DEV Community: Arbuckle

Star Sailors Returns

Arbuckle — Sat, 21 Jun 2025 22:31:08 +0000

Hi! I'm Liam, and I'm the lead engineer behind the citizen science protocol "Star Sailors". I've decided to start writing development logs again so our new players and partners can stay up to date with what's going on - and hopefully I can bring a few new users in along the way.

The last post I wrote about Star Sailors was way back in the middle of the Buildspace S5 Incubator...here's a link if anybody wants to check that out. We've come a long way since then, with over 400 commits, around 1 million lines of code changed...and plenty of other meaningless stats.

I'm going to do my best to keep this post relatively short, but I'll be aiming to provide updates every week and will provide more comprehensive documentation on our development website.

What is Star Sailors?

Star Sailors is a full stack application that allows anyone, regardless of education level or occupation, to participate in citizen science research projects, by either classifying data from research projects, or providing data from the real world.

What is citizen science?

Citizen science is a term used to describe projects that members of the public can participate in. You don't need to be a "real scientist" or working in a specific field to participate and help out.

Citizen science projects often involve providing participants with large datasets (usually images) and asking users to describe and annotate what they see. This can be useful for projects across a wide range of fields, like Astronomy (for example, Exoplanet Hunting) and Meteorology (Cloudspotting), and is incredibly accessible - all you need is an internet-connected device.

Some projects may involve users going out into the real world and looking for data that researchers have requested. This can be useful for conservation projects, where researchers are provided with sightings of rare birds and their locations (as an example), helping to track their migrationary behaviour. These projects aren't always accessible to everyone, however, as they often rely on participants being in a particular location.

Why Star Sailors?

There are a few websites where you can contribute to citizen science projects already - the most famous is Zooniverse. Some of these websites have hundreds of thousands of users - and dozens of completed, successful projects. So a reasonable question might be - "why bother making yet another citizen science website"?

These platforms are amazing, and I've spent plenty of time on SciStarter and Zooniverse, so it's not because I feel that these platforms are bad. However, my issue with these existing platforms is they're very static, in terms of narrative & environment. If you discover a planet in "Planet Hunters" project, you can't then watch it collide with an asteroid you discover. You can't populate it with alien messages you find in SETI's "Are we alone" project.

In short, there is no gamification and no link between projects. And I feel that this affects the ability for these other platforms and websites to attract more casual users, and to expand their data collection abilities.

The idea for Star Sailors (at least, for the near future) isn't to create some interactive, open-3d world to explore and build and compete. It's just going to be a shared map, with a lot of data, where you can jump on for 20 minutes when you're on the train, or walking outside and see a bird that interests you.

The goal is simple - Star Sailors is a world composed of data from various citizen science projects, and the narrative and events that occur in this world are shaped and driven by you - our users.

What's next?

We're heavily focused on user growth, so we're working on a few key elements:

Adding more projects & more content - primarily across the astronomy & geology sectors
Increasing incentives for users to invite their friends
Improving the tutorial & onboarding experience

On the technical side, we're working with our partners to improve the process of sending our users' classifications & discoveries back to researchers so that they can benefit quickly from Star Sailors.

If this is something that interests you, check us out at https://starsailors.space

Star Sailors V2 is publicly available...sort of

Arbuckle — Mon, 01 Jul 2024 05:30:17 +0000

(and so begins my long-await return to Dev.to)

Hey. Welcome. Bonjour. My name's Liam, and we've just published the first pre-release version of #Star-Sailors Version 2. So I'm going to be talking a little bit about that here.

What is Star Sailors? Well, the simplest way to put it is that it's a series of protocols and practices for meaningful citizen science games. Right now, its main form is a web application that allows you to visit different planets and collect all sorts of resources and data. We've used the highly appropriate domain name starsailors.space which I am still surprised was available when I registered it in September 2023.

I've been working on Star Sailors for a number of years, and last year I came out with the first version of...well, anything. Before that, it was a few years of code snippets and not much else. Version 1 of Star Sailors was a website that showed you lightcurve data from exoplanet candidates and allowed users to classify them. It had a simple interface and some basic gamification and community mechanics.

One day I will go into more detail about what I was doing in the years prior, but right now I want to focus on the big announcement: Version 2 is available for basic testing.

What is version 2?

It's still a web application
Users still get given data to classify

But...

It's more like a game (now)

There's resource management
and more stuff
like

More citizen science modules (mars rover data, martian cloud spectroscopy, wildlife tagging & conservation, etc)
Collaboration between users
Missions to complete
Vehicles to travel in
Bases & structures to build

What does "pre-release" mean?

Essentially
we have something working
that I want to share

but...

it doesn't meet the goals/features I want for Version 2 (all the features described above, and a few more surprises)

Right now, Star Sailors is going through the fifth season (our third overall) of Buildspace's Nights & Weekends incubator.
As part of this incubator, we have to get users to test out what we're building
Additionally, we don't want to put in the huge amount of work to get to V2, release it, and find that it has a bunch of problems or users find it confusing

So...
We're doing pre-releases

We publish a new version weekly
With new content, items, things to explore, and hopefully some fixes/improvements from the week before

This will culminate in a full release in about 4 weeks (from time of publishing - 01/07/24)

But you can start playing it now

So, please do

And let us know what you think

And come back every now and then to see how we're getting along

And, feel free to get in touch
liam@talonova.space

Democratising space research

Arbuckle — Sat, 17 Dec 2022 15:29:43 +0000

*This is an excerpt of an article I'm writing for Developer Economics about the importance of free information & focus on science in schools.

I live in a city where there's very few opportunities to meet and learn from world-class and influential figures in the scientific community. Australia has a very embarrassing space agency, and for years we didn't even have that - which is why so many of our brightest and best go overseas for opportunities. Whenever I hear about an influential scientist who I can learn from visiting Perth, I immediately have tickets to that presentation or lecture. When it comes to my love of learning, I am a perfectionist and must have access to the best sources of information.

This is why I was so excited to be able to attend a lecture by the Director General of the European Space Agency, Johann-Dietrich Wörner, when I was in high school. Due to attending his lecture, I became passionate about the advancement of aerospace as an industry in Australia. I asked questions about space policy and about opportunities for Australians. It was made very clear that people from all countries SHOULD, by right, have opportunities to contribute and even more importantly, that scientific information should be shared freely, to all people.

My eyes were opened to the fact that so many people are genuinely interested in science and space travel, but it's presented as a media novelty to most of us. Astronauts get their 15 minutes of fame and Nobel-prize winning scientists will make a special speech, but none of us see the years of training, R&D and passion that goes into each media story about the Curiosity Rover or the fight against cancer. As a consequence, information that should be shared freely is locked behind a paywall, and the opportunity to really and truly learn and contribute is left to the elite few who were given the sort of opportunities many can only dream of. This is why the anti-science movement has started - because the masses not only do not understand the science, they are actively being forced to not understand.

When I went to this lecture, I became aware of the basic right for every country to have a place in space. I decided that I wanted to be part of the process to democratise science and provide equal opportunities for everyone. Because the only way for extraterrestrial life to be discovered, or for more effective cancer treatment to be utilised, is for science to be given to the masses. Who knows, if my dream comes true it may happen in Australia.

Combatting misinformation

The internet was envisioned as a place where knowledge could be shared freely and openly. Instead, large parts of it have become havens for hate speech, misinformation and destruction.

In 2023, I'll be providing my second Astrophysics course (my first one finished back in September 2022) and one of my goals is to track down and destroy as many misinformation sources on the internet as possible. One of the things I'll be doing in each weekly talk is answering questions about specific issues (like vaccine hesitancy) and using science and logic to debunk the lies.

“If you’re not allowed to question it, it’s not science, it’s propaganda”

This is why scientific journals are full, cover to cover, with people questioning absolutely everything—mostly other scientists.
This is why, whenever a scientist says something, a dozen other scientists pounce on them, eager to find any fault or flaw, no matter how insignificant it might seem.
The thing so many people fail to understand is this:
The only way to question science is to do better science.
You don’t “question science” by making a YouTube video claiming that the Earth is flat, or that vaccines don’t work. You question science by carefully collecting and analysing data, and present the conclusions by submitting it to scientists for criticism. That’s how all scientists do their science.
Nothing in science is published, much less accepted, until a whole flock of scientists have failed to find any flaw in it.
So when Dr. Fauci says something on TV, there’s not much point in questioning it anymore. It comes pre-questioned by hundreds or thousands of people who have studied the issue for decades, and whose entire careers were based on finding faults with it.
And some tings, like gravity, quantum mechanics, relativity, or evolution, comes pre-questioned by quite literally millions of people, who have all repeated all the experiments involved and have very explicitly been encouraged to find a better explanation.
That “better explanation” often involves a return ticket to Stockholm, a gold medal and a million dollars, so it’s not like people don’t try.

Exploring what my game will be

Arbuckle — Thu, 16 Sep 2021 16:31:29 +0000

I've been wanting to tell a story set in my own science fiction world for the longest time. After countless times restarting the "world building" phase of writing, I decided that instead of simply writing a story, I wanted to make a video game based on my story. It helped that at the time I was going through a computer science course at high school, and due to my Wordpress website hobby my interest in programming was at its highest point so far. I continued writing the story down as if it were a narrative, but really just so that when I actually got around to developing the game I'd have a full plot and world built, so I wouldn't have to waste time doing anything in that area.

If we cut out the time since 2019, when I really started thinking along these lines, I'm now at the stage where my team and I are finally able to develop our dream game. We've gone through a few plot changes to make the game more accessible (which is ultimately why I'm writing this comprehensive narrative to completely identify the story we're trying to tell), however these original plot points from the first narrative idea are still remaining:

An original ball sport based on jetpacks & simulations that would be played by different aliens around my fictional galaxy
The home planet of the protagonist would be "destroyed" (or rather, captured) and a key part of the plot/quest would be finding it again
There would be a focus on cooperation with other alien species and emphasis on how "newly discovered" species (such as Homo sapiens) could/would be integrated into a "galactic society"

As time went on, I started wanting to use my passion and skills in development to contribute to the scientific industry, and at the same time I began working with my new team on various blockchain-related projects. Recently, we've decided to make a "suite" based on the game, rather than simply the game itself. The first product is the game, and the game AI components. We're wanting to directly integrate citizen science (more information at Zooniverse into the game, as well as make every item in the game (which will be a controlled sandbox/open-world) an NFT available to be interacted with on an online marketplace (if you were to look at my previous discussions around a proposed software platform called "Arcadia", this is literally just an extension of that). I also knew that I wanted a companion app so that users could have minimal interaction with their character/game from wherever they are, and for this I started reaching out to various developers such as Myriad Creative Services to integrate parts of our product into their games, in return for creating a companion app for their product to increase user engagement.

Anyway, we've striked partnerships, received massive investment (yay!), got a partnership with Swedish band Veonity and are planning to do a Kickstarter campaign for our game in early December, as well as adopting a similar approach to Star Citizen in terms of funding our game through an online marketplace (our marketplace, however, will be based around NFTs & smart contracts and designed in a way that makes the game still mod-friendly (open-source) and not pay-to-win).

I wanted to share the narrative with you guys so that if anyone is wanting to get inspiration for their own projects, I can contribute to that.

Here we go!

In the beginning, there were the forces. Electromagnetism, the phenomena of interaction between magnetic and electric fields. The Strong Force, which binds the quarks that make us together. The Weak force, which is responsible for the eventual decay of all matter. And finally gravity, the force that binds everything together. Over time, .....

Read the rest of the article here

A comparison of web platforms

Arbuckle — Sun, 11 Apr 2021 10:42:06 +0000

I've been fortunate enough to work with a number of platforms for website development, including Wordpress, Jekyll and Bootstrap. However, I've gone through several "phases" or periods in my journey as a developer in terms of what I do and want to accomplish, and as such my needs when it comes to web development have changed since I created my first Wordpress.org site in 2015. Today, I want to talk about what I use and my opinions/overall thoughts on these platforms.

Starting out

In 2015, I was starting to become a bit of a tech geek, interested in new phones and computers, and I started learning about websites and wondered about building one. I was working on building a replica astromech droid and thought it would be cool to build a weblog, or blog, about this topic. The first few websites I built were simple drag-and-drop generators - uKit, Wix and Weebly were the platforms I used. However, right from the start I knew there were a few things that I wanted included in my website:

Forums/Discussion board
Photo gallery
Newsletter subscription

Number 1 was especially important to me. I'd shared my build log on sites like astromech.net (a story that I'll share another time) and I wanted to create a community of robot builders - not necesarily just astromechs - that discussed the technical workings of these beautiful creations. However, almost all platforms I found that were "free" either didn't work due a lack of features, or they were ad-filled messes that looked horrible. I learnt a little about web layouts and UI/UX by using drag-and-drop solutions, however without access to a credit card I was never going to get very far (never mind the fact that my data would never be owned by me and instead would remain the property of the website provider).

Long story short, I found someone willing to sponsor me for a wordpress.org site and after some trial and error I got together a good-looking website with plenty of plugins and functionality. However, there were a number of problems that I'd frequently run into:

Long loading times - partly due to plugins and the hosting provider
Frequent errors and breakages - plugins
Unable to backup properly due to the size & complexity

For all the benefits of Wordpress as a CMS and web platform, there's plenty of downsides as well. Its reliance on plugins is both good and bad - it means that the platform doens't need to have tons of mostly unneeded features built in and opens the doors to thousands of developers - but the open-source nature of Wordpress means that these plugins vary in quality and support:

A free plugin is shared on the wordpress network by a single developer
The developer treats it like a side project, maybe getting some donations from a slowly growing userbase
People start finding bugs with the plugin and feature requests start popping up on the support page
The developer gets burnt out and starts spending more and more time on his plugin(s) until eventually he loses interest

There were a huge number of plugins that I used (mainly oriented around Buddypress) on my wordpress install that were a few years outdated, which led to security risks, a slow website and plenty of other issues over the four year life-time of my website. Eventually I started programming and developing themes and plugins for Wordpress, but another problem was becoming apparent: after a few years of trying to build this "robotics network", the website was 1) rapidly morphing into something else and 2) there was virtually no user engagement (this was discussed in more detail on the SK devlog, linked above).

I had, as I mentioned been introducing coding into my Wordpress experience for a while, but I wasn't skilled enough to get my site to where I wanted it to be by maintaining everything by myself. Add to the change in purpose of my "web network" (including going into game development with my newly created team), and a major breakage of the Wordpress site, and I decided to go it alone and venture into the world of Bootstrap & static site generation, which I had a little more knowledge about in terms of development.

Tomorrow, I'll talk about Jekyll & Hugo, two developer-friendly platforms often used for journals/blogs & documentation sites.

Hashnode Elearning Platform - AWS Amplify 2021 Challenge

Arbuckle — Mon, 01 Mar 2021 07:04:25 +0000

Something my team and I have been wanting to add to our software suite for a while is an elearning platform. We've all used one, whether it's to learn new skills for work or maybe to get help on a side project that you're hoping could sprout into something bigger.

Having said that, we feel that there are a number of flaws with existing platforms, including high price barriers for [school] students, a gap in the market for intranet/extranet usage, and poor user engagement/retention userflows. With this in mind, we decided to start working on an elearning platform at some point in 2021.

The Amplify Hackathon was something that was perfect for this. We often have trouble getting started with entirely new projects, so I thought this would be a good opportunity for me to get back to messing around with hackathons and be able to have something tangible by the end of it, something I could then take back to my team after the hackathon finished and keep working on it.

Just to reiterate: this was an entirely new project.

I decided that the backend would be created using Flask and I experimented around with React, Vue and Shoelace.style trying to decide which would be the best solution. I spent so much time trying to make pages from scratch that before I knew it, my home city was back in lock down, it was February 15th and I hadn't started creating anything with Flask either. It got to the point where I just had to rush everything, so I started designing things in figma and then exporting them to plain html & css. Unfortunately, amina didn't do a very good job with the Javascript and I wasted valuable time messing around with that. This was a massive project for me and with everything else that was going on I just wasn't able to get anywhere close to finishing. However, I'm excited with the groundwork I laid during this hackathon as it means I'm that much closer to adding another eventually finished product to my team's suite.

Tools, front-end and libraries
The app was design using Figma and the front-end in Microsoft Studio Code (unfortunately just the login screen for now). The first choice was to customise elements from the sholace.style library. Shoelace is very light and works smoothly across any web application and it is likely that we will be trying to bring this library into SK’s eLearning platform to cut development time.

Another library used for illustrations, that I highly recommend is Undraw.co . There are some very useful illustrations and all of them are in vector format, which allow us to edit each layer of the illustration.

User flow

There is not much about to say about the user flow. It is very intuitive, simple and not different than other eLearning platforms or even social media. The idea is to make the platform easy to use and manage, accessible to every level of familiarity with webapps and the internet. The user can create a new account by defining a user account (no email necessary) or login into an existing account. Also, users can be invited by content creators to the platform via email or by a shared link.

The first screen that the user see is a dashboard to monitor progress and access the courses enrolled. Relevant information such as percentual of courses completed and access to the latest courses, displayed at the beginning of the list. For content creators, or teachers, the courses created by them will be available on the dashboard. Information about each course will be presented, such as the number of students enrolled, engagement and if the course is published - available for search in the platform. Users can also explore other courses by accessing the tab Explore, giving it more opportunities develop new skills.

Courses have no limit in the number of chapters or individual lessons. Chapters can contain videos, quizzes, tests, text or zip files to be downloaded by the user – resources for instance. Each course can also have deadlines by chapters or the whole course. The course structure is very flexible, accommodating any type of subject.

Visual identity (the fun part!)

The general visual identity of the application reflects what is trending in the web/app development, making it friendly and visually pleasant to any type of user. In order to make the application easy to navigate, many design principles are applied. For instance, most elements, including typography, present elements with smooth edges, following a principle called contour bias. This principle states that element with round edges can be perceived as friendly, and therefore, can make the user more comfortable using the application.

Back to typography, the main font used is an open-source Sans-Serif from Google Fonts with smooth edges called Open Sans. To create contrast, another Google Font with serif, creating balance and a visual balance among elements that represent different level of information.

A light pastel colour scheme also features the platform, avoiding eyes strain and reinforcing the friendly image of the application. These colours are also presented in gradients, a recent trend in the market. This also complement the idea of fun and friendly represented in other elements in the composition.

I personally had never used AWS Amplify before, however I had used Robomaker & Sagemaker on a previous machine learning hackathon, from NASA with my team in 2019-20.

A flask full of Pythons
I haven’t really been able to get much of the flask backend working yet, but here’s the eventual plan for the flask-sqlalchemy-based database:

User - Username (Done) Password (Done) - hash function with werkzeug Email Address Local time (Done - automated) User ID (Done - Primary Key) All this user’s posts (Done - linkes to “user who posted” from Post class) Enrolled courses (as a student) - linkes to “Course ID”/”Enrolled Students” in Courses class Posts (like status updates) - Title Body (Done) Time of post (Done) User who posted (Done - Foreign Key) Post ID (Done - Primary Key) Courses (One user can have many courses) Title Description (like Body in Post(s) class) Teacher/Instructor - user foreign key Enrolled students - user foreign key Course ID (Primary Key) Video ID (Foreign Key) Any extra content…. Videos (One course can have many videos) Title Desc Course ID (Foreign Key) Video ID (Primary Key) Extra files/content

We hope to save the watch state of a user/their progress through the course so that on their student dashboard they’re able to jump straight back in.

User bookmarks could possibly be implemented with a form and then a db.commit().

Resources
Documentation & Files

Elearning platform

Arbuckle — Sun, 21 Feb 2021 07:33:24 +0000

https://findcollabs.com/project/elearning-platform-UoJJdzbbtUSNXdfszZo5

My team, Signal Kinetics, is working on an open-source ELearning platform that is designed to be a one-size-fits-all solution. Companies will be able to upload content to a private intra/extranet for their clients/team to learn new skills, while teachers or instructors can create and distribute courses freely with minimal effort.

I originally came up with this idea as I'm doing freelance work for a consulting company that wants to have an onine portal for clients to browse content, and I thought that it would be great to create an open-source platform for anyone to create courses and learn new skills. I'm planning to build it using Flask and in the last week we've made a little bit of progress with it.

We're always looking for new teammembers and we'll take anyone!

Stay safe,
-L

Hacking Rainmeter for my setup

Arbuckle — Wed, 16 Dec 2020 21:02:55 +0000

Rainmeter is a free and open-source desktop customization utility for Windows, released under the GNU GPL v2 license. It allows users to create and display user-generated customizable desktop widgets or applets called "skins" that display information.

Wikipedia entry for Rainmeter

Table of contents:

Suites
NXT-OS
Gemini

There is a huge community of developers and artists that create the skins that make Rainmeter...rainmeter. You don't start off with much after installing Rainmeter, just a few system monitors:

And that might be enough for you. Some people just use Rainmeter to keep an eye on system resources, and in fact a large number of the most aesthetically pleasing and most popular Rainmeter skins contain skins that are used for that very purpose. An example would be NXT-OS:

Sometimes you want to have a skin that can remulate a movie or video game:

A very popular choice amongst Rainmeter hackers, the Jarvis/SHIELD OS skin copies the GUI from the successful Ironman movie franchise and while it's quite full on, if you're a fan of Ironman it's something you'd like. There's other "suites" of skins like the Jarvis one for Star Wars (for example).

I've messed around with Rainmeter for a good four years now, and I've shifted my design mentality towards a more minimalistic style than I had before. I've got a ton of skins and I'll be sharing them to Github if you'd like to check them out; however there's only about 4 or 5 skin sets (more commonly referred to as suites) that I have in my current setup.

I'd like to get into designing and creating my own rainmeter skins one day, but for now I make do with browsing the Rainmeter category on DeviantArt and there's tons of talented individuals there. I'd also recommend checking out the official Rainmeter group on Deviantart.

Suites

Here's a list of the suites that I use (now, I don't use an entire suite, rather a collection of skins from different suites):

NXTOS

This is one of the largest and most popular suites across the entire internet. It really fits the bill as a suite as after installing it a settings pane opens up and it sort of feels as if you're installing a software update on your computer. Some of the best skins it includes are:

Steam game drawer
System monitors
Circular dock
Circular clock

Right now, I only use the Steam game drawer - which is really elegantly designed, by the way - and it sits in the lower-right hand corner of my setup.

Gemini

One of the really cool things about this is it comes with two different modes: "Pollux" and "Castor", with the former being (by default) for games and the latter being for work. When I first read about Gemini I heard that it included an automatic linking to your steam account (like NXT-OS does), however I had to edit the .ini files instead as I couldn't figure out a way to add my games and apps automatically. I'll probably need to edit it again in the future as I didn't realise there was a quick toggle option to switch between Pollux & Castor, and I ended up with 4 games and 4 work apps on Pollux; I'll be fixing this up and sharing it with you in an edited version of this article.

It also comes with a simple, minimalistic clock. I only discovered it today and it's one of my favourites already.

Here's the (so far!) finished product:

Taking a new design approach and welcoming two new members

Arbuckle — Wed, 16 Dec 2020 03:34:00 +0000

Over the last two weeks, I've been meeting and interviewing several Perth-based programmers and designers looking to join the team at Signal Kinetics. I'm excited to be announcing two new teammembers, but first a lot has been going on at Signal kinetics so Iguess we'd better talk about that.

Like we've been planning to do for a while we intend to make our venture not only fun and innovating but also profitable. Immediately profitable. Due to this we've been expanding so that we can contine participating in our hackathous and creating our games as well as going into the commercial side of things. This includes freelancing and setting up an indie game development branch at Signal Kinetics. We've started the process of moving into a new studio and are still continuing with all of our old open-source projects, but now we have the capacity to make money as well, helping us support ourselves.

I've started learning different design techniques to be able to go from idea to fleshed-out concept much quicker than usual. usually I imagine in my head what I want to create and I just start coding. Oh, of course I look around for site inspiration on platforms like Envato, but I never use things like Adobe XD. At least, not until now. I've started the process of redesigning Our team's dashboard to have less reliance on external tools and Adobe XD is a major part of that design process.

In other news at Signal Kinetics, we've started preparations on a kickstarter to fund our opensource game engine. We've also opened up contact with satellite company Ambasat and the looks to be a fruitful venture.

Finally, I'm excited to announce our team's latest, and most promising recruit: Arthur Passos.

@everyone, please welcome our newest members:
@arthur Passos, our CIO (Chief Informations Officer) working with us (locally, Perth) on web apps, freelancing and a little bit of gaming. @elijah Gardi, our new CMO (Chief Media/Marketing Officer) working with us (locally, Perth) on game desgin and content creation. Eli's been working with us for a few months, while Arthur comes highly recommended from the beautiful facebook jobs database.

Arthur's joined as us our new Chief Informations Officer, where he'll be working alongside myself and the rest of the development team including Eli, Rishabh, and Edwin. Our first step is to move into our new studio and start attracting more clients.

stay safe, and we'll have more updates coming soon

--L

Game engine & future innovators award

Arbuckle — Fri, 27 Nov 2020 01:37:25 +0000

It's an honour to have been nominated for, and to have won, the Australian Defence Force Future Innovators Award for 2020. I wanted to write a little about how I won this award and who I have to thank. This year, I've collaborated with and worked on projects with companies/groups like NASA, Microsoft, & IBM and I've become a part of a team that is going places. I'm really excited for the future, both for myself and for Signal Kinetics.

Last year, I joined forces with Rishabh Chakrabarty and Edwin Montgomery to compete in the NASA/JPL Open Source Rover competition for 2019/20 (a competition that will likely be repeating and a vibrant community that I'm looking to continue working alongside and with in the future) and we did really well. It was my first coding competition (hackathon) and my first project to use machine learning (specifically reinforcement learning). I learnt a lot and this not only started our team, it also led me to pursue a slightly different career path than the one I had outlined for myself.

Since then, I've accelerated my knowledge of Unity (C#), Python (ml/web dev), and JS (web dev) to be able to continue working with my team on projects that make a difference in the world and are fun to work on. I started working on introducing myself to companies who made software or games that I loved, so I'd be able to get secret tips and ideas from the best in the business - as well as a chance to put my foot in the door of the industry.

Towards the start of the year I recommitted to competing in the following school-run competitions:

Australian Space Design Competition
Microsoft AI For Good Challenge
Fremantle Young Writers Conference
Sadly due to the intensity of Year 12 I was unable to compete in many other competitions that I had done over the years previously, however these three - especially the Microsoft AI For Good Challenge - would be crucial in me receiving my award. I led a team to design a self driving car network based on reinforcement learning principles and wrote a paper on it; the main ideas discussed were reducing emissions, traffic jams and destruction to native environments through the construction of increasingly large road networks. As part of the competition we designed a raspberry-pi based model & simulation set showing how such a system could work. For this I was nominated by my supervisor, Isaac Kigodi, for the Australian Defence Force award and later on he discussed with me my other accomplishments throughout the year.

My entire list of projects can be seen on my portfolio, github profile and dev.to profile. Each and every one of these has helped me achieve my goal of expanding AC0/RD Software into a viable company and that is therefore something I'm immensely greatful for.

I'd like to take the time to thank the following people for helping me in 2020 to get to this point:

Microsoft AI For Good Challenge -

Ethan Yong
Olsen Chen
Isaac Kigodi
HeroX:

Rishabh Chakrabarty
Ankit Grover
Edwin Montgomery
Mikhail Asavkin
Ambasat
Gaming projects:

Dylan Vekaria
Martin Evans
Dervish Musovic
Savy Soda
One Lonely Outpost

The transition from group to collective
I always smarted at the idea or notion that AC0/RD was a company or corporation. The whole point was that it was an open-source initiative; while we would try and make money from our projects this was less about profit and more about making projects that made a difference in the world, and helped us achieve our goals as well. We're big believers in the idea of global access and due to this we never really described ourselves as a company.

Over the last few months we've been attempting to find our identity to create a viable future. When I joined forces with Rishabh (this meeting really allowed AC0/RD to spring from being a primarily 1-person team to where we are today) he already had his own initiative, a UROP/X entity that had a vastly different path to what AC0/RD did at that point. Furthermore, he was planning to switch into indie game development as a main career path. To try and manage all these passions between our small team has been awkward, as we want to ensure that we can embark on the professional career we wish to while still being able to collaborate as a team.

Until the latter parts of 2019, AC0/RD was primarily about robotics. I mainly did most of the work on my own, occasionally getting friends to become part-time members on certain phases. Since then, we've moved away from hardware into software, primarily in the scientific industry. And this journey has allowed me to realise what I want AC0/RD to be.

I play a lot of Eve Online and other sci-fi games, and one of the great things about EO is that it allows players to contribute to real-world scientific research by playing the game by partnering with the Zooniverse. I feel that at AC0/RD, creating to, or adding to a game engine that enables this "scientific communication" between games is one of the greatest things we can do:

Multiple projects can be constructed in these applets, whether it's piecing together the parts of a Mars rover in a construction game or programming it to drive along the plains of Mars and find the fastest route to a destination by utilising machine-learning principles. There's plenty more ideas we have for these applets
It provides a source for education that is built into the game
It provides a larger source of traffic to the Zooniverse, allowing it to grow
We'd even be able to utilise this engine for many of the competitions we take part in, which is something we want to continue
It could be created in the form of mods like traditional minecraft mods or overlays like in Eve Online
Our open source license for this means that the user is fully informed of where their data is being sent and what data is being sent; and what will be done with it.
At AC0/RD, we've been fortunate enough to work on projects relating to web development, however we're putting that on hold - and shifting it to the new company I'm sure that you've heard rumours about - and focusing on dissolving AC0/RD. What do I mean by this?

Having too many companies is bad for vision
Due to the small size of our team, we can't split ourselves into multiple companies to keep different projects separate. Obviously some projects are very different, but as we've decided as a collective to prioritize moving into indie game development as a long term goal, we needed to restructure our team.

Here's a quick run-down on all of our open-source projects we've worked on as a team, throughout 2020:

Arcadia - an open-source games platform & community; also including our games and collaborations with other publishers/developers
Our as of yet unnamed game engine
Our web starter kit, with plenty of plugins to extend websites regardless of platform or framework
Machine learning plugins for voice assistants (Voicebox) and communication between applications (APIs), for example Github's very own bot service Probot
Exoplanet machine learning research
All of these have obvious applications to our sci-game engine, and we're obviously still taking part in competitions, primarily:

NASA/JPL Open Source Rover Challenge
HeroX Exploring Hell Challenge
All the competitions we've taken part in in 2019/20 have allowed us to further our goals in the software development industry, however....

At AC0/RD we're effectively ONLY focusing on the parts of these projects that have real use for our game engine. We've formed a new company, Signal Kinetics, which will be taking over the other four projects overall. Once we've got a successful prototype of our game engine, AC0/RD will be completely dissolved into SK and we'll work on integrating this engine into the rest of our ecosystem, focusing on ways to make sure that it works well with everything we do. We'll prioritise competing in competitions that allow us to further these projects with the aim of creating a more connected environment for discovery as a collective.

Finally, we're also forming a freelancing team to create websites and other mediums for users. Tentatively named "A+", it's planned to be the main money making service for us; while everything we make is open source, not everyone has the time to make a website for themselves. That's where we come in, selling a service rather than a product.

Thanks for taking the time to read this post, and I'll come back with a little bit more about what's been going on with my end

opensource #innovation #microsoft #ai #team #future #cloud #blockchain #cybersecurity #projects #aws #digitaltransformation #azure #community

My "Terminal" Setup

Arbuckle — Tue, 20 Oct 2020 10:39:08 +0000

As someone that customizes my devices frequently, it's surprising that I've never really touched the terminal, or as it's known on Windows - the command prompt. It's probably because I've never used it very much - it's only when I started learning Django that I used the command prompt on my computers for programming.

In 2016, I discovered the joy of customization when I started installing launchers on my Samsung Core Prime smartphone. From Nova to Microsoft Launcher, I played around with icon packs, the app drawer and the notification center, and I'll be doing a post describing some unique setups for Android devices some time in the future. In 2017, I started using rainmeter, which I still use today.

When I started programming, the main applications I used were Visual Studio Code, Unity, Github/Gitkraken and a web browser. It may seem odd, but until recently my main projects were bootstrap/react-powered websites and a series of interactive games (again, I'll be writing about my journey through programming soon). I never touched the terminal.

I'm now part of a proper team - for a long time, AC0/RD Software was primarily me - and just me - 99% of the time. When I added more people to my network, I started getting involved in python competitions and I learnt a lot about Raspberry Pis, robotics and programming in general. Right now, we're working on API services and I thought getting started with Django would be a good idea. I actually started with Django in April, when I was building the foundations for Arcadia, but after finishing the demo project I hadn't touched it for a while. I'd used the command prompt for things like pip and setting up virtual environments. When I determined that I'd need to use the terminal, or a bash application, frequently, I quickly came to the decision that I needed to beautify my setup.

I have 4 weeks left of school (including exams), and after Year 12 finishes I'll be getting a new laptop. I don't want to have to go through the aches of re-learning Linux (I haven't used it since May, and the dualbooting process took several years off my life) or getting a Macbook while I'm preparing for my ATAR/WACE exams, so I decided to try and find terminal applications that work on all platforms - that way I'll be able to bring my setup along with me after school.

For now, I'm using HYPERJS. I discovered starship.rs and I fell in love with it, but it was a pain to install - well, I still haven't installed it (ha ha). Hyper was recommended to me by a friend and it looks just as good. If you haven't heard of Hyper, it's a terminal replacement that works on Mac OS, Windows & Linux. It supports a wide variety of open-source plugins and themes, which can be managed in the .hyper.js file. Here's mine for reference:

// Future versions of Hyper may add additional config options,
// which will not automatically be merged into this file.
// See https://hyper.is#cfg for all currently supported options.

module.exports = {
  config: {
    // choose either `'stable'` for receiving highly polished,
    // or `'canary'` for less polished but more frequent updates
    updateChannel: 'stable',

    // default font size in pixels for all tabs
    fontSize: 12,

    // font family with optional fallbacks
    fontFamily: 'Menlo, "DejaVu Sans Mono", Consolas, "Lucida Console", monospace',

    // default font weight: 'normal' or 'bold'
    fontWeight: 'normal',

    // font weight for bold characters: 'normal' or 'bold'
    fontWeightBold: 'bold',

    // line height as a relative unit
    lineHeight: 1,

    // letter spacing as a relative unit
    letterSpacing: 0,

    // terminal cursor background color and opacity (hex, rgb, hsl, hsv, hwb or cmyk)
    cursorColor: 'rgba(248,28,229,0.8)',

    // terminal text color under BLOCK cursor
    cursorAccentColor: '#000',

    // `'BEAM'` for |, `'UNDERLINE'` for _, `'BLOCK'` for █
    cursorShape: 'BLOCK',

    // set to `true` (without backticks and without quotes) for blinking cursor
    cursorBlink: false,

    // color of the text
    foregroundColor: '#fff',

    // terminal background color
    // opacity is only supported on macOS
    backgroundColor: '#000',

    // terminal selection color
    selectionColor: 'rgba(248,28,229,0.3)',

    // border color (window, tabs)
    borderColor: '#333',

    // custom CSS to embed in the main window
    css: '',

    // custom CSS to embed in the terminal window
    termCSS: '',

    // if you're using a Linux setup which show native menus, set to false
    // default: `true` on Linux, `true` on Windows, ignored on macOS
    showHamburgerMenu: '',

    // set to `false` (without backticks and without quotes) if you want to hide the minimize, maximize and close buttons
    // additionally, set to `'left'` if you want them on the left, like in Ubuntu
    // default: `true` (without backticks and without quotes) on Windows and Linux, ignored on macOS
    showWindowControls: '',

    // custom padding (CSS format, i.e.: `top right bottom left`)
    padding: '12px 14px',

    // the full list. if you're going to provide the full color palette,
    // including the 6 x 6 color cubes and the grayscale map, just provide
    // an array here instead of a color map object
    colors: {
      black: '#000000',
      red: '#C51E14',
      green: '#1DC121',
      yellow: '#C7C329',
      blue: '#0A2FC4',
      magenta: '#C839C5',
      cyan: '#20C5C6',
      white: '#C7C7C7',
      lightBlack: '#686868',
      lightRed: '#FD6F6B',
      lightGreen: '#67F86F',
      lightYellow: '#FFFA72',
      lightBlue: '#6A76FB',
      lightMagenta: '#FD7CFC',
      lightCyan: '#68FDFE',
      lightWhite: '#FFFFFF',
    },



    // the shell to run when spawning a new session (i.e. /usr/local/bin/fish)
    // if left empty, your system's login shell will be used by default

    // Windows
    // - Make sure to use a full path if the binary name doesn't work
    // - Remove `--login` in shellArgs
    //
    // Bash on Windows
    // - Example: `C:\\Windows\\System32\\bash.exe`
    //
    // PowerShell on Windows
    // - Example: `C:\\WINDOWS\\System32\\WindowsPowerShell\\v1.0\\powershell.exe`
    shells: [
      {
        name: "CMD",
        shell: 'cmd.exe',
        args: ['--login'],
      },
      {
        name: "Bash",
        group: [
          {
            name: "WSL",
            shell: 'C:\\Windows\\System32\\bash.exe',
            args: ['--login'],
          },
          {
            name: "Git",
            shell: "C:\\Program Files\\Git\\bin\\bash.exe",
            args: ["--login"],
          },
        ],
      },
      {
        name: "Powershell",
        shell: "powershell.exe",
        default: true,
      },
      {
        name: "Cygwin",
        shell: "C:\\cygwin64\\bin\\bash.exe",
        args: ["--login"],
      },
      {
        name: "Visual Studio",
        shell: "cmd.exe",
        args: [
          "/k",
          "C:\\Program Files (x86)\\Microsoft Visual Studio 14.0\\VC\\vcvarsall.bat",
          "amd64",
        ],
      },
    ],



    // for environment variables
    env:{ TERM: 'cygwin' },



    // set to `false` for no bell
    bell: 'SOUND',

    // if `true` (without backticks and without quotes), selected text will automatically be copied to the clipboard
    copyOnSelect: false,

    // if `true` (without backticks and without quotes), hyper will be set as the default protocol client for SSH
    defaultSSHApp: true,

    // if `true` (without backticks and without quotes), on right click selected text will be copied or pasted if no
    // selection is present (`true` by default on Windows and disables the context menu feature)
    quickEdit: false,

    // choose either `'vertical'`, if you want the column mode when Option key is hold during selection (Default)
    // or `'force'`, if you want to force selection regardless of whether the terminal is in mouse events mode
    // (inside tmux or vim with mouse mode enabled for example).
    macOptionSelectionMode: 'vertical',

    // URL to custom bell
    // bellSoundURL: 'http://example.com/bell.mp3',

    // Whether to use the WebGL renderer. Set it to false to use canvas-based
    // rendering (slower, but supports transparent backgrounds)
    webGLRenderer: true,

    // for advanced config flags please refer to https://hyper.is/#cfg

    hyperTabs: {
      trafficButtons: true,
      }
  },

  // a list of plugins to fetch and install from npm
  // format: [@org/]project[#version]
  // examples:
  //   `hyperpower`
  //   `@company/project`
  //   `project#1.0.1`
  plugins: [
    "hyper-midnight",
    "hyper-tabs-enhanced",
    "hyper-active-tab",
    "hyperpower",
    "hyperborder",
    "hyper-letters",
    "hyper-dark-scrollbar",
  ],

  // in development, you can create a directory under
  // `~/.hyper_plugins/local/` and include it here
  // to load it and avoid it being `npm install`ed
  localPlugins: [],

  keymaps: {
    // Example
    // 'window:devtools': 'cmd+alt+o',
  },
};

And here's what it looks like:

Just amazing.

A look at the Mars Rovers

Arbuckle — Mon, 03 Aug 2020 11:19:45 +0000

At AC0/RD, we've been privileged to work on rover projects with NASA's JPL. The long-awaited Mars Perseverance Rover launches tonight, and we'll be exploring the red planet, and the rover's mission

pathfinder-concept.jpg

It’s fair to say that NASA has been at the forefront of off-world driving for a number of years now. After the Pathfinder mission in 1997, NASA’s sent 3 more rovers to the Red Planet and each of them has been extremely successful. A fifth American rover - nicknamed Perseverance - will be joining them in 9 months if - and that’s a bit if - all goes well. While one common misconception is that the primary focus, or objective of each current rover was to search for life, the Mars Perseverance Rover is the first that has the main goal to find biosignatures that signal life once existed on the surface of Mars - and may still.

The UAE launched their Mars orbiter successfully on the 19th of July, however, most government space agencies haven’t had much success in launching missions to Mars - almost one-third of probes and rovers sent to Mars have failed. Regardless of the reasons for failure, one thing is certain - getting to Mars is certainly not a cakewalk.

All the signs point towards a successful launch, however. NASA’s learnt a lot since their catalogue of failures at the dawn of the space race; this has led to the last 10 missions to Mars (launched by NASA) being successful, with this statistic dating back to 2001. 7 hours ago, NASA announced that the launch is going ahead and the rover’s Twitter profile is sending out a lot of exciting news.

At AC0/RD, we've been privileged to work on rover projects with NASA's Jet Propulsion Laboratory (JPL). This includes the custom version of the Open Source Rover, as well as adding Arduino tutorials to the official OSR Repository; however, our highlight was competing in this year’s JPL Space Challenge.

A brief history of our exploration of Mars Since mankind first began to look at the stars, we’ve always been fascinated by the “wandering stars”; today known as planets. The reddish tinge of Mars gave the planet its name - after the Roman god of war. The planet of Mars has had more missions sent to it by humans than any other planet.

NASA’s Mariner 4 was the first successful mission to Mars ever, in July 1965. It was a fly-by mission - which basically consists of a probe flying past a planet, taking a bunch of pictures and other readings, and then shooting off into space. It wasn’t until 1971, 6 years later, that a man-made object entered orbit around Mars. Again, it was a NASA/American probe - Mariner 9. It orbited roughly 9,000 kilometers above the surface of Mars, and we got our first true close-up images of the Martain surface a few months of dust storms (arriving in May 1971, a few months of dust devils meant that we wouldn’t get too many clear photos). Over 7000 photos were taken over an 18-month period. It’s expected to orbit Mars until at least 2022, however it’s currently inactive.

The Soviets were the first nation to send probes to Mars, however none of them were successful (in fact, the Mariner 9 probe was also the first man-made probe to orbit another planet).

In 1976 the two Viking landers landed on the surface of Mars, along with two orbiters, with one of the goals being to search for evidence of life on the Red Planet. While unable to conclusively prove that life had ever existed, they sent back the first colour panoramas of Mars.

The Phobos program by the Soviets had the main objective to explore the two moons of Mars (likely captured asteroids) - Phobos & Deimos. Phobos 1 lost contact with Earth en route to Phobos, while Phobos 2 was marginally more successful, taking photos of the moons but failing just before it was about to launch a lander in 1988.

NASA’s Pathfinder lander arrived at Mars in 1997. Currently called the Carl Sagan Memorial Station, it was the first mission to involve an automated rover off-world. It paved the way for future Mars missions.