I just redesigned what an offer letter looks like. [1] I think it's cool for a few different reasons: (1) most people don't truly understand the value behind an offer from a startup. Equity for example is tricky for newcomers. (2) Most companies don't share this information unless you push them on it. I wish more companies would be transparent about it, and I think doing this will help.
[1] As with many things in design, I think I should attribute the source of inspiration, which is Carta. They wrote about a new offer letter format back in 2016. I'm basically taking their Apple Keynote slides and moving them to the web.
http://tracket.com - It's google news + the internet archive. You can go back in time and see what the top breaking headlines were on a given day. I've been running it since July 2016 so there's nothing before that, unfortunately.
http://ripplescan.com - A website to visually explore the Ripple XRP network, lookup accounts, and transactions, and subscribe to XRP news in your inbox every day.
https://exponentialbackoff.substack.com - A newsletter where I write about the soft skills side of software engineering that I've learned throughout my career. Hopefully it's helpful to some younger engineers that are just starting out in their careers.
I'm hooking up a piece of kitchen equipment to a cloud backend. It's for a major restaurant chain.
I've done this probably a dozen times over the last couple of years as a demonstration of the technology.
It always ends the same: Every party fights for ownership of the data, but nobody can figure out what to do with the data once they have it. Nobody wants to pay for it, much less support a subscription model. The free sites are held together with duct tape and chewing gum. Nobody wants to support it in the field. There are too many variables in network infrastructure. You can't use the network in the store because the cashier system is on there and PCI says no way. Getting a solid 802.11 signal in a room full of sheet metal is a nightmare. Customers want to use off-the-shelf cellular hotspots that are pieces of garbage and couldn't hold a TCP packet if it was covered in crazy glue.
But the demonstrations always go well. The customers think it's cool.
So on we go, boldly onward into the IoT future. I just pushed the next release up.
I spent 3 months scraping 6 million pages from an extremely old political web forum (I think their backend was written in perl and hasn't been updated since 2001 and the response times for these pages were extremely high) in an effort to learn NLP stuff with a working database. Its cool because you get to see how old Boomers digested huge events during 9/11 up until now and the slow decline of web forums as a medium. The pain is trying to parse the actual pages since the dev never learned how to liberally use div tags like we do now.
I once scraped a large forum off Yuku, the HTML was easily the most bizarre I've ever seen (i.e. multiple different HTML formats for posts on the same page for no apparent reason).
* Every cell is live, reactive code that re-evaluates automatically whenever any of its inputs do. This includes awaiting promises and yielding new values from (potentially async) generators.
* A notebook can import and reuse any cell from any other published notebook.
* You have access to all of the computational goodies that the web platform gives you: graphics capabilities with <canvas> and <svg>, the GPU, the world's largest open source library with npm, and web APIs.
It's been interesting thinking about what it would mean to have a read/write medium for the web — where you're working with live values, and not in a "dead code" text editor. For just a teeny hint of some of the possibilities, check out Tom's autocompletion gif: https://twitter.com/tmcw/status/966073331207233536
I'm working fulltime on a tool called TimeShark (https://timeshark.io). It generates invoices based off color coded events from your Google Calendar.
It's cool because I as well as many others got tired of spending time...logging time. A lot of us live inside our Calendars so why not build on top of that behavior?
Very much in listening and feedback mode right. If TimeShark is interesting to you, please reach out (contact details in my profile).
It's cool because programmers should be able to easily watch their algorithms transforming the data they operate on.
I think this is an area with a lot of potential for seriously improving computer programming: we need better instruments for making observations of dynamic program state (or dynamic database state, etc.). It's kind of an inverse problem to what we've needed in the sciences, though, with microscopes/telescopes etc.: (one of) our problems is that we have too much data, and we need ways of automatically abstracting and presenting just the relevant parts—something like 'abstractoscopes'. This is my attempt at making one.
I am working on testing ("ground truthing") common observational astronomy modeling techniques against hydrodynamic simulations. The basic approach is to generate synthetic/mock observations from the simulations, apply typical model-fitting approaches, and then see how the model results compare with the properties of the original simulations. Our first paper on this has recently been accepted (https://arxiv.org/abs/1803.07084) and we're preparing a second paper looking at emission from dust in new simulations which more realistically treat the cosmological history of galaxies.
It takes designs for mobile apps (from Sketch files), and tries to convert those into React Native components. It's still early days, but it's a promising start.
It's cool because if I can get it working fully, it can replace a large chunk of my consulting projects (taking designs and converting them to the initial code)
An ongoing comics series. Elevator pitch: “Imagine you’re watching funny-animal Star Trek... except every other episode is from the point of view of the Borg.” Also the Federation is actually a quasi-monastic order of Space Jesuits chasing after whatever remnants they can find of the vanished Ancients, and the Borg is a bunch of goth hippie dryads dreamed up by massive shipminds, and did I mention everyone is a cartoon animal with no pants, and it’s intentionally designed to be too big for just me to draw and I’m really looking forwards to/dreading when it has enough inertia for me to actually hire some friends to draw parts of it, and...
I’ve been spending the past couple years kicking around a massive space opera. I’m finally starting to finish pages of its comic-book incarnation. It’s good to start moving forwards on a story again, wrapping up all the loose ends on my previous graphic novel took far too long.
http://egypt.urnash.com/parallax - very much a placeholder right now, just the last iteration of the pitch bible and some links; once I have enough pages drawn I’ll start having them show up there. Right now they’re just on the Patreon.
I am working on my own startup (since 12 months, but we incorporated only in August 2017). I am not here to publicize what we are doing, but rather to share a very important consideration.
I used to work at Amazon Web Services (6 years), then VMware (2 years), then at a startup as an early employee/CTO (1 year). My job satisfaction decreased at every step.
I feel so lucky and blessed to be able to work on something that I truly care about, without the usual corporate BS.
I've wanted to build something similar but haven't gotten around to it.
Couple questions:
What do the icons at the top of the page do? (The music note, rain drop, anchor, etc) I think they are some sort of filter but click them doesn't seem to do anything. A tooltip or quick note might help the user know what they are for.
would you consider adding soundcloud integration? I have a bunch of playlists/music that I can only find on soundcloud and want a way to integrate with youtube playlist for continuous play/mixed playlists.
are you open to outsider help? I've been looking for a project to help contribute to and really like what you have started. Wouldn't mind helping with some PRs if you are open to it.
So I just broke those this week with an update & some refactoring, but they are for soundscapes, like rain, ocean, cafe background noise ect. - still thinking about how to group things
You know, open sourcing with multiple devs is something I haven’t put much thought into - at least for this project. I will have to move around some configs and do some readme stuff but growing that collaboration muscle seems important. Ill let that stir in my head, its still my baby, but I don’t see any harm, so maybe in a little:)
Soundcloud is a great idea:) I didn’t think of that either! - probably bc I don’t use it enough. Im definitely going to look into that this weekend.
Thanks and I’m really glad u guys like it this made my day:)
However, because of Apple's new strict rules on what it considers user aggregated content, privacy, and etc. It's been stuck on beta for a while. I originally wanted to create a YouTube aggregator (like yours) however, I'm almost certain Apple will reject it.
Not really ¯\_(ツ)_/¯
- personally i use adblock, and most of these mixes are 1-2 hours long, so even people having an add in between each would seem not too bad
Also i havent really put this site out there before, so any feedback or ideas are welcome:)
I'm working on a simple cooking site targeted for millenials and beginners. No long blog posts about how you used to make this with your (insert family member here) and (insert more useless information here), no bloat. Just ingredients and steps. If a recipe suggestion is voted high enough, that new recipe will replace what was originally posted.
I'm making a 2D casual/puzzle game similar to scrabble but instead of building chains of letters to form words, you build chains of words that rhyme with each other based off the picture on the tile.
Some tiles have multiple labels. A picture of a cat can be rhymed with bat or hat, buy it can also be a "pet" and rhyme with net, etc.
Please don't! Efax owns patent for sending faxes online and they troll/bully other website into sometimes buying them out sometimes shutting them down. There is 99% chances you will receive their C&D letter :(
I think this new operational take on CRDTs (which is not my original idea) has the potential to be quite revolutionary, combining all the benefits of CvRDTs and CmRDTs and allowing brand new replicated data types to be created rather intuitively—all while maintaining near-optimal time and space complexity and providing a common interface for things like past revision viewing and garbage collection.
Hope to publish the article for real in the next few days.
No, I think interest based matching is overrated. I think shared experiences, storytelling, setting common goals will be more effective. I have a list of a bunch of mechanisms I want to test this with.
I'll throw y'all a curveball - IRON CROWN (https://www.ironcrowncomic.com) is an action-adventure webcomic about mercs, monsters, and a dictator’s daughter fleeing a coup - with all of the above standing against a much darker enemy on the horizon.
Also updates every Monday, rated T+, and published through Hiveworks which is a neat indie publisher that probably hosts that other webcomic you love. Since I'm the creator (and a designer/illustrator by trade), I'd be happy to answer any questions about getting a foot in this sort of business. Can be incredibly demanding, but creating your own stories is almost always worth it.
The software runs on an industrial controller (basically an Intel Atom with a proprietary bus to connect tons of I/O modules), driving a bunch of electrical and hydraulic gear, and logging data for analysis.
I like Albatern's idea for a wave power array based on lots of small, easily-deployed modules, and I'm hoping that their tests pan out and get them to production readiness.
While I have just spent 2.5 days freezing my ass off coding in an unheated industrial shed in a Scottish winter, it was a shed full of massive wave power components - hydraulic hoses as thick as your arm, man-height generators, and so on - so that's pretty fun! When I wasn't coding, I also got to make myself useful by stripping cables and wiring an emergency stop box.
In previous client work, I scrambled around wiring up sensors and programmed control code and UIs while standing in a small shipping container floating on a tidal platform in the middle of a Scottish estuary ( https://sustainablemarine.com/plat-i ); and more recently, learned how to do mean-time-between-failure analysis for an industrial Ethernet network that needed to live at the bottom of the sea with ten-year maintenance intervals and a >£10,000 bill to bring it up for component replacement.
I'm building https://timeguard.io, which is an iOS app that acts like HN's "noprocrast" setting but for the entire internet.
You pick distracting sites/apps to block, and then you choose when to pause the block and access them. I've found that little extra bit of friction in checking FB/HN/Twitter is enough to keep me far more concentrated and present throughout the day. Would love feedback from other HN users!
Hey Kyle, this looks nice. I use RescueTime[1] for this purpose on desktop, but since it doesn't easily interop with mobile/other devices, it's still easy to slip into procrastination by pulling out the phone. I'll try TimeGuard as a way to fill the gap.
For me, the killer feature for this type of tool would be a blanket anti-procrastination mode that works across all devices. Maybe it could even make sense to integrate somehow at the router level?
A website that shows if there's a scene after credits for movies in theaters. So long for all that people standing on the stairs because a scene just started.
The best way I can describe it is: imagine Python but statically-typed with type inference, proper concurrency support (via greenthreads), sum types (like enums in Rust), and immutable data structures in the standard library (like Clojure). I may even add optional checks that a function is pure.
The stage I'm at right now is learning how type inference works well enough to be able to implement a correct type inference engine for it.
Outside list is cool because it (hopefully) encourages people to travel and get outside more.
Hollaback is cool because it's sort of an "about" page as a service, and lets people know you made the thing they're looking at, not a team, not a company, you.
I've spent the last few years building a product to help every user and business own their data. It solves Identity, Language, and Hosting, with customizable UIs.
It's cool because it can change the way we all communicate; instead of interacting with 3rd parties, you always interact with your AI that handles everything for you.
Just coming out of stealth, launch posts and whatnot coming soon :)
https://token.beteleported.com - It's a cryptocurrency token for a VR marketplace. Store 360 degree videos and photos on the blockchain, for 50 years, just $1. It's cool because you never have to worry about losing or deleting your memories.
Customers can meet with professionals virtually, without having to travel anywhere. Professionals can bill their clients, with the cryptocurrency token.
It's also an interactive map, to view published 360 photos and videos, called moments. Use the token for storing moments, viewing moments, giving tips to moment creators, and paying for professional services in VR.
https://get3dtours.com - a SaaS product that makes it super easy to add virtual tour experiences to any site.
I'm building a command line tool in python for front end development. Im self taught and never actually have built anything. Besides that, I was unhappy with everything else out there. I could either learn someone else's tool or learn to build my own.
I do have a question. I'm using urllib.request instead of using requests to download normalize and I found conflicting information regarding if it was safe or not.
Finally, can I get feedback on my website. Gotta lose the fear.
Edit: for the life of me I can't find how to solve a full height div on mobile, specifically in Firefox. It happens regardless if I use 100vh or 100%. It seems that it can only be solved in JavaScript (which I'm still searching for a practical solution), tough, I would rather avoid a js solution.
I quit my job as a sysadmin to be a stay-at-home dad.
In my spare time I'm working on a console-based mail-client, scriptable with lua, some golang-based projects (object-storage, puppet-dashboard replacement, and similar things)
Beyond that this year I'm mostly relaxing and pushing a pram in the snow. Fun times :)
No link, as this is a scratch my own itch thing for now, but I'm working on a personal 'hub' web app that is part UX/UI experiment, and part a way to integrate various tools I've created over the years to ease various pain points in my life.
Some of this has to do with solving the problem of the deluge of information I consume, through psychological 'intervention' as well as some way to store all this stuff and make it easy to filter and find the useful bits.
Some of it is journalling-related, and vaguely in the Quantified Self area.
A significant portion is about finding ways to actually use this amazing, powerful computer that I have in my pocket for things that actually, quantifiably, provably improve my life (and hopefully, eventually, that of others). This is mostly in the direction of autism-related issues, anxiety, maintaining social connections, and so on. Applying the (relatively) well-supported lessons from psychology to the bafflingly advanced technology that is part of my daily life, and that of most people around me.
Mostly it's a vanity project though, that I can currently afford indulging in. I figure that after years of working and making money off building stuff I don't care about much, it's time to let my more creative side express itself in ways that might prove useful somehow. I'm not an painter, or a musician, or a poet, but I can code, I love to code (when it's on my own terms), and I'm still amazed that for less than five bucks I can have a server running that does whatever I want it to do, and as a 'full-stack developer', that's quite a lot!
Part of me wants to work on things that are more directly beneficial to others, but 1) I have enough 'runway' to do that too, eventually, and 2) I've grown weary of always trying to make sure that what I do is 'useful' by some measure that doesn't quite feel like it originates from within. Many of the best things I've done started out as scratches to personal itches, so perhaps it's okay to 'indulge' for now.
- You install it with `npm install server` and use modern ES6+ including async/await very naturally.
- Node.js is the most popular "framework, library or tool" [1], but getting started with most of the current Node.js server frameworks has a really steep learning curve.
- Plugins are coming! They will be a really easy way to extend server in ways that middleware cannot by adding some lifecycle calls.
I'm working on https://www.framedtweets.com — it's an easy, beautiful way to preserve and remember your favorite tweets forever (to save them from the internet's ephemerality).
It's cool because REST/HATEOAS is the original and unique software architecture of the web, and building client/server apps in Javascript/JSON throws that architecture out:
https://BitBank.nz a cryptocurrency prediction dashboard with live updating charts, API access, bulk data.
Its cool because i get deep into data engineering/analytics, machine learning/AI. Its using latest Google Cloud tech and my customers are always pushing me to better the product, myself and have done lots of work in open source.
Its interesting seeing what the customer perceptions of data science/prediction systems are and dispelling some of the rumors and snake oil e.g. forecasts get less accurate with time so people shouldn't trust out of date forecasts that are forecasting a long way into the future.
https://vrsketch.eu/ - Editing and viewing virtual reality plugin for sketchup. It's a lot of basic research on UI/UX in VR.
http://pypy.org - Python interpreter with a just in time compiler. It's the only alternative python implementation out there, after failures of unladen swallow/pyston etc.
http://bloc11.co.za - Own climbing gym. It's cool to work directly with people and pursue own ideas in a small space.
The goal is to visualise all the different topics you can learn about and provide the best resources available for learning the topic.
It's also Open Source (https://github.com/learn-anything/learn-anything) so the cool part I think is that the users can actually be developers and submit changes they wish to see on the website.
I'm working on a technology that will enable people to easily distribute and copy data in a secure and more performant way, the more people are using it. It's basically throwing a local cache over the web, so when you neighbour already have the data, you can download it from them without being afraid if the data has been altered in any way. I'm super excited to work on it, as it combines my two favourite areas, open source and decentralized. The project I work on is called IPFS, the InterPlanetary File System.
I'm building a mobile app that's a bit like the anti-Instagram. It gives you confusing set of shutter/filter options that post weird, uneditable and uncanny photos to a feed whose sole mode of interaction is scrolling and drawing.
Mostly it's a conceptual art piece that's meant to offer a different model of being social online — one that's wholly about ongoing presence and participation, not accumulation and curation. I think I'll try to sell it for a buck in the app store.
https://pagedraw.io/ — keeps me from ever having to write HTML and CSS for web apps again. Enables solo developers to do more by letting us design UIs in a visual tool and not have to re-code them later. Enables developers on teams import their designer's Sketch or Figma files into code, skipping the tiresome designer-developer handoff full of tediously re-doing their design in JSX/CSS.
https://www.venturecoffee.xyz/ - A platform to support indie makers in building successful businesses. After reading Andrey Azimov's post ( https://www.medium.com/AndreyAzimov ) on medium, I was really impressed and thought of building something to help him and other makers like him.
It supports MySQL and Postgres, supports organizations (invite your non-engineer colleagues), sharing to google sheets, JSON, CSV, email, and is secure (encrypted credentials, read-only transactions, etc).
I have much bigger plans, but for now I'm bootstrapping it and so far customers seem pretty happy! Give it a try!
hey this is really cool:)
I like looking at what bags people use, its nice to see some that are 'battle tested' - for ideas on what to pick when I go travel hiking:)
I've been working on Fructika https://fructika.com for the last few months as a little side project. It's an app that helps you manage your fructose consumption (my partner suffers from fructose malabsorption).
It's cool because it makes her life better and I've had some great feedback from other people trying to manage their intake of sugars about how it makes their lives easier :)
https://yumefood.com.au - a platform that mitigates surplus food from going to waste, saving CO2 emissions, landfill and water usage in the process at a massive scale.
It also enables organisations like schools, universities and missions get access to bulk food at a fraction of the usual price.
I never thought it’d be possible to affect this type of change using code, which is quite cool IMO!
It's a little music production community. You know how everyone on HN laments that small quirky websites have been eaten by bigger ones, so they don't exist any more? My website is cool because it has that sort of community - intelligent and good at music but also funny and super weird. We all kinda know each other, which makes it fun.
Also, a bunch of us have gotten a lot better at making music, which is another nice win. :)
It grabs what I've read from Pocket and makes a page for each week.
It's cool because what I read is better than what I can write, and I think this is true for others. I've got loads of ideas for it, it's going to be massive, as long as HN like it.
https://github.com/PetrochukM/PyTorch-NLP -- Following writing a NLP research paper. Working on open sourcing the tools that helped me rapidly prototype. PyTorch-NLP comes with pre-trained embeddings, samplers, dataset loaders, metrics, neural network modules and text encoders.
I think it's cool because it's almost definitely the most complex Slack bot out there (though I admit that's an arbitrary metric), but it's for sure the most advanced game on the platform.
Very similar idea to Multi User Dungeon (MUD) RPG games back in the day. Was that your inspiration? If you're not familiar with those you should check out their history.
Absolutely something I took inspiration from, though the primary inspiration was more along the lines of Legend of the Red Dragon, a BBS Door game that hit the height of it's popularity back when MUDs hit the height of theirs.
It's an Ohm-inspired packrat parser for Emacs. It's cool because it uses the packrat cache to allow efficient incremental repairs, so your 20k line buffer doesn't have to be completely reparsed just because you added a one line function in line 7
I’m working on a technical analysis alert site for cryptocurrencies. There are a few out now, but they are limited to a few TA algorithm’s and a handful of intervals.
I’m aiming to offer as many algorithms/candle intervals as I can and make them fully customizable. Also implementing a screener and the ability to test your alerts through papertrading with historical data.
Needed a way to keep track of TV shows I watch and when see when episodes air. All existing sites I found were so messy and bloated with stuff I don't really care about.
Wanted to have a list of new episodes and a heads up about the next few days whats coming up so https://boldshows.com was born.
We value full and unreasonable transparency both in our internal and external communications. We believe a world with complete and open transparency is a better world. It has been very interesting to try to make the search engine data world more open.
It's cool because... well, it makes it easy for programmers to fill out and sign PDF forms. So it's not really a "cool" product, but I enjoy working on it, and it saves people a lot of time.
On some level I feel this question is just here to advertise the poster's project.
I think it would have been better if they had put their project in the comments like everyone else instead of giving themselves more visibility by putting it in the main question.
They may not have intended it this way - but it's possible.
I am currently working on building a system that processes and provides analytical data for a major semi-conductor manufacturing company.
It is cool because we have a serious stream of data to collect, stream, process and store, and provide users with very fast access to massive datasets.
https://www.telltali.com/ – Voice-first timekeeping for Alexa and Google Assistant. I think it's cool because we're solving real business problems with voice. Win/win.
I've founded a startup with kick-ass people to build the future of Electronic Medical Records. And it looks very bright, transparent, intelligent, and automated :)
Working on an SMS crypto trading bot where you can text it trading commands according to whatever exchange you connected to. I've been in many situations where I needed to trade and logging into the exchange on my phone is cumbersome and could sometimes take 3+ minutes, and by then the opportunity is gone.
Here's a demo: https://coinbase.onsites.co/o/aurora-h
I also redesigned the onsite experience. Demo here as well: https://airbnb.onsites.co/joe-zadeh
[1] As with many things in design, I think I should attribute the source of inspiration, which is Carta. They wrote about a new offer letter format back in 2016. I'm basically taking their Apple Keynote slides and moving them to the web.