I'm planning to mainly target developers that have resources spread among different providers and need a secure way to either connect to them or connect resources to each other.
It will be easy to setup and seamless to use. You'll just have your VMs connected between them on the same local network, so all your application will seamlessly work with it. This will potentially unlock other ways of architecting your apps.
E.g. you might have some VMs in Amazon and some other in Azure, Linode, Rackspace, Digital Ocean, bare metal in any other hosting provider (OVH?) or locally in your DC. Each of those providers might offer something different and it'll be great if you could just use them all together without opening your services to the whole world or without having to tinker with IPsec VPN tunnels or firewall rules. You just need to create your new network in our system and deploy the client with the provided config file on each of them. I'm making it as simple as providing you with a DHCP server already, so you're ready to go after launching the client. Every machine you join will be part of this virtual network in a completely transparent way.
http://www.mailscope.io adds profile data to your existing mailing list. Why? Well, if you don't ask for firstname and lastname on signup, you'll get increased conversion. But if you then want to get increased open and click rates when you actually send email, you should start personalizing (one way is to use firstnames in the body) and segmenting. That's where MailScope comes in. Each new subscriber you get, MailScope will automatically add firstname, lastname (and other profile data).
It's just the start and I've been having great fun expanding the possibilities - alerting you when an 'influencer' signs up so that you can reach out directly; auto-following subscribers on twitter when they signup. I've already got a dozen or so paying customers who use MailScope to enrich their mailing lists and increase their revenues. It's awesome to have learnt so much here on HN and finally be able to start offering something of real value back to business owners.
What did I finish? I've been rebuilding my larger quadcopter and I finally got it up in the air and it flies well enough for my needs, though admittedly I did trim my friend's tree with it a bit. I'm chalking that up to pilot error, however. I hadn't flown anything in a few months and it turns out that line-of-sight orientation on quadrotors isn't just like riding a bicycle. It could still use a bit of tuning -- projects like those are never really finished.
I also have a few fairly interesting/exciting projects which I desperately need to write up. One such example is a wide field-of-view stereo head-mounted display for FPV flying which can be built in an evening for under $200 USD. Another is my as-of-yet fruitless efforts to build a very low latency HD digital video transmission system - also for FPV flying. If these are projects which you'd find useful, or which you'd simply like to read more about - leave me a comment or flick me an e-mail . Encouragement always helps when it comes to getting things written up.
2: My username at gmail.
- We must stop writing UI by hand
- We must adopt a new language that's less ambiguous
- We must prefer graphs to trees
- We must not build service-specific apps and websites
- We must stop thinking of computer hardware as a personal device, and make switching from a device to another completely seamless
- Local storage must be nothing but a local cache of sections of the ONE universal knowledge base
- Speech and text must not be the primary way to interact with a machine/AI
- Service providers must never have a say when it comes to UI/UX, as services must be completely decoupled from UI
- Everything is an agent, the system shouldn't make a distinction between human users, AI, smart contracts, APIs, etc.
- There must be no difference between creating software and using software
- Money shall be replaced by a social currency, a form of trust score (based on reliability, honesty and predictability)
- Businesses must adapt their practices and models to software, not the opposite
- Brands must be eliminated and replaced with a trust score
- We must allow for non-precise facts to exist (ranges, set of weighted values, conditional constraints)
- We must allow for non-universal facts (opinions, contradicting values, different weight/credibility based on the user's trust graph)
- "Undo" and "Predict next" functionalities must be ubiquitous and present everywhere
- Intentions and Predictions are communicated by logging events in the future (where the date doesn't need to be precise as allowed above)
- A cooking recipe, GPS directions, task dependencies, a playlist, IKEA directions, a tutorial, must all use the same model
http://orangemind.io - my personal project, comics series(started only recently).
It currently requires something akin to EigenTrust++ implementing in the DHT namespace, except EigenTrust++ requires information about the amount of successful downloads peer nodes have made, so it's going to require minor adjustments for decentralised HTTP.
On the frontend it requires a way to insert arbitrary elements into the DOM using something akin to Mediums' impressive little editor.
Also missing RPC_EDIT, so there's no inter-instance web page editing /just/ yet.
Added a bunch of new features over the break.
Its a website that calculates the best time to exchange one currency for another accounting for the various exchange rates involved.
I'm getting interested in super low frequency signals so I looked up the E202 Very Low Frequency (<10kHz) receiver and laid out/built a variation of it. Right now the whole thing is a broadband receiver with no antenna (obviously) and the whole circuit board assembly is functionally acting like a microphone. I can hear when I touch any component or move my hand around in the air. I'm going to add a 60Hz notch file and then take it out to the middle of nowhere.
I think it would be awesome to go find a pipeline to use as an antenna...
Next project is to take my BlueROV and build a hydrophone array for it so a friend and I can see if a underwater acoustics engineer friend and I can use it to track other objects (like a remote-controlled toy boat) in the water. I've been doing some Kivy visualization of an accelerometer and gyro (MPU9255) and I think we could use matplotlib's interactive mode or something in Kivy (maybe) to visualize it all in realtime.
There's nothing cutting edge here but I've done a bunch of radio frequency (RF) stuff like GPS and WiFi and I'm really enjoying how tangible audio seems in comparison. Just having fun with low frequencies, basically.
The idea being that even if we do great things and are very active in different areas of our life, we actually end up doing the same thing again and again. If we track the things we do, and look for something that was different and special, will that entice us to do more diverse and interesting things, and get outside of our comfort zone?
It's very MVP, just seeing where it goes at the moment, but so far, it seems people like the idea.
I also built it in Meteor, which I only tried for the first time on Dec. 30th, and I have to say, for prototyping something basic like this, it's been really great. Some of the poor quality of the site (like slow load time) is probably due to my inexperience with Meteor.
I started in a few years ago, but over the Christmas break I put a lot of time into it and it's growth has spiked quite well as a result. I need to upgrade my servers now :/
As the organizer of HelsinkiJS and a new dad I find it takes too much effort to organize monthly events. Meetabit makes this easier by letting companies offer sponsorship & speakers submit talk proposals. All that organizers need to do is pick a date. The service handles sending out invitations, handling registrations, providing a wait list and even getting speakers to add links to their slides after the event.
To see some of the features available, check out the HelsinkiJS community profile: http://helsinkijs.org. If you're an organizer yourself, it would be great to hear from you - just drop me a line via the feedback link in the footer.
Also started working on an Android app that talks to my Anova 1.0 (bluetooth-only) because I wasn't happy with the official Anova app. Was great to learn about controlling a device over bluetooth from Android, which was simultaneously a pain in the ass and easier than I expected.
1: Cheesy Example, but shows the basic premise of the goals interface: https://vimeo.com/58997158
Not sure if it would actually be useful to anyone in the real world, but I've learned a lot about writing PostgreSQL extensions and such.
https://github.com/chc4/urporn the name is tongue-in-cheek, please don't actually use it for porn)
For fun, check out https://github.com/chc4/urporn/blob/master/home/ape/porn.hoo... and see if you can actually figure out what the hell is going on :D
This means all the posts are cryptographically signed and stored in the blockchain.
I was quick to build the Nodejs part of the framework. I have been taking a longer time building the Go part. Mostly because I am new to Go and am trying to figure out whats the best way to go about it.
These days I am reading a bit into Dagger, and trying to figure out if I can put some learnings from Dagger into Archiejs (or particularly the Go part of ArchieJS).
Many improvements to shoebot - (A cairo port of nodebox).Experiments in graphics for python - started a midi mapper / OSC controller in kivy.- learned about bezier curves.- wrote various music vis experiments in shoebot and nodebox-gl.- wrote a couple of simple VJ apps and actually VJd with them.- Experimented with building native android apps using SDL.- Did numerous opengl + shader tutorials (and contributed fixes).
Currently finishing off a sort of 'leaderboard' of chapters for a book writer (to choose the order of the chapters).
A bunch of other things - vext - a way to use libs like Gtk from virtualenv)- time learning about 'boring' stuff, testing, packaging etc.
It's been quite a bit of fun ... hopefully when I start back at work can use some of this knowledge and not be just 100% back on the backend work.
Doing all this has been really good, and if you are a contractor and have the chance I'd recommend doing something similar, have been very lucky !
https://jjuliano.github.io/markdown-ui - Create Beautiful and Responsive Websites in Markdown Syntax
 http://www.fullstackpython.com/change-log.html https://github.com/makaimc/fullstackpython.com/commits/gh-pa...
http://www.peergym.com let's you search for quality gyms in your area by membership price and amenities - the kinds of things services like Google Maps and Yelp don't do. Most people do a particular kind of workout (running, weights), and need special equipment (treadmill, barbells), and you can't always guarantee you'll know what you're getting just by the name and a few pictures.
It was mostly an excuse for me to learn Elixir and Phoenix. I've tackled auth, uploads, geolocation/geospatial DBs, SSL and more, so it's been a lot of fun and hopefully I can turn this into some sort of tutorial series on building out a real-world app.
For the future I want to add reviews, community edits, and advanced filters to make them easier to search and populate. And hopefully, accept payments if people want to buy passes to their gyms online (or automatically renew their memberships.
A product update timeline (change log) that you can quickly add to your website: http://productmap.co/
Will be ready for a beta launch at the end of this week!
Now struggling to get REPL working properly.
I have plans for pricing, but before I can approach that, I need more data on how people will use it.
For technical folks, it uses Phoenix and Elixir, and a lot of JS.
It's somewhat like the "Releases" feature of Github, without everything else. A minimalistic selfhosted-internal-appstore-slash-customer-service-desk if you will.
I started it about 6 months ago on Leanpub and iterated it openly. Interesting journey so far.
It's still early so i only have cameras listed, and things may break. I'm adding tvs next. Let me know if you have any requests.
Additionally you get Trusted Advisor checks and CloudTrail event notifications, which you can e.g. use to get alerted on unauthorized API access.
I think the biggest barrier to privacy online right now is how inaccessible applications like Bitmessage are to the average user. Having to install a local python code base and store gigabytes of data that takes potentially hours to download sucks.
In summary, an application that:* Provides completely anonymous, encrypted, untraceable, uncensorable messaging between people and groups (using Bitmessage as the data store)* Is accessible like any other web application and provides pretty good security in doing so* Can be used as a browser extension if you want virtually guaranteed security/privacy* Relies on a data store that no one owns, everyone can access, and everyone can forever contribute to
There are questions about the security of Bitmessage, but I know it will be improved over time.
It is a struggle to decide how much time to dedicate to this application, though. I'm a somewhat underpaid developer with aspirations of actually making money to support myself with side projects, but at the same time wanting to contribute with open source applications like this to make the web a better, more private, place.
- First blog post, benchmarking AWS S3, Google Cloud Storage and Azure storage, http://blog.zachbjornson.com/2015/12/29/cloud-storage-perfor...
- MongoDB driver for Mathematica, https://github.com/zbjornson/MongoDBLink
- Started packaging bioinformatics tools for Mathematica (a la Bioconductor), https://github.com/zbjornson/BioTools
* https://twitter.com/hn_frontpage displays Hacker News top 30 (to deal with ranking volatility, an article is tweeted once a day) and has both article link and discussion link. Haven't found an existing one that is both the front page and the discussion link.
* https://twitter.com/tic_tweet_toe for people to play games of TTT. It remembers each person's W/L/D record.
Working on a web version of cribbage that isn't a Java applet.
Researching idea for my next bootstrapped company.
A microsubscription (cf. Spotify, Google Contributor, Apple Music etc) system that works on the open web while remaining robust to attempts to siphon payments through fake visits.
As is typical of an engineer, I spend entirely too much time running down technical rabbit holes. Thanks a lot, ADHD. My current rabbit hole is getting everything into Concourse CI.
It uses shared memory to store the stack, thus making it "global". Still needs some polishing.
Check it out. feel free to give feedback, pull request, whatever.
- Snub: Manage .gitignore files from the terminal or from the status menu bar. Free and open source: https://github.com/ashokgelal/Snub
- LightPaper: Finished rebranding, just released 1.2 alpha and continuing the development: http://lightpaper.42squares.in/
I got frustrated with GTD a long time ago. I was super frustrated that my todo apps would eventually be full of useless information. I loved the ideas of some bloggers to basically just pick a few tasks for the day and do them. I thought: "Why not mix the calendar events and tasks in such a way that you have software that actually encourages you to focus only on the tasks you have time for?" Hence, you have your "real day" as opposed to the day you thought you'd finish 20 huge tasks and end up with none attempted.
realday.co (Had MVP up like a few years ago but scraped in favor of current Golang/ReactJS version I'm building now)
1: Work in progress
I was inspired by posts like this one  to give the free monad a spin and after ~300 lines i'm almost done. But then I found other posts that talk about free and cofree  that I still can't really understand so I guess there still is some room for improvement.
 http://www.haskellforall.com/2012/06/you-could-have-invented... http://dlaing.org/cofun/posts/free_and_cofree.html
I used to be one.
https://kernl.us - Come check us out!
It will give you tips on what to change and let you time your query and see other resource info.
Anyone interested in being a beta tester?
Right now I'm focused on what we need for https://foldapp.com, supplementing what we've already built (first Python, then Scala).
Coffee subscription service based in New Zealand.
Tech wise it's all Go on Google App Engine with a dash of jQuery and I had to extend the library for Braintree payments to work with GAE.
Anyone interested in using/testing/anything please drop me a mail: nsglynn at gmail
2.An app for salespeople. It helps track prospects, keep sight of targets and communicate with home office staff and other members of the team. You could say I'm a SalesForce competitor.
The twist is, it's a single page app that doesn't use any JS tags.
You can play it on my test server here . Consider it a puzzle to figure out how it works.
If you want to see the source, it's on GitHub here .
 http://220.127.116.11/ https://github.com/pimlu/connect4
The idea is that machine-hours are cheaper than man-hours, and the hardware to run real browsers is cheap enough that for non-trivial apps, it often makes more sense to run real browsers.
Intro here https://m.youtube.com/watch?v=FJTYcCjYo2g
Sign up at https://app.hapii.co
The biggest hurdle in the design right now is figuring out a mechanic to encourage people to visit the site regularly
Out of curiosity, what kind of model (loss function) are you using?
 https://github.com/samuelngs/go-mithril https://github.com/samuelngs/go-sphere
I'm building infrastructure that I'm hoping will be needed to complete the Stockfighter trading puzzles.
(Hoping, because it's entirely possible that I will be lied too far less than the documentation leads me to believe, and I won't need all of this sanity checking.)
In summary, buy reliable used cars that are rough around the edges and with low asking prices. Spend leisure time doing necessary repairs to ensure reliability, safety. Spend time doing detailing work, which in the future may include repainting panels, and basic underbody rustproofing. Use as a car for 2 to 3 weeks as a burn-in to ensure I have something that meets my quality standards. Sell for 125-140% of cost. Limit yearly sales to keep under the transaction limits for a non-dealer.
* Excluding first car, cost me ~$2500 for good tools and safety/disposal equipment.
* Usefully leverages my vast knowledge of the automotive landscape.
* Improves my sales and negotiation skills, which will benefit Business #1
* Improves car repair skills and detailing skills, which is useful since I'm a hardcore car-nut.
* Turns a hobby into something that makes money. I'm having a blast and making money doing it.
* Unable to move a car, reducing profit, which means time not well spent.
* Poor assessment of a car I purchase for resale. Risk eating all profits or taking a loss.
* Losing interest. Though I'd have all the tools I'd ever need for my existing toy.
* My spare time I could spend on other things.
* Fraud, which I'm taking precautions against. This includes things like payment issues, or buyers not completing title transfer and doing terrible things.
Where it goes next:
* Nothing exciting for a while. Sticking to things like Corollas, Civics, Camrys, Accords, CR-Vs, RAV-4s, F-150s, etc. Known quantities with easy parts availability and consistently strong demand that are easy to refine my process on (assessment/inspection, negotiation/paperwork, repair, pre-sale QA), and learn how to properly do bodywork at an acceptable level in a private garage.
* In 2 years the hope is to move into still easy-to-move cars with a higher profit-per-vehicle. Lexus ES, Lexus RX, Acura TL, Acura MDX, Mazda MX-5, etc.
* In 4 years the hope is to start going after cars people really desire on the used market. "Affordable dreams." Things like older BMW M cars, AMG Mercedes, Subaru WRX/STis. Late 90s Japanese sport coupes. This has the potential to bring in enough money to cover rent, utilities, health insurance/care, and groceries in full.
Business #1 will always bring in 4-5x as much money, but having a hobby pay the bills 5 or 6 years from now? I'm stoked. Since there's no pressure for this to pay the bills, I get to enjoy it too.
- aseprite (http://www.aseprite.org/) is a cross-platform animated sprite editor by David Capello. I got FreeBSD support working  and added a shortcut to center the canvas . This was my first C++ commit.
If you like retro game art, definitely stop by and check it out. It's GPLv3 and you can build it for free, check out their [github](https://github.com/aseprite/aseprite/).
- uMario (https://github.com/jakowskidev/uMario_Jakowski) is an NES Mario clone in C++ and SDL2. Check out the youtube video . I got it building on Linux, FreeBSD and OS X 
- Then a couple of tiny C projects with CMake that are sort of boilerplate ATM, but they're my first C programs. The cool thing is they use permissive licensed libraries and build across platforms since they use CMake. https://github.com/tony/sdl2-playproject / https://github.com/tony/reST-lex-byacc / https://github.com/tony/ncurses-example.
On that front, I'm reading a book called "Compiler Design Using FLEX and YACC" by Vinu V. Das, which has been going good. As well as Lazy Foo's SDL tutorial (http://lazyfoo.net/tutorials/SDL/).
- Another thing to mention is automatically rebuilding / reloading scripts when a file is saved. I started using entr(1) for that, http://entrproject.org/. Previously on projects like tmuxp and vcspull I've used sniffer (and looked into watchman) but have found this works best cross-platform. FreeBSD has file watching a bit trickier since we don't have inotify or fsevents.
- On the dot-config front (https://github.com/tony/.dot-config / https://github.com/tony/vim-config) got virtualenv + python 3 + vim working together , as well as neovim fully compatible with my standard vim-config. I'm now using neovim full time and all my plugins without any problems. They loads asynchronously with NeoBundleLazy and the autocompletion is async thanks to Shougu's deoplete.nvim .
 https://github.com/klen/python-mode/pull/609 (https://github.com/klen/python-mode/compare/develop...tony:p...)
Also there's kaggle - https://www.kaggle.com/
For front-end stuff: http://2016.render-conf.com/
For people who lead tech teams: http://2016.theleaddeveloper.com/
Disclosure: I help run them ;)
If you're an iOS nerd, you might like these conferences:
Yosemite (March, USA): http://cocoaconf.com/yosemite/
NSNorth (April, Canada): http://nsnorth.ca
UIKonf (May, Germany): http://www.uikonf.com
360iDev (August, USA): http://360idev.com
iOS Dev UK (September, Wales): http://www.iosdevuk.com
Release Notes (September, USA): http://releasenotes.tv/conference/
Cocoa Love (October, USA): http://cocoalove.org
The 2016 conference details will be announced shortly in the new year.
e.g., DEFCON, Chaos Communication Congress, HOPE or Demoparties from the DemoScene or BioHacking conferences?
There are also really wonderful tech talks, and this has pretty much become the focus of the event. In addition to dev talks - which range in topic from crypto to mesh networking to solar power monitoring - there's also material on 3D printing, drones, beer brewing automation, high-tech gunsmithing, and radio communication.
There's also a beer exchange cum key-signing party which has become a hillariously awesome tradition.
It's great fun and a great place to learn things you didn't know you wanted to learn.
Other than Porcupine Festival, I'll also echo other people's suggestion to attend PyCon. It's more of a cultural event than a dev conference per se, but it's a really great gathering. And being in Portland, it's surely going to be quite a party.
Speaking for my own field, Real-World Cryptography should be mostly understandable (and entertaining) to a programmer and enthousiast cryptographer. (CHES and EUROCRYPT are also very interesting, but require a lot more background.)
(Also consider e.g. ACM, Usenix, and any local interest/user groups.)
June 22-24, 2016 - Denver, CO
If you live near by and want to do deliver a short talk or lead a workshop, have a look at the Call For Sessions.
Video from last year: https://www.youtube.com/watch?v=LI2yOM4tODw
Disclaimer: I'm co-organizing this event.
The sheer size of vendors and attendees is staggering. They keynotes are polished and generally reveal exciting things. The tech talks are numerous and sorted into 100, 200, 300 and 400 tracks based on how technical they are.
If you're building stuff on or for AWS you won't go home without learning something new.
Develop Denver - https://developdenver.org/
Rocky Mountain Ruby - http://rockymtnruby.com/
It is mostly Linux but a lot of other related stuff gets in.
I've heard very good reviews about previous editions.