Ask HN: What are you working on and why is it awesome? 2017 Edition
A similar question was asked just under a year ago. So I thought it is about time to ask again for 2017.
2016 edition is https://news.ycombinator.com/item?id=11259746
2016 edition is https://news.ycombinator.com/item?id=11259746
49 comments
[ 1.9 ms ] story [ 105 ms ] threadIt's a growth tools suite for bloggers (as explained on the landing page.) Right now we're in pre-launch but we're launching in late Feb / early March (a couple weeks). It's awesome because:
1. It gives bloggers social/growth tools never before accessible.
2. It provides a new/different source of revenue.
3. It can replace the redundant/impractical "timeline" widgets offered by FB, Twitter, etc. Those currently act more as a glorified social channel promoter than anything actually beneficial to the blog itself.
4. It's free (for the basic version).
5. It's powered by Snapzu, our Reddit/HN like voting community platform.
Would love any feedback, and am open to answering any questions. Cheers.
I did a Show HN about three months ago (https://news.ycombinator.com/item?id=12879179) and was so grateful for all the great constructive feedback I received, it really encouraged me to keep evolving the project. And I really have come very far since then.
So I thought I would share what I have achieved so far:
1. The original release was iOS only, but so many people told me I should build an Android version, so I did. Now both iOS and Android are completely open source on Github and they both use the same JSON grammar. Here's a recent post where I talk about how I designed the JSON-to-Native mappings on both iOS and Android: https://medium.freecodecamp.com/how-to-build-cross-platform-...
2. I've implemented functional programming in JSON. Yes, a function in JSON. I had to come up with a JSON grammar that can represent all aspects of a function like invocation, arguments, callbacks, call stack, recursive calls, subroutines, etc. If you're interested, check out the post I wrote about this: http://blog.jasonette.com/2017/02/15/functional-programming-...
3. I've implemented a "require" for JSON, so it can be used to parallel process multiple JSON objects from different sources, which I think is great for people who want to build a mobile app for their decentralized apps. I personally think decentralized apps will take off in 2017 and going forward for many different reasons, so this means a lot to me: I wrote about this here: http://blog.jasonette.com/2017/02/17/require/
These are just some I have worked on personally, but the REAL cool part is that the community has really picked up and many people have been contributing to the project. And I think THAT's the main reason why it's awesome. The fact that some nobody with no track record can just release something that resonates with enough people and anyone from around the world can just jump in and help. This is my first ever experience with an open source project and everything feels so magical to me. I hope to update you guys with even better news in the future!
The update will feature new legal analysis, including references to court decisions that have come out in the meantime, fewer elaborate technical suggestions, and more concern for the problem of how border agents may react to noticing that you've took precautions against searches (not only for whether the precautions are effective).
There's been a lot of awareness of the risk of searches of electronics at the border lately, so hopefully we can get some up-to-date and informed guidance out there.
There are a ton of math learning solutions but most of them are pop-up worksheets slapped on top of a game. In SMW the math is integrated so every interaction has a mathematical property.
The "making games" part is possible because we expose the editor in WebGL game, because we really want to put kids in control of their experience--nothing helps move a kid's mindset from "math sucks" to "math is fun" like being able to create your own world of math toys.
We're in beta, launching June 1 and always looking for collaborators! Reach out if you're a math or physics geek, we can make some pretty neat visualizations/interactions using our engine.
http://hackertribe.io/ - a community of hackers and founders.
http://lumiverse.io/ - discover great educational videos.
http://fictionhub.io/ - fiction publishing platform.
https://github.com/andrew-lucker/rust-openai
I started working on this recently: https://github.com/joshuamorton/tut, and might go back to it. After seeing Kenneth Reitz's Pipenv I wanted to do something similar but a bit more extensive that included testing tools and some other stuff. And tut is the WIP result. Think Cargo for python.
I made it my mission over the last year or so to build something better than rosunit and rostest, because having to more or less copy launch files for every test seems arduous. I ended up with the test_utils library that we'll probably spin off into its own package soon, but its a much, much nicer interface for writing tests for ros, and at this point seems to be relatively stable, with one or two minor changes I need to make.
Glad its helpful!
We just launched Optimail (https://optimail.io) - it uses reinforcement learning to automatically optimize drip email campaigns.
We only just launched yesterday, so I think we are still working to find that exact product-market fit. At the moment, however, we're targeting medium- to large-sized businesses that use drip email marketing campaigns, such as onboarding campaigns, lead nurturing campaigns, and retention campaigns.
Thanks for the feedback on the copy! If you are feeling extra generous and have a second, shoot me a PM and let me know what parts you found confusing. We want it to be accessible to non-technical marketers who might be using older software like Mailchimp, etc. (After all, a main benefit of using AI here is that you don't need to get into confusing automation-building stuff like multi-branching decision trees.)
> Our suite of advanced analytics, including Cohort Analyses, Sequence Analyses, and Algorithm Metrics will unlock new insights in to your customers and campaigns.
> We use a combination of deep reinforcement learning and hierarchical Bayesian models to quickly and accurately learn about your customers and optimize your campaigns in real-time.
I think you could simplify those a bit, maybe hitting from a higher-level with a link to a more detailed explanation of how Optimail does its thing. I totally get the value in using these terms, but for the landing page I would imagine using simpler language may result in non-technical users being able to grok the benefits of your product a bit better.
But you're absolutely right - our goal is to appeal to developers as well as marketing types, so perhaps this level of detail is just confusing and unnecessary on the landing page.
Feel free to shoot me a message if you have any other feedback - I really appreciate it!!
Been in beta for the past few months and am excited to finally launch, having spent nearly a year developing it any chance that I get.
I’m currently working on wrapping up some great business-ey features such as an admin dashboard that will allow users to visually monitor metrics for each product to get insights for popular license types, as well as other crucial metrics like license churn, overdue license check-ins, usage across different machines and more.
Been a long road, but I'm excited to try and put my sales hat on soon. It's a service I've always needed for developing small niche desktop apps.
[0]: https://keygen.sh
Site caching is so underrated I had to do something about it. In fact, I recently announced its first public beta. The one USP is this: you create a Cachoid and it automatically starts caching your site. No need to get into the nitty-gritty details of caching or get your hands dirty. It just works!
The difference between Cachoid and other platforms is that we guarantee memory allocations per Cachoid (Varnish instance container on steroids), which is crucial to achieve speed. A multi-tenant content distribution network doesn't always have a memory-store SLA for its tenants.
More info: https://www.cachoid.com/
Quick overview of its architecture: https://www.cachoid.com/blog/2017/01/15/high-level-architect...
Docs: https://www.cachoid.com/support
You're right, I didn't see it from that angle. Plus the registeration link would auto login already registered users anyway.
Thanks!
I'm also hoping to use it as a platform to test some of my theories on how to apply traditional IDE search techniques to help find and understand code snippets.
Plus its just a fun way to see if I can write code that "scales".
I create hobby-projects for fun, I have two (I think?) that I charge money for. They'll never be "real" companies, but they cover their costs which is all I wanted.
I see broad trends on sites like this, for example you can pretty much pinpoint the moment that every other Show HN started adding "Deep Learning" to their submissions. That kind of herd-mentality is pretty depressing, but also fascinating to watch in action.
Startups as a meta-topic are fascinating, although again I have that same feeling of depression and sadness when I read "Growth hacking" tips that basically boil down to fancier ways to spam and mislead, but I appreciate reading about real people and real-projects. The posts that start out "I posted on my blog, which has 50 million readers, and suddenly I was profitable" are pretty unrealistic.
Simple projects that have slowly become more and more popular, like the guy who is frequently talking about his webcounters, are a breathe of fresh air from the hype.
I'm sure I could code a web-counter. I'm sure that it'd never become super-popular, because long-tail counts for a lot. Then again I do have a few readers of my blog ..
Hope this helps.
The email consists of a brief, 5 question interview, asking about a problem someone faces at work and the software they wished they had (that they would buy) that would solve the problem.
Each day I receive a handful of responses from developers who want to learn more, and I try to connect them to the interviewee
We are using various ML modelling methods to understand each sensor's normal operating behaviour - more info on the tech is at https://www.shprd.com/technology
We are just coming out of our R&D phase and into full commercialisation. Any critique of the explanation site or other info welcomed!
Wow, mfw I'm still working on the same projects.
What was going to be a Berzerk clone[0] has turned into an attempt at a more general purpose framework for games in C++ and Lua[1]. Eventually, I'll get around to actually finishing something in it. Mostly it's just a time sink for self-education.
And I'm still slowly working on a HN-like forum in Hack[2].
[0]https://bitbucket.org/kennethrapp/berzerk
[1]https://bitbucket.org/kennethrapp/sdl_framework
[2]https://bitbucket.org/kennethrapp/basedforum
Also, off-topic, but why bitbucket?
I tried getting it to run on OpenShift but their HHVM cartridge wouldn't allow me to run Composer through it instead of PHP, which I had to be able to do, and their version of PHP was out of date.
Given the state the project is in, I haven't bothered trying anywhere else - it wouldn't be worth spending money to host.
>Also, off-topic, but why bitbucket?
Free private accounts, not as much social media nonsense as there is with Github.
Thinking of selling something but hate working just by gut feeling? AiHello will give hard starts (sales, competition, price, profit etc) and whatever you plan to sell.
We help online sellers launch their product quickly on multiple channels like Amazon,Ebay, Jet.com etc as we have inventory management built in.
It's going rather well - 87% positive reviews on Steam and people saying things like "In years to come, I think this will be seen as one of the seminal VR titles" (paraphrase from a recent review on Reddit.)
A programming language that I've been designing for about two years now. I'm in the initial bootstrap phase, and about to rewrite the compiler in itself. The goal is to use it as an experimental test bed, as an example, the compiler should use a tracing inference model I wrote a thesis on, for automatic type inference, and function overloading. The overly-ambitious goal is "As easy as Python, as safe as Rust".
Offers are hand-picked/filtered and companies are verified (lightly, as much as one can tell from a fast look at public info; mostly regarding remote environment). It should also be fast, with well-formatted data - so far, so good.
You can find it on https://matchr.in
My main concern is that I won't be able to get the user's picture and basic info.
It will destroy millions of jobs.