu/ [2] http://blog.pshdl.org/
This lets the restaurant increase their profits by serving more parties through their tables at peak times - maybe 10-15 min per table that uses the order system.
Some variant of this might also work for busy bars too.
They don't do while you wait ordering AFAIK. Well maybe they do if you "wait" in the bar!
http://www.paywithcover.com/ New York
http://www.tabbedout.com/ Austin
Open table is also doing it in select markets.
Possibly a week-long summer camp, weekend hackathons/fitness outside major cities, weekly runs or bicycle rides.
- a "Rap Genius" for crowd-translating doujin manga (kind of exists on danbooru but not quite)
- A collaborative gaming site for pen-and-paper RPG and boardgame players which would let you design and run your games as a virtual representation of the physical game (probably exists or else is a bad idea)
Adding the support systems around this type of table, like tracking stats for various RPGs, could be a way to dominate the market. Most of the other solutions focus on the stat-tracking externals, and only have a very simple table or rigid grid, not in keeping with how real games go.
[0] http://www.berserk-games.com/ts/ - this link shows board games but there's a user-created library of game pieces[1] and some built-in RPG figurines.
I have heard Roll20 is pretty popular and fun to use. I have never done it myself but it seems like a really cool way to play.
Now I just need to be able to not be the last to have it.
That being said, it's a great idea and I would love to work on this one day.
Doesn't seem to really take off - 6000 users in France after 3 years of operations (data from a french article a year ago, not sure if reliable), and their FB page has last posts about a year ago too - I would say it's dead now.
I'm gonna take a look why it didn't take off.
Btw, the same idea in UK - Cookisto.
Secret exists to confess anonymously within your circle.
General social networks to connect to your connections (FB, LinkedIn, etc).
An insane list of failed startups that try to get strangers to connect but no one actively seeks meeting strangers for the sake of in general.
I am having a hard time seeing why I'd want to connect to anyone anonymously in my area that don't fall into the above categories but I'd love to hear why you think it makes sense
Although I guess you could do it on github if you had technically inclined people doing the traveling and wanting to post about it.
Http://hopon.com
Then you simply specify: "I want to make dinner, what can I cook?" The app links you to the recipe and any videos for making that recipe right to your device.
Maybe a smart refrigerator will figure it out some day or Amazon with their automated service/scanner combo. It certainly could prevent food waste.
I have a loyalty card for my supermarket so they'd know almost everything I buy. If they made that data available to their users, this space might be more accessible.
The problem is that you need to reproduce the local Excel environment, or the local video rendering environment, in the cloud. That is either something you do on a per-vertical basis or perhaps with some very advanced system administration tricks.
Also, video is probably not a good candidate, because it's data heavy and relatively CPU light. The good candidates for cloud offloading are lots of computation on small data, which does not fit the class of things that most people do on their desktops these days.
I asked some commercial video friends about this once and they didn't seem too interested. Their company invested in a private network for their office and a lot of CPU muscle.
I have two specific applications for VR: the metaverse, and really good porn. (These do not need to be combined though I suppose they could.)
I would argue that Second Life failed mostly due to execution issues. I'd love to see a virtual world where I can socialize, and where I can build cool spaces to hang out with my friends / hold meetings / work.
Phillip Rosedale, the guy behind Second Life, is working on a next-gen equivalent. Janus VR, meanwhile, is busy working at turning the Web into a metaverse, and getting some rave reviews.
As for porn, http://www.reddit.com/r/oculusnsfw/ is your source there... "Really good" porn is, of course, highly taste-dependent.
You'd make a killing.
I can't really see myself ever doing this though because you'd got a chicken and egg problem, plus it'd probably only ever be used by tech companies unless there was an easy to use API.
We aim for the cards that will be useful post-rotation as many card sale sites do an okay job estimating release price. Also, we mostly played legacy.
Snapcaster for instance was ~$25-30 on release (although some places sold it for as low as 15). Then it was ~20-30 for a while until it saw a lot of legacy play and jumped up. It is currently 30-35. ROI probably similar to an index fund.
Tarmogoyf was probably our biggest 'buy' since our legacy staples. We didn't buy that many (5 play sets). And sold as soon as the price hit $100-120. This we risked due to rumors of 'modern'.
Force of Will, a legacy staple for a long time. 8 years ago we paid ~$10-15 per FoW. Now they're going for ~$90. The S&P is up ~70%, FoW is up ~566%. We bought 20 play sets (80 cards) for roughly $1000.
We're still holding many of our original collection. But, sold off enough to get our original $10k back. At this point the price keeps going up, many cards fall out of circulation (destroyed/lost/etc). The only "risk" is reprints.
One issue is unloading many cards. With stocks I can create a sell order and sell it for roughly asking price. Whether I'm selling $100 of shares or $100k of shares it will go through instantly. With magic it isn't hard to deal with ~500 cards or so. But, once you scale up it is a full time job. If we bought $10k worth of FoW vs $1k we'd have 800 FoW to try and sell at some point. This would take a long time.
You either pay SCG prices to get cards quick. Or you use ebay for a 10-20% discount. But, takes longer and no guaranteed sale.
As far as logistics we each put in roughly 5k and split the original buy cards evenly. Since then our collections have fluxed. He bought/sold a Black Lotus at one point, I never did. He bought individual power 9 cards and sold a 'power 9 set' (5-10% markup) a few times. I was more passive.
A friend of mine was a full time magic trader, and he hired a kid to sift through mountains of commons that he bought in bulk.
Basically I hate maintaining an AWS instance that is idle most of the time. And Heroku will only get you so far.
It would probably be difficult to make profits on because you'd be charging slices of a penny at a time...
Percona has a remote DBA service that gives us 24/7 access to a team of proficient DBAs for a fraction of the cost of hiring one.
I'd like to see the same product for cloud sysadmins.
Popcorn Time for quality children's programming - Bill Nye, Mr. Rogers, Sesame Street, Avatar. Shows that are entertaining AND educational, none of that advertising filled, sassy attitude, Disney Channel crap.
It seems like you need a lot of money and experience to build a successful kickstarter?
What if there's a different service where a person can put up an idea with minimal cost and flare, and just raise enough funding to hire a video team, PR team, etc to then run a kickstarter campaign?
Through the app, you can apply for warranty , see when it is expiring etc.
http://en.wikipedia.org/wiki/Magnuson%E2%80%93Moss_Warranty_...
<br/>
Granted, the movies and shows on Play have at least a context to do the pattern matching with - a cast list greatly reduces the possible matches.
I have boxes and boxes of tapes, disks, notebooks, books, and photos. All of which I'd prefer to have digital. Other physical things could be digitized with video, photographs, and scans. I would take them to my street corner and a driver would pick them up and deliver them to a digitizer. They could show up in the cloud a week later, or be delivered back to me in a hard drive.
Once the digital copies are received, the customer can request to have the goods responsibly disposed of. The service could also cover long term storage for customers who want the originals to remain intact. Possessions can be returned to the owner within a day or two's notice.
A service like this would allow people to minimize their physical existence while preserving the memories associated with physical possessions.
I ran into a big problem: how do I label/organize this mess of both physical/digital memories? What are your thoughts on that specific part of the problem? How would you solve it? More importantly, how would you automate it?
There is actually a huge potential to improve the worker side of the current status quo. Right now, these guys have a ton of idle time[4] and there is a pricing opportunity[5]. There's a lot of other opportunities in this, I've been mulling it over for the past few years, I can build it just don't want to market/grow it[6]
[0] for projects that I don't necessarily need a contractor for [1] or, work that a regular contractor doesn't do, like; my lawn guy doesn't clean gutters, my housekeeper doesn't clean windows [2] taking them home is worse - they're probably dirty & sweaty and I don't that in my car [3] can't have high expectations, these are generalists don't expect them to do high quality tile or carpentry [4] they sit in front of hardware stores for hours just waiting, some days they have no work, there is never a guarantee of work [5] they would take less money for guaranteed work, they could build a reputation and charge more for work, they could have their own transportation; saving the buyer the hassle of playing taxi service (often the buyer is a contractor, not a lone home owner like me) [6] if you do, let's talk
An "eliza bot" like service that doles out Freudian dream analysis when you tell it about your dreams. Maybe even combine a logging service so you can log your dreams (ala the way some people keep dream journals).
I really have few ideas for monetizing the thing, I mostly just think it would be fun to do. But possibly you could do some cool targeted advertising based on the "discussions" you have with the dream-analysis-bot.
Instead of just makers saying what they can do it would be based more on what people say they want, and attach a value to say they'd pay a certain amount for it. You could also declare that given x, y and z you could do a, b or c and thus giant chains could be resolved.
All a bit GOSPLAN like though.
I've also considered Tinder crossed with auctions: bid according to how hot you think they are, with highest bids getting to meet (and pay!)
You open an app. You're played three different pieces of music. You're asked which one you like best. The program branches and plays you three more pieces of music. Again they're different but from a more similar selection.
At each point you can highlight bits of music to go back to - to buy that music or to start the chain from that point.
One example would have this tightly connected to one particular publisher's catalogue.
It would eventually teach about music, giving comprehensive sleeve notes about the composer or the piece of music or the history or music theory or etc.
The system will have dedicated "containers" on corners or squares where you drop your garbage and an underground network of carts delivers it to a central point where it can be picked up. Or even to the dump if possible.
This can be fully automated and I think will save a lot of money in manpower and gas (for driving the garbage trucks through the whole city).
I've made a naive summarizer that seems to get the job done for the summaries part: http://breue.com/summarizer
I'm just not sure what the final product would look like or if there would be enough of a reason for people to prefer it over Pocket.
For example, I do save articles I can just read. For that, a summary service would be great BUT that is true with or without Pocket. On the flipside, if I saw stuff like a blog post on technical code someone wrote about for something, more than likely I want to save that reference for later (true for a huge chunk of stuff I save or the type of things I save).
That said, I came across several people who have either worked on summarizers or talked about it but nothing seemed to have ever took off on those. I can't speak for others but in general I prefer to read content for myself.
Yes, it would probably be more expensive than people's current policies.
The neiches are small. One for example is model engineering and related subjects. Books with plans, drawings etc. Construction methods.
In CeeBot you learn programming concepts and whole languages by writing instructions for virtual robots. You can see how they move around and perform different tasks. (Actually Mehran Sahami from Stanford teaches the programming methodology course with a little virtual robot named Karel with exactly the same principle).
The player in my game would be a spacecraft captain. But unlike in other games, where you just press a button and the vessel goes to any direction, this ship had been hit by a meteoroid and its main computer is broken. Therefore all the commands have to be done manually and any computation is performed on a piece of paper and just put into the command line.
There could be no graphics at all. Just the roar of your enginges.
In the beginning the tasks are simple, but the more you play the more complicated the calculations become. It begins with simple arithmetics and trade. Later you need trigonometry to fire a "torpedo". It would be great if you could progress it even further, with advanced math and phisics, and also chemistry - you need to combine different substances in order to burn them as fuel or to produce oxygen to breathe or combine nitrogen and carbondiaxide in order to grow food in the farm.
It would be great if instead of taking tests the teacher would just say: "John, you are still on level 8, you should go to Alpha Centauri and fight with pirates. Play more!" - which would mean - learn to solve problems with two unknowns and calculate volume of spheres.
And imagine a multiplayer with students on the same level who have to make accurate calclulations fast because without it they would just float in the dark and cold outerspace.
I will never make it - I don't know math and programming that well - but I'd play the game!
Example: My AC has its own remote. With my device I can record the wireless signals it outputs, similar to a garage opener in a car. Then with my mobile app I can create my own interface to power on/off, set temp, etc. I now use just my smart phone to turn on my AC. I press power on the custom UI I created in the app, it will send a packet to the hardware in my LAN, and that hardware sends the matching wireless signal to the AC.
Some things might still need physical fittings and cannot always be wireless devices such as power switches. New power splitters with this functionality would also be a good way to control appliances that just need to power on and off.
Dawkins' book "the Blind Watchmaker" introduced "biomorphs". These are intended to show the power of repeated random change and selection. He uses a small computer program to draw six images composed of lines. The user selects one and the program redraws another six images, making small changes based on the image the user choses. This is repeated many times.
The new version is pretty similar except it uses Lego pieces instead of lines.
It's kind of like a parking lot for business opportunities I didn't go through with.
The idea is to catch people and entice them to eat differently right before they make the decision to eat fast food.
Fully automated and fully mobile.
How bout this idea..
>> already exists, here is URL
