Ask HN: What projects are you working on?

40 points by RomanPushkin ↗ HN
I think its good having this kind of thread at some period of time.

Previous posts:

https://news.ycombinator.com/item?id=8561842 https://news.ycombinator.com/item?id=7957989

59 comments

[ 2.1 ms ] story [ 72.7 ms ] thread
I'm a business guy not a technical guy, so these projects will reflect that. By day, I'm a consultant helping mid-sized companies build growth management systems, so these projects reflect that a little as well.

There are 3 markets here - Coaching, Tourism, Barristers - so I'd love to hear from anybody also playing in those spaces.

1) I'm building the business system for a business coaching organisation. They're nominally competitors of mine, but I think they have a great niche and expertise and more clients would benefit if they had a systemised approach. This might scale for other coaching / consulting companies.

2) I'm researching micro-operators in the Travel and Tourism space, to see if an approach to strategy and traction I built for a similar organisation can be shared at scale and at their price point. Finding it challenging, because many of these micro businesses don't want to grow / invest in growth.

3) Early market feedback from Barristers indicates their 'business' training is inadequate. I've developed an overview for building a Bar practice from pre-startup to a thriving specialisation, and am currently talking with potential partners in their area of specialty (eg, personal financial planning).

So my focus is seeking scale. These aren't going to disrupt the world, but they're also 'not another to-do list app' so I feel I'm helping.

1) http://trevorb.cloudapp.net:3000/ personal dashboard to read hackernews/reddit/web comics without seeing the same post twice (I swear I can waste time 10x more efficiently)

2) http://trevorb.cloudapp.net:3001/ mmo type game exploring webgl, socketio and just experimenting

+1 for Personal dashboard.
The personal dashboard is awesome.

Suggestion: put a count on the comments icon to know if there are any comments on the article (I find the comments more valuable)

Something of a cross between reddit and craigslist. Not live yet, still sorting out voting. Hopefully eventually though.
(comment deleted)
I recently launched Steam Profile Lookup!

https://steamprofilelookup.com/steamid/76561197982486517

Fraud is a big problem in the steam trading scene and this tool basically allows people to do a quick background check before they trade.

This is my first real project that I've been able to put in front of users and it's been a blast. I've found it's a lot more enjoyable for me to have actual users trying stuff out rather than just building internal tools. Getting feedback, trying stuff, and running tests is just deeply satisfying. It's also really forced me to write good, production grade code.

Credit scoring and fraud detection/prevention are big problems in games and virtual economies in general. Keep me posted about your progress (DM @fsimoneschi).
This is very true. Currently the best communities can really do is name and shame. Obviously this means there will always be a larger percentage of scammers here than normal. The other big issue with the community driven method is that people can't really take their lumps with a temp ban. Make one bad decision and it will haunt you

As people spend more and more time inside virtual worlds the natural progression is for people to trade in-game items for cash. If someone could make rent just playing WoW and selling items, there in-game time would skyrocket! Unfortunately game companies are pretty against this with the only real exception being Valve.

The biggest argument against cash trading is that it ruins the markets for 'normal' players who don't want to grind gold. Markets in general are hard but throwing in the requirement that they also have to be fun makes the challenge extremely difficult. Competing against bots just ruins the market aspect of the game for most players.

I'm not sure what the solution is but I think a core issue is that most companies don't think of structuring the game around the economy and item entropy. This leads to weird inefficiencies and massive inflation when players and bots alike start grinding gold heavily. The only game that really comes close to fixing this issue is EVE Online. Their market is fairly well built and expensive items can disappear in an instant making runaway inflation not as much of an issue (though still very present).

It's an annoying problem from the traders prospective because fraud runs rampant and there isn't a ton we can do. Sites like mine help raise the bar required to scam large amounts of money but without the game companies support it will always be an ongoing battle.

This kind of app can be super useful to do background check in other situations (startups, so called head hunters).
I'm working on wits (written in the stars) - an app that takes western astrology into today's mobile platforms. It's kinda sad that there aren't any proper astrology apps in stores today. http://astrowits.com
I'm working on a network flow monitoring system called Cistern [0]. Part of it is a time series storage engine called Catena [1]. I've also written a few protocol implementations for it, including sFlow v5 [2] and SNMP [3].

[0] https://github.com/PreetamJinka/cistern

[1] https://github.com/PreetamJinka/catena

[2] https://github.com/PreetamJinka/sflow

[3] https://github.com/PreetamJinka/snmp

interesting! I am a network engineer so I am curious to ask - what benefits would your tool bring me in my environment? 100+ branches, 3 large data centers, etc...
I'm trying to make it very scalable so that you wouldn't need a big machine to monitor many systems. Using Go helps with that, but I've had to make protocol implementation efficient too.

I want this system to be very developer friendly. I think I could make this more like a database instead of a monitoring system. You wouldn't insert data into it directly; that happens with network flows. You should be able to query it like a database with rich data models and aggregations.

I don't think it would be a direct benefit. I'm hoping that it will become a great framework so people can write tools on top of it that would be way better than anything I can make.

- Building an entity-component framework in C++ and making Space Invaders. No code because it's literally terrible but here's some "gameplay footage" https://www.youtube.com/watch?v=uxQZz0yr4V4

- Building a forum that's a sort-of-but-not-entirely HN clone in PHP. It's a free app on Pagodabox and still mostly doesn't work but assuming it's awake what there is of it is here: http://original-oak.gopagoda.io/ This might wind up being my final project this semester, and if it is I will post it here, or maybe not I don't know.

- Unity tutorials and ragequitting/restarting Handmade Hero all the time.

A proof-of-concept for a multi-platform, secure, distributed, client-side chat application called Toc Messenger: http://lewisl9029.github.io/toc

Built with web technologies such as Ionic, Telehash and remoteStorage.

It's very bare-bones and severely under-optimized right now (have fun downloading 4MB of code lol...), but I have some pretty ambitious features planned for launch such as simultaneous logins, group chat, voice/video chat, file transfers, some kind of email-based user on-boarding process (I'm looking at integrating Persona), and peer discovery using the Bittorrent DHT over the internet as well as automatic discovery for clients on the local network.

And yes, of course it will be open-source. =)

A service which is a cross between git and dropbox with client side encryption and deduplication.

The clients are GPL open source with a cli client underway here https://github.com/buppyio/buppy

My server is private but based on a kubernetes cluster. I hope to add a server version people can setup themselves if they don't want to pay for the official server.

Still too wip for alpha, but I want to announce and get alpha testers soon.

I'm going to be making my first Ruby Gem. It's going to enable easy lookup of weather forecasts. I want to unite geocoding APIs and weather APIs, basically.

It started from a simple "whats the weather" app I wanted to make to learn Sinatra, and then I realized I could turn everything I was doing into a gem and that would be kind of cool, and definitely new for me. It's simple, but I figure if I do a good job it will be useful to someone, as there don't seem to be a ton of (well-documented) options for this kind of gem out there.

That's awesome! Good luck, post it back here when you have it.
A personal banking assistant you communicate with using natural language. (Using speech recognition, NLP, and ML).

https://www.wealth.ai/

Will launch in a few months.

Extremely cool. It'll be nice to see it catch on.
I am not sure about the domain “personnal finance”, seems to narrow. It looks like a small H.E.R. application.

The design is convincing.

I went through your blog posts, nothing about your stack. Could you explain what do you use of “speech recognition” (and “speech synthetisis” if any)?

I'm making a countdown timer for my university's transit system.

https://github.com/jbwashington/BusHog

I find myself making a series of countdown timers in the morning based off my class/bus schedule, I figured I would cut out all the bloated map UI and pull data straight from the school's bus API.

Been working on https://fris.bi for a while - it's a WebRTC group video chat app.

Would love anyones feedback!

Need demo on the main page without registration!
Together with another dev I am buildung the file sharing and editing service we always wanted to use.

Editing-support for lots of file-types, encryption, git-like workflow, are some of the things we Plan to implement.

Together with another dev I am building the file sharing and editing service we always wanted to use.

Editing-support for lots of file-types, encryption, git-like workflow, are some of the things we plan to implement.

Been working on http://www.soundshelter.net

Sound Shelter is a platform for electronic music lovers to recommend their favourite vinyl records to each other.

We then use the recommendations to build a personalised feed full of music we think they'll love.

An ebook that'll function as a guide as to how to quit law and code based on my experience.

http://www.quitlawandcode.com

This is interesting to me. I'm a nurse who recently made the transition to developer and I think others could benefit from my experience.

Did you validate a need for this book prior to starting? Do you see an exodus of lawyers leaving the profession? I ask because I don't necessarily see nurses leaving their professions although most everyone I know is unhappy with the job.

You aren't Jared Sinclaire are you? http://jaredsinclair.com/about.html He's a registered ICU nurse who became a developer - I thought this transition is more unique than mine.

More than a quarter of licensed attorneys who graduated from law school in 2000 are no longer actively practicing for a myriad of reasons. A significant number from my year, 2008 hold jobs that don't require a law degree.

These articles should give you a glimpse into the situation. There were several others I really like from over the years, but I can't seem to find them at the moment.

http://www.nytimes.com/2015/04/27/business/dealbook/burdened...

http://www.slate.com/blogs/moneybox/2015/04/30/job_market_fo...

I don't believe the general job market for nurses is as bad as it is for attorneys. People also have a great difficulty accepting a potential sunk cost, or refuse to believe that there's greater opportunity elsewhere once they have both feet committed somewhere.

Nope, thats not me, although I'll probably reach out to him soon.

The market for nursing isn't bad, although it's not as good as it was 5 years ago. Job satisfaction, however, is horrible.

I'll start a similar book and see if I can drum up some interest. Two things going for me: Nursing satisfaction is pretty low and burnout rates are pretty high.

http://www.everydayhealth.com/news/why-americas-nurses-are-b...

http://www.indeed.com/forum/job/nurse-rn/Want-leave-nursing-...

On a list of unhappiest jobs in 2013, nurses came in at No. 4. First place were associate attorneys, btw. I thought that was pretty funny.

A lot of things at once. I'm finally committing to finishing them in order.

A) My personal CMS. Standard CMS but outputs a static website. Kind of a static site generator without going through all the "put file in folder run [jekyll/bootstrap] whatever to generate site and upload via ftp" hassle. Just blog like you normally would if using wordpress or medium.

B) A couple of SMS applications to generate some money on the side for me. Junk apps along the lines of "send 'quiz me' to 7890 to get a new quiz everyday". They are big (huuuge) in some parts of the world still.

C) An android app to create a single image from two images. This is just to learn android development.

D) A firefox addon for...

Sigh. I should probably shut up and go back to working on this stuff.

Heaps.

1) A framework (edencoder/eden) 2) A (sort of) imageboard https://www.1eden.com 3) An online marketplace (adlist) 4) A rental management software

... and a full time job

Paginate (imageboard)
I'm working on a sort of DAW-ish thing for iPad. Except it's not really a DAW — more of a musical sketchpad. When I was sitting down and thinking of what I would like in a music creation app, I thought of two things. First, the typical DAW grid structure (and sheet music before it) is far too rigid for modern music. It's really hard to annotate music with syncopation, pitch bends, and fluid rhythms unless you already have the song in your head — and even then, it takes forever. Second, the feedback loop in most DAWs is really large. By the time you've pecked away at your piano roll and fiddled with the dozens of switches, your song is no longer fresh in your head.

In other words, I wanted something akin to musical clay: a very simple, very malleable representation of music in space that focuses entirely on ease of interaction.

With my app, I try to solve both of these problems. In regards to the former, you have the option of drawing every note arbitrarily in pitch or time, as simply as in an ordinary drawing app. (Snapping is an option, too.) And as for the latter, the whole thing is a scroll view, so panning and zooming are incredibly simple. Furthermore, you have immediate access to undo/redo and can also rewind to the last place you played from, meaning that iterating on your ideas takes only a few seconds as opposed to the arduous process of controlling a DAW. The UI tries to get out of your way as much as humanly possible.

I should emphasize that this won't be production software; the music you create will sound like MIDI. Rather, my goal here is to make an app that you can freely sketch and create musical ideas in, which can then be used as a rough draft for working with a more featureful DAW.

Performance is a top priority, and I'm targeting 60fps on my iPad 3.

Screenshot with rough "programmer" UI: http://cl.ly/image/1I0y0G2v0b2T/IMG_0281.PNG

Depending on my work ethic, I expect to release in a month or three.

What do you mean by work ethic?
Well, hah, "depending on how lazy I am"!
Building a platform to assist in marketing some vacant office space we are buying in Atlanta. Will then try to take on CoStar.