Ask HN: What project are you most proud of? | Hacker News Reader

Ask HN: What project are you most proud of?

30 pointstech_crawl_27 comments

Been working on some awesome stuff and wanted to see what everyone else has been a part of...

Lifetime: Khan Academy ported some code I wrote; it now runs all of their internal user engagement metrics.

Recently: Erin and Thomas are still busy putting the finishing touches on their half of Starfighter, and I'm basically just on client communications duty, so I gave myself a two-day engineering project as time off for good behavior.

Certain events which happen when players play Starfighter games are significant. We want to congratulate players when they happen, ideally in as close to real-time as possible. Also, since "strike when the iron is hot", if that player happens to be a job seeker that would be a really good time to do an introductory phone call if they're willing to do one.

We have heuristics which identify these important events and pipe them into Slack. The plan was originally "If we see one of the notifications, send them a person-to-person email."

But if you're playing Starfighter in your browser, you aren't in your email client, so you might not see the email until e.g. the next day. What we really want is like Facebook messenger, except inside our own application, hooked up to Slack, with the ability to be selective about who it is open to.

Enter Apollo, messenger of the gods:

Front-end: React and SockJS Back-end: SockJS-go, some Go glue code, and NSQ Back-end for Slack: three very trivial endpoints in our main Ruby on Rails application Interface on Slack: two slash commands and one asynchronous incoming webhook

How it looks in practice:

In-app message window: https://www.evernote.com/l/Aaf2wn2dyxhPDoHAzSYymDcNIasqcmj91...

Slack: https://www.evernote.com/l/Aadc-vDMIpxNc7z6XZ1_xH1jMSNOVJ-58...

Not bad for two days of work. (Starfighter will likely OSS the only hard part of this, which is the NSQ-to-websocket piece.)

Looks really good! Another benefit of this approach is that it will probably be quite obvious to your users that the messages are not canned.

I just wrote my first golang library: https://github.com/jcuga/golongpoll

I'm proud of it because its a general purpose library (http longpolling) versus random prototypes that I normally make. Also it has a lot of unit tests, code comments, and a decent README. I can appreciate the amount of work it takes to create a polished library for others to use after making one myself.

My work on Disney's movie "Enchanted". Went to so many damn meetings and hardly got to do what I love, but I learned so much about the process and loads of respect for the leads and higher ups. Also lots of people have seen it, as opposed to the huge pile of schlock I've worked on that nobody has seen.

That is really awesome! I am a little jealous that you got to work on that.

Thanks yo! Disney is actually a-okay to work for. I've done a few of their shows. So, probably best not to say what I'm working on now then ;)

Its really not worth mentioning with all these awesome projects in the comments, but i created a Pure CSS slider, which i think was away for me to show people my level in CSS. Plus it got a lot of stars on github haha. https://github.com/Elitewares/CSS3-Slider

Of course it's worth mentioning :) Even small projects are worth posting, like this recent one by Holman: https://github.com/tholman/github-corners

Not for anything else, but I'm proud that I have been able to stick through my current project (http://www.hellobox.co) for more than 2 years. I used to be very distracted with all sorts of ideas and abandoned projects.

Cool! Any plans to open-source Hellobox core? Do you have a pricing model already? If I start a community (linked to a custom web domain) and it grows well over time, who owns the content/data in that community? Can Hellobox snatch the community or hold it ransom for a price?

I guess unless these are made clear, potential community builders might have lot of doubts on Hellobox as a platform.

Good work though, the stuff seems pretty neat.

I know it's small and probably not worth mentioning, but I created a bitcoin exchange's api aggregator (my first Node.js app). Basically it attempts to wrap multiple exchange's api's into 1 common library, so you don't need to constantly look up the docs across different companies.

https://github.com/jxm262/xchange.js

I built another project to analyze prices across the exchanges using the xchange.js library. Surprisingly both have 13 stars on github :) I'm planning to add _alot_ more functionality over the next month or so to both projects.

https://github.com/jxm262/btc-stats

Those are both actually really awesome.

My own project: http://www.appointmind.com/ Panasonic and Honda among others use it.

For a client I've built and maintain part of the backends of the largest speed camera app and data provider in Europe.

These are my babies currently:

http://www.jque.re/

https://www.unicod.es/

I could keep making these, but it's do or die with side projects. You either commit to them with a full heart, or they begin to bit rot.

Currently I just maintain the projects and promote them the only way I know how (Hackernews and Twitter). There are other outlets for promoting your projects, but they tend to be very niched and specific.

You can see the Twitter accounts here:

https://twitter.com/unicodetools

https://twitter.com/jQueryRepo

I only get about an hour a day to work on it during my commute, but I am having fun with my side project bestfoodnearme.com I am hoping to have a new look for it in about a week, more visual, more pictures.

I like that concept.

Kudos for finding the time for your side project, even if during a commute.

A book series and website where I explain LSAT practice tests. The LSAT is a standardized test used for law school admissions. It's an analytical test with very precise English, and my explanations help people make sense of it.

A significant portion of the material was done in the early years, when I worked like mad. Now I'm running the business at a more comfortable pace, though ramping things up to work on a new product for the site.

http://lsathacks.com/explanations/

Wow. This is a huge project.

Thanks! It took years. Revenue generating, fortunately. Otherwise there's no way I would have built it. But I like that I'm able to keep it free.

Wow very impressive, looks lie you have 1000's of pages very well organized. You'll definitely be able to sell this at some point I think.

Any plans to monetize?

Thanks! The site already makes money, actually. Plus the books sell in hard copy. I'm working on other things to sell on the site, because I think that has more revenue potential than the pdf versions I have on sale currently.

Once upon a time I made some timelapse software for building videos of hackathons (ludumdare, specifically). It got pretty good traction and the quality/quantity of the timelapses in that community went up, which was really great to see. Now it is fairly obsolete (and so ugly!), but over its lifetime it saw about 50,000 downloads and helped build timelapse videos for everything from game dev hackathons to professional car engine builds to medical research.

I'm one of the authors of QuickSynergy: https://github.com/cesarbs/quicksynergy-gtk

I was quite happy to learn recently that it's pretty popular with my wife's team at Amazon. She had a pleasant surprise when she installed it on her machine and noticed my name in the About dialog :)

It's due for an update though...

My Android location sharing app[1], while still barely known, has generated a lot of great feedback and personal "thank you" emails.

It's barely made any money (40€ in a year or so), but the great feedback makes it all good.

[1] https://graticule.link/

I just finished writing Python Crash Course, which is coming out this month from No Starch Press. I'm looking forward to hearing what people think once it's out.

http://nostarchpress.com/pythoncrashcourse

None. I have impostor sindrome.

Reply on news.ycombinator.com