A peer-to-peer Web browser. We let users create and host unlimited sites from their own devices. We think the technology can support thick hostless applications, which eliminates data silos and makes every application forkable from the browser. No blockchains, just an improved variant of BitTorrent called Dat [1].
We've been quiet the last 2 months but our 0.7 release is coming up, and we're excited about it.
I loved the premise of Unite; it was a home HTTP server with dynamic DNS via an opera-run service. Very cool idea, but it wasn't that good of a product experience. (Maybe somebody should try it again using Electron; "node in your browser" isn't that bad of a pitch.)
But, one of the big problems with a bunch of home HTTP servers is availability. The crypto-p2p network lets you push to a public peer and solve availability and improve transfer speed, without having to change the URL. That's a pretty big deal when you consider how unreliable the edge network is.
The other really key difference is that P2P removes the service dependency; sites/apps sync to your device, and then use device APIs to get work done. It's a different view of the applications model, and it means recipients can modify the application after download. This is a really big part of our focus; all sites and apps are user-modifiable.
Zeronet made decisions with their technology stack that I disagreed with, but it's in spirit very near to us, and I've always loved the polish of their applications and demos. P2P is pretty competitive right now. Some projects involve blockchains, some involve BitTorrent, some involve IPFS. We use Dat, and there are technical reasons behind why we chose Dat, but in the end this space will differentiate by the end-user experience.
That'd be really interesting, splitting social media control out of the existing silos. The problem with that seems to be that when you try to read someone's profile and they're not online, it wouldn't work - what would happen in that case?
An out of the box platform to make deep learning easy. It has a drag and drop interface, connectors for common databases and blueprints for common tasks.
SignalBox is being used for: Forex trading, Genome visualizations, customer segmentation, large recommendation systems, fraud detection, realtime anomaly detection, bot detection and geo-dataset market prediction.
As well as the general platform I'm getting some much larger customers onboard where we build out their entire data environment - HDP deployments alongside SignalBox instances work great.
We are profitable, have some great customers on board, and are growing rapidly - and most importantly making our customers happy! :) Please reach out if you would like to work together!
You get a phone call every night and record your 60-second response to the question "How did your day go?". From there, your response is archived into your private online journal and displayed alongside your photos, twitter posts, check-ins, etc from that day.
Soon, you'll start getting Flashback emails (ex. "Here's what you were doing 6 months ago") with all of those cool things so you can reflect on your past.
We want to help people stop forgetting 90% of their lives.
"We want to help people stop forgetting 90% of their lives."
I also saw this tagline or similar on PH, and I got to say I'm not a fan. If you look at everyday on it's own, for most of us, it's pretty boring stuff. Why would I want a record of that? If something hugely important happens, sure maybe I'll want to be able to access that memory / my thoughts at that time in the future.
But 10 years (let alone 50) from now, I might not even know what year big events happened. And so I'll be digging through maybe 100s of hours of recordings/text if I'm looking for a specific event.
There's definitely some product in this space that could be great, but I don't think 60 second recordings is it, especially without some way to filter through the boring crap. I imagine looking through my old calendar would be more helpful than my 60 seconds of nightly drivel.
Of course most of it will be "drivel", but it's _your_ drivel, the drivel that your life is made up of. For most people that _will_ be interesting, or at least emotional, as they look back on life later on. For starters this group includes everyone with children. Just think of all the "pointless" photos we take of our children, they mean so much to us.
But apart from this, very soon (if not already) it will be possible to transcribe these messages and extract some useful metadata from it, even put together a kind of biography. That will be interesting even to sceptics like you!
Also consider journaling, many people, that live relatively uneventful lives, do that. This sounds like an alternative for them.
I don't journal, but from what I understand, products like the 5-minute journal ask you to journal some things you are thankful for, something positive/good that happened today, what you could have done better, etc.
This exercise seems more about focusing your day/priorities, clearing your thoughts and taking some time to feel some good vibes. It's not about remembering 100% of your life.
Yeah - If I'm going to look back on life later on at things, it will be 100x in favor of things like my kid's photos/videos vs words I spoke that were transcribed to text.
We're building a new type of wind turbine that generates energy using significantly less material, making it much cheaper to install. We do this using huge fixed wing kites made of carbon fibre, and drone technology.
There's huge interest in renewables here in China and a great deal of windy Himalayan mountain peaks that have been covered with wind generators here in Yunnan province (far southwest China) in the recent past. Also solar and hydro is everywhere. Lots of e-vehicles too, everything from golf buggies to cop cars to buses to taxis to bicycles to cars to old-person mobiles to tractors.
Also coming from Australia with Scottish heritage, I wonder if you could use this to power long distance autonomous e-vehicular trucking through remote high wind land areas?
We also have awesome high altitude lakes for kite surfing (Fuxian in particular), I used to sail Weta trimarans there, but sold them last year. Motor boats are banned. You guys should come visit!
+Wiener's Eighth and Final Law: You can never be too careful about what you put into a digital flight-guidance system.
We're actually in talks with a fellow Scandinavian that lives in Yunnan right now!
When I lived in Shaolin a long time ago I did spend some time exploring Yunnan, the mountains are amazing, some of the most stunning views I've ever seen. This was more than 15 years ago and there seemed to be very little interest in electrical vehicles or renewable energy, I'm happy to hear things have improved!
SkySails have been exploring pulling large ships with kites, but I don't think it's very practical for trucks, then it makes more sense to go via electricity. Apart from the impracticality, there are also pretty subtle reasons of efficiency. They did use kites to pull vehicles in Ancient China though!
This is an awesome idea. I remember reading about kite-based propulsion systems for large ships a few years ago, do you see any overlap with that area?
How do you deal with the possibility of the kite breaking free of the tether, or breaking up in the air? I guess it could be quite dangerous if there's a populated area downwind.
There is definitely some overlap with SkySails (as that project was called), particularly in the control software part.
But as far as kite power goes the principles are quite different. They use the pull (or "lift") of the kite to directly help propel the ship, while we have generators installed on the kite and harvest energy that way. This is called "drag mode" kite energy, vs "lift mode" where you let the kite pull on the tether and let it reel out and thereby generate power (or pull a ship).
We haven't looked into it but I have a feeling it would be more efficient to use drag mode even on a ship and generate electricity that would be fed into a hybrid engine system.
The scenario you mention is not really a major concern, for several reasons:
First of all the tether is of course designed to withstand all reasonable situations, with a wide margin of safety.
Secondly, the way the kite normally lands and launches itself as the wind drops and returns is by flying as a drone using the rotors-generators (see the website), so as long as the kite isn't completely broken it would just return to earth in a controlled manner.
Furthermore, in a worst case scenario, an untethered kite is actually much _less_ dangerous as it can't travel faster than the wind, whereas while tethered it can go 10 times faster or more. (As a kite surfer the first thing you learn is that if something goes wrong you should release the kite.)
While in use the kite will be some 200 m up, but several kilometres from land, so it will not be able to get very far in the time it takes to fall to earth.
Thanks! The size we're targeting now is a system of two kites on one tether, each with a wing span of 20 m and weighing 1650 kg. This would generate around 1 MW.
The largest off-shore wind turbines right now have 80 m blades and generate 8 MW at a weight of 1300 tonnes (excluding foundation), but the main reason they are so large is economies of scale, off-shore wind turbines are expensive.
Kite wind turbines do not have to be as large to be economical.
But with a traditional wind turbine with 80 meter long blades, the total "footprint" for each one is roughly a 160 meter wide circle. In principle you don't need much space between those circles in a bigger wind farm (but because of turbulence and the wind shadow there usually is).
What is your "wind farm" model? Reserving a kilometer wide circle for each kite pair does not really seem feasible even off-shore.
The wind shadow from a traditional wind turbine is actually very significant. For example, at the London Array, a very modern wind farm, they have 175 wind 3.6 MW rated turbines over 90 square kilometres, so about two per square kilometre.
Some scholars believe that kite energy can best this by a wide margin. For example, you can put the kite systems at alternating heights to combat the problem with wind shadow, and there is no reason why you wouldn't be able to have a significant overlap in the circles, since the wind will be blowing in close to the same direction throughout the wind farm.
They are cheaper to build because they require about two orders of magnitude less material for the upper parts, and at least one order of magnitude less foundation (for the same power generation). This saves on both building and setting it up and I would say there is pretty strong consensus on this part.
You are correct however that it is less obvious that it makes sense when maintenance is factored in, but we are convinced it does. For perspective, large modern wind turbines are much more sophisticated than one might think.
I'm glad you asked ;)! The short answer is that we are going to become the first company to succeed in putting two kites on one line, which is much more efficient but of course more complex to pull off. We will achieve this by iterating much faster, learning from the community and taking advantage of the explosion in drone technology.
Makani is a very cool project but we think we can improve on it further!
https://sherloq.io
We detect and stop online hate speech, harassment and other kinds of cyberbulluying, allowing companies to have toxic free discussions and environments.
I tried your demo and it worked great! I tried saying similar things to see if it was just word matching, for example:
"You should crawl under a rock and die." -> offensive
"You should join a rock band, that music is to die for." -> fine
My only complaint with the demo is that you keep trying to refresh the result as a I type, and I quickly overwhelm whatever rate limiting you have in place and get a "please wait before trying again" message. If you just gave me a "submit" button rather than refreshing as I type, that would fix it.
What I'm curious about is, how do you plan to sell this? Is reddit or HN going to pay for it? Are you going to make a paid vBulletin plugin? Something else I'm not thinking of?
That is good feedback! Thanks!
We are still exploring selling ideas. We are currently integrating Sherloq with the coral project, which will allow companies like The Washington Post use our service. We are also exploring online games and integration with big social networks, where we can help the companies moderate content generated on their Facebook page, for example.
Also we are trying to contact Clever, that will allow us to offer this service for schools.
As I said, lots of possibilities. Currently trying to get as much feedback from as many users as we can.
CoWriteStory is a platform that let users cooperate to develop a story collectively. Users can start a story with a paragraph or more and other users can continue the story line that they like. Users have the opportunity to expand their favorite branch of story or create a new branch with different story line. Writers can get feedback from users via voting and comment system and get reputation and badges for their contribution to the community.
so this is actually a relatively large market for adult themed material based on this idea and there are several different versions out there, each with its own focus.
there are other, non-adult, versions as well but they don't see as much traffic.
*i just realized i definitely went to that site to check that the link was working on my work computer, so thats going to show up as a flagged site in my history now. great :/
reddit is not build for this, you can't create long stories. following stories with branching is very hard, posts showed on first page are hyped based so a story with potential get buried in couple of hours. when people use comment to submit paragraphs of stories and another person use the same functionality to post an actual comment it is getting nasty very quickly.
in general you can't do anything serious on Reddit.
I'm developing a programming language called Morph inspired by LISP, Smalltalk, Eiffel JavaScript, Python (and others). It's extendible, homoiconic and designed to be transpiled to other languages, so that you can build entire projects with multiple applications on separate platforms from one code base. This is then coupled with a symbolic AI engine to generate, build and deploy applications at an increasingly higher level of abstraction.
The code base (and libraries and packages) serve as a knowledge base of definitions (what things are) and procedures (what things exist). The symbolic AI engine can then take information from this (essentially code snippets) and combine them in novel ways. Example: If you know what a RESTful API is (how it is defined), what MongoDB is and which data models you want to deal with, you can generate the source code for an API handling those data models with MongoDB backend from this. Other example: If you know what requests your API supports, what the semantic data types of the objects and fields are and what a UI-based application is, then you can generate a web, mobile or desktop application from this providing a wrapper around this API.
The goal here is not to replace programmers entirely, but to alleviate the pain of having to code things up manually, especially boilerplate. Because you can deal with issues at a higher level of abstraction now, this can make programming more approachable to (so far) non-programmers.
The language itself draws some inspiration from LISP, Smalltalk, Eiffel, JavaScript and Python. I guess, SICP was a huge influence on my thinking about this problem. Compilers (in particular cross-compilers) do something similar, but otherwise code generation has come a bit out of fashion (due to the limitations arising if you don't have a Turing complete language as a base). Some people have also looked at NN for code generation, but nothing usable has come from this so far afaik.
I'm building an inexpensive, autonomous robot that take cares of your laundry. While you're at work, it will wash your clothes, fold them, and put them away.
We are building an orbital communications network and linux based satellite operating system that lets you program your small sat in widely used programming languages, and gives you access to unprecedented, low latency compute and storage on orbit, and enables uninterrupted connectivity to the ground.
what are you doing differently to solve the predictable latency problem? having all the processing happening in the compute cluster so that there isn't enough back and forth to the ground to notice?
Tuiqo allows you to work on many versions of your text at the same time, from within one document. It is the easiest way to revise your text and organize document versions.
I'm interested in everyone having their own personal bot, and a professional bot, and they talk with each other behind your back finding opportunities to collaborate without violating privacy or confidentiality as they are all sandboxed.
We are launching a tuition-free and remote software engineering program based on ccollaborative and project-based learning.
Our mission is to train 1 million software engineers by 2030.
Not everyone can afford to spend thousands of dollars or was born in the right place to get access to a college degree.
On the other hand, most people don't have the superhuman willpower to learn alone from home watching videos online for 8 hours a day during a year.
We use Collaborative Learning (e.g. Pair programming between students) to create a learning experience that is as supportive (e.g. Mutual accountability) as a traditional learning experience (e.g. College or bootcamp) and as scalable and affordable as online learning platforms (e.g. Coursera, Udacity).
Well, to the extent that a programme such as this will increase supply, you could argue that it's currently kept artificially low because of the costs of education in some countries.
I don't think that is quite the situation we are in though. Computer programming is an unusually easy industry to enter. I didn't study computer science, only started working as a developer much later and my experience is that potential employers and definitely clients if you're freelancing care very little about _how_ you learned it, they just want to know _what_ you can do. A portfolio is much more important than a diploma.
This is why i really take issue with this statement: "On the other hand, most people don't have the superhuman willpower to learn alone from home watching videos online for 8 hours a day during a year."
If learning programming entails you watching videos 8 hours a day for a year, I submit you are have chosen the wrong profession. You will be competing for jobs with people who love coding, people that it would require a "superhuman" effort to _keep_ from programming.
Anyone with a real interest and access to a computer with internet, can _easily_ learn to program. Not so much by watching videos full-time, but by a combination of actually programming, reading tutorials and yes, watching some videos.
Then there are people who are not passionate about programming, people who are not naturals but just want a decent job. I suspect this programme is aimed at them. There is absolutely nothing wrong with that of course, but it's unlikely that they will be competing for the $200k jobs in Valley startups.
That's short-sighted. Eventually, things that are highly skilled become less so, and eventually integrated into the baseline running of society. At some point in history, the fact that you can read and write would be a big fucking deal, but it isn't today. At some point in the future, making the machines do things will be in everyone's skillset to some degree. This just seems like a step in that direction.
To use an example, this sort of thing has about as much impact to data engineering as someone coming to the table with Excel skills does - they both generate some lines of code to work with data, but we're not working on the same level of issues. Doesn't mean I don't want to raise the floor, though, that has value to the whole organization.
EDIT: To be clear, I'd never heard of Microverse until 10 minutes ago. I have nothing to do with them, but I think it's a cool idea.
A free course schedule planner for college students, platform to connect with people within classes and soon the way graduates will find the job they are uniquely qualified for.
Write me at kaiser@plan.university if you think your college/university would need plan.uni as well :)
As a class project we wrote a Chrome extension that transformed our university's terrible course offering page into something like this (but less ambitious) It was a pretty big hit and continues to be used even 3-4 years later.
The goal is to create a social media platform that does for content creation what previous social networks have done for distribution. Create in an open space, collaborate dynamically, and remix any other content on the platform.
https://typito.com/create - Help creators produce videos for web fast and easy. Provide them with simplified online work-flows to publish videos that give better ROI on online platforms like YouTube and Facebook.
You can check out the demo - https://typito.com/demo. The tool currently helps with one use-case - adding good-looking motion titles, images and text animations on your video before publishing on YouTube with a few clicks. Adding these elements, we have seen, improve the watch time and engagement on YouTube videos and hence give better ROI.
We went live 5 weeks back and have 15 active users (who publish more than 4 videos / month via Typito), mostly YouTube creators - the market we are looking at.
A smart calorie counter app with AI - Analyzes the foods you eat, how active you are and the progress you make to automatically set nutrition targets designed for you, every single week.
how do you guys do things differently compared to MyFitnessPal? I recently started using MFP and it's been useful in that I know how much I'm eating (whereas previously I was just wrongly guesstimating with some mental math).
I love your concept, just curious what sets it apart.
FYI, you will probably hear from 1000+ more people asking the same question: Why not use MyFitnessPal? My advice? Don't compete against them. Everyone, from moms to entrepreneurs to wall street folks use them, know them, and recommends them. They're too big, and you need to be 10X better to go up against them, and even then it might not be enough.
I feel like that is there biggest weakness, they are too big. There is lots of room for advancement in this area and they are far too comfortable. There calorie counter is also critically flawed, when a user plateaus and stops losing weight they are left in the dark, also they may not lose any weight at all based on there estimate and carry on eating the incorrect number of calories for far longer than they should, this happened to me personally before I built Poundaweek. Also the amount of calories the other apps recommend is static, and only adds calories based off of very flawed estimates of calories burned from exercises. Poundaweek learns and adjusts dynamically and specifically for each individual user as well as letting you know why and what you can do to keep on track towards your goal.
I feel like you're up against an app that's synonymous with tracking food/calories. However, if your app does indeed do what you say well, you have something. However, in my opinion, initially you would be better off going after users who don't use MFP or any app than convincing them to convert. Though I'm only a two-week user of MFP, there's too much switching cost I feel. And I'd really have to hear good things about it from others/in the news to convince me to switch at this point.
I'm going after anybody that wants to lose weight safely and consistently, the people that don't know how much of what to eat every week and the people that get stuck after using other calorie counters. Simply put Poundaweek works and continues to work - It was built out my own frustrations with the other calorie counters that stop working or never worked at all.
Haul will be a city marketplace of local retailers' stock with on-demand delivery, offering an alternative to shipping.
The goal is to offer products more reliably and quickly than Amazon or individual companies' online stores while charging less than the user would've paid for shipping. Each city will have a separate marketplace with immediately deliverable inventory. The model largely prevents against counterfeits, enables delivery within an hour, and more.
We also have plans with some retailers to offer an alternative to lugging bags around after shopping in-store, via using "HaulHome" at checkout to have your purchase delivered home within a timeframe you specify.
Would love to receive input from the community here!
A way for Instagram users to sell their photos to brands who want authentic stock photos. We're also building an influencer search/compare/booking engine.
The company is based around Traccar open source GPS tracking system. Providing SaaS and professional services to our customers. We develop everything from back-end to mobile apps.
Although we don't use any machine learning at the moment, GPS tracking is a great area for it. If we get funding, that can be one of the directions for future development.
SnapCycle is a modular personal transport vehicle. Can be reconfigured into an ebike, self balancing board, skateboard, and multiple variations of each. All existing electric personal transporters utilize the same tech and components but have only one possible configuration. Our modular system allows one to ride a bike to work and a skateboard home.
Priceline for Uber and Lyft. Location and price comparison for your ride within walking distance.
Since transportation is usually a top 5 expense for most people, and more people give up their car for ridesharing services, saving 10% on your rides could save you hundreds of dollars every year.
We are going to send out a "one time use" Sim Card sticker to registered voters that you can stick on your regular sim card and Vote in the next election from your phone. Great for absentee ballots, disabled community, lazy people :-)
That is a great question. It is plastic, so I would assume it is but I will confirm as well as find out if they are made from recycled material. Thanks for the comment!
It's probably an algorithm to get you some decent % of views early on to see what % of upvotes you get so it can move your post up or down accordingly.
A peer-to-peer Web browser. We let users create and host unlimited sites from their own devices. We think the technology can support thick hostless applications, which eliminates data silos and makes every application forkable from the browser. No blockchains, just an improved variant of BitTorrent called Dat [1].
We've been quiet the last 2 months but our 0.7 release is coming up, and we're excited about it.
1 https://datproject.org/