For the main explorer challenge OZ Minerals released over 4 Tb of private exploration data, but understanding this avalanche of information is a challenge without prior experience. When we reached out to our community of thousands of innovators globally we realised very quickly that we needed to provide an easier way for non-geoscience innovators to get on board.

To help teams get started with exploration data, we created a ‘data science stream’ in parallel with the main Explorer challenge, and asked innovators to use publicly available exploration data to predict mine locations from across Australia. For more details on the challenge, scoring and dataset, see the full challenge description.

As a base dataset we provided data from 25 x 25 km ‘stamp’ areas selected from across the continent. Each stamp dataset contained known deposit locations and aligned geophysical and remote sensing data from publicly available geophysical exploration data across the Australian continent (over 47,000 raster and vector layers total). These were taken from national datasets managed by Geoscience Australia and the national research data collection hosted at NCI.

We also provided a publicly available Python package for teams to generate their own coverages, and support for innovators through our community platforms, where non-geoscientists could get access to geoscience or data science expertise to help them compete.

Although we provided a benchmarking/leaderboard facility for teams to test their predictions against other teams’ approaches, our aim wasn’t to create a pure machine learning competition, but rather to provide feedback on the teams total data science approach. We judged submissions not just on technical accuracy, but also on how teams broke down the problem, how they handled data quality and uncertainty and how well they communicated outcomes to non-data science audiences. All of these challenges are familiar to anyone who has had to convince an exploration manager to drill a hole into a given target.

We don’t want to give away too much information about the technical details that teams used since that might give away their advantage in the main Explorer Challenge. But we thought it worthwhile to consider the approaches in aggregate to see if there were any interesting commonalities in approach or direction.

This wasn’t a straightforward challenge to complete. There were a few known problems that we were interested in seeing teams tackle. For example, there was a massive imbalance in numbers of each deposit class. This was due to the different ways that deposits get recorded by Geoscience Australia (our source for deposit locations). For example, many gold lodes are recorded as separate deposits, even when these can be closely related. In contrast larger deposits such as Broken Hill or Prominent Hill are recorded as a single deposit. Teams needed to handle this issue both in defining a good metric for their models, and also when stratifying samples for cross-validation. Most teams went with the resource groupings provided by Unearthed but there were some interesting unsupervised approaches to label grouping as well.

Unsurprisingly, teams that did well on the benchmark also produced submissions which showed they had spent a lot of time time exploring the datasets to deal with the garbage-in, garbage-out problem. Although we provided a relatively ‘clean’ set of data teams still had to content with missing values and skewed distributions, not to mention the problem of extracting useful features.

There were a range of approaches taken to data/feature selection, from relatively manual criteria based on geological expertise, through to fully automated ML feature selection. In particular, there were a number of interesting ways to encode the vector data before training a model, including some interesting dimensionality reduction techniques. We didn’t see as much feature engineering on the raster layers themselves, possibly because a lot of gradient features fall naturally out of the neural net approach, but also possibly because of time limitations (some teams reported using hundreds of hours of CPU time just to preprocess the data).

For actual ML models, teams generally took one of two approaches — either deep nets using the stamp layers as channels, or a pixel- or superpixel-based feature extraction/boosted tree approach (CatBoost was particularly prominent — thanks Kagglers). The best accuracies in this competition came from the neural net approach although the judging panel thought that being able to mix-and-match some of the more interesting feature engineering approaches used by the point-based models would improve this further.

It’s interesting to look at the scoring of submissions to the benchmark in total and see what this says. We provided a metric that had two components: (a) a commodity label metric which measures how well a model predicts the likely deposit type(s) in a given stamp, and (b) a location metric, which measured how close the predicted deposit locations were to the actual deposit locations. Both components and the combined metric had a range of [0, 1] with higher values being better. These metrics were designed to penalize both false positives (saying there is a deposit when there isn’t one) and false negatives (saying there’s no deposit when there is one), with false negatives being penalized much more heavily.

We expected that part (a) would be relatively easy to achieve since there are lots of geological signals (in lithology etc) that tell you whether a particular region is prospective for a particular commodity. Part (b) was much harder since the location information is very sparse (same number of deposits but now spread over 250000 pixels per stamp). Roughly these parts correspond to the way we approach exploration in industry — it’s pretty easy to get yourself into a prospective area, but really challenging to decide exactly where to put the drill rig.

For the combined benchmark score we weighted 75% on commodity label and 25% on location, however in the plots below we show each of these metrics on separate axes. We also provided three ‘benchmark’ situations to see how far random guessing gets you:

1. Optimist: The optimist says there is a single gold deposit at the centre of every stamp in the dataset. The overall score for the optimist was 0.34, with a distance score of 0.56 but a commodity label score of 0.27. A lot of the submissions were worse than random guessing based on these metrics, which gives you a clue as to the difficulty of the task.
2. Pessimist: The pessimist says there are no deposits in any of the stamps. All scores are 0, since it fails to predict any deposits at all and hence gets penalised on all stamps which actually do have deposits.
3. Super-optimist: The super-optimist places four deposits (Au, Pb-Zn, PGE and Fe) about the centre of the stamps. You can see that the super optimist does worse than the optimist on both labelling and distance metrics, since it is being penalized for more false positives than the optimist in the commodity label, but also because the distances to actual deposits are likely to be higher given there are four deposits rather than one.

The best scoring teams’ submissions fall in the top right-hand corner of this plot. The best overall marks were just over 0.5 but a lot of submissions fell in the 0.4 to 0.5 range. From the overall scatter you can see that while most teams were able to effectively learn a commodity classification, a lot of teams struggled to get the location information much above random guesses (the optimist), and a few submissions noted that their models basically learnt to place deposits at the centre of the stamp. The teams that did do better on location used some interesting priors to force their models away from the centre of the stamp, and it will be interesting to see whether this improves in the main competition.

So if you’re a data scientist looking at this and thinking about the approach for the main Explorer Challenge, what should be your takeaway?

• Sampling bias and covariate shift are not your friends. There is a strong bias towards areas with low cover (cover == regolith and other stuff that hides deposits) in Australia — simply because it’s easier for a geologist to fall over something if it’s sticking out at the surface. In contrast, the area under Mt Woods has thicker cover sequences, requiring you to use geophysical features rather than surface information like Aster. Most teams found the geophysics more useful than the remote sensing data at a national scale anyway but this is something to keep in mind.
• Higher resolution data will make a big difference. The datasets used in this competition cover the Mt Woods tenement but you’re likely to get better results using the detailed regional geophysics (requires competition signup) in the data release for the main challenge. As most of the tenement is under cover, the geophysical potential field datasets (magnetics, gravity, IP) are likely to be more use than some of the remote sensing datasets used in the data science stream.
• Find some geological expertise. Teams which scored well in this competition had found a geologist to bounce ideas off. If you’re looking for some extra help, come and ask questions on our community slack — we’ll try and point you in the right direction.

Explorer Data Science Stream Review was originally published in Unearthed Community on Medium, where people are continuing the conversation by highlighting and responding to this story.

Written by Holly Bridgwater and Jesse Robertson, compiled by Yhana Lucas for the Explorer Challenge

Over the life of a project, exploration geologists collect a vast array of different datasets, all aimed at providing them with different information on rock characteristics. They use this information to assess an area of land for the potential to host economic quantities of the mineral or element they are looking for, like copper or gold. Economic mineral deposits are rare, and over time are becoming harder to find (all the easy ones have been found), so geologists are continuously looking for new data sources and techniques to give them an additional edge.

These datasets can feel a bit intimidating at times, as it may appear that you need an unattainable amount of specialist knowledge to interpret them. Here, we give an overview for those new to geology data — of some commonly collected datasets, how they are obtained, how geologists typically use them and what to watch out for.

The Exploration Lifecycle

Before we jump into the datasets, it’s useful to get a snapshot view of how exploration geologists approach looking for deposits. A really great infographic on mineral exploration is provided here by Visual Capitalist and Orix Geoscience. It’s a great place to get started to get a feel for the overall process.

Planet Earth itself

If you’re totally totally new, here’s an exciting (we think) video on how earth was formed.

Okay, so onto the data.

Geological Reports

In Australia, exploration and mining companies have been required by the government to submit annual reports of their activities back into the 1970’s. In the last 15 years or so, they have also been required to submit all their datasets digitally, so you can imagine there is a wealth of knowledge in these historical exploration reports! The government have done their best to digitise most of this data, but geologists will always review the historical reports to look for extra clues that might have been missed!

The reports that companies have submitted, will relate to the land which they have been exploring over. Companies are granted a permit to explore over a specific area of land, which is called a tenement. Each report is associated with a tenement, or multiple tenement numbers.

Geophysics

As the name implies, geophysics is the study of physical properties of the earth. Ore bodies often (but frustratingly not always!) have distinctly different physical characteristics to the rocks around them. They may be more dense, more magnetic, more conductive or resistive. Geologists use a suite of different airborne and ground surveys to measure these characteristics both at surface and at depth. The needle in a haystack analogy is pretty applicable in this case, considering the very different physical properties of the needle and the hay.

Common geophysical surveys include; seismic, magnetics, radiometrics, gravity, IP (induced polarisation), resistivity, EM (electromagnetic induction) and magnetotellurics.

Geophysical surveys are carried out on a range of different scales. You’ll find merged Australia wide and state surveys at Geoscience Australia and the SARIG portal, and small scale surveys carried out over project areas.

When a geophysical survey is conducted, we are typically putting energy into the ground and recording the response. That response is caused by the distribution of physical properties in the rocks. What we really want to know though, is the physical properties that gave rise to the response, not the response itself. The process to get to that is called inversion, which is a mathematical process applied to the raw data based on prior knowledge about the region. Several methods effectively measure the same physical property but at different scales to trade off between the penetration and resolution of the method (generally speaking, longer wavelengths penetrate deeper in the earth but at lower resolution). In the various datasets you will notice you have access to a combination of the raw and inverted results, and gridded images of those inversions.

Remote Sensing

Remote Sensing as a general term refers to the acquisition of data without physically touching the object you are acquiring data from. In geology, this is largely satellite data, or data captured via airborne surveys. Although remote sensing would also include some of the geophysical surveys mentioned above, geologists are usually referring to a couple of types of data: Landsat (satellite imagery) and spectral data like ASTER that give us information about the earth’s surface, usually at a 10–50 metre resolution (for currently-available public data). Geologists frequently use spectral data to indicate what minerals are present on the surface and subsurface. Different minerals have distinct emission and reflection spectra. This type of data is most regularly used to identify alteration minerals around deposits, which may extend many 100’s of metres away from the deposit itself, but can also be used to provide predictive covariates for surface geochemical samples.

Surface Geochemistry and Geological Mapping

From the early phases of exploration geologists will often take field surface samples, from rocks or soils. Surface geochemistry results will indicate if anomalous levels of an element are present, be that the ore element itself or proxies/vectors. This is a fairly cheap and quick way to assess an area of land. Geological mapping often occurs in conjunction with surface geochemistry surveys. Geologists record their observations in the field to create geological maps which can be used to identify prospective features for mineralisation locations.

At many sites, the majority of the tenement is undercover, by this we mean that the older, prospective rock units are covered by more recent sediments. There is therefore little surface geochemistry or field mapping for these projects.

Drillhole Database

This is where you get to dig a hole!

Geologists often referring to drilling as the truth machine. It is the one surefire way to demonstrate if mineralisation is present under the surface, through collecting subsurface samples. Drilling is expensive, so geologists will only carry this out once they have used some of the techniques above to identify prospective areas.

Drillhole databases (DHDB) include all the information that was recorded about the samples collected and the drilling process including; chemical assays, geology logs, surveys and often much more.

For chemical assays, when drillhole samples are submitted to a laboratory for analysis, the geologist will select what elements they want to know the concentrations of and what analysis method they would like used. Different analysis methods have different limits of detection (the lowest concentration detectable) for different elements. This has also changed through time, so historic results will usually have higher detection limits than more modern results, where the detection limits have improved with new technologies and methods.

Assay results are almost never normalised for a project, so always look out for different analytical method codes and detection limits when you are comparing results. The is particularly obvious in gold exploration, where different methods may be used for surface sampling compared to drilling samples.

It’s also worth noting that drillholes rarely go as smoothly and straight as we’d like — you can read more about how coordinates are computed to allow for this here.

Want to have a crack at picking where to dig? Join the Explorer Challenge, main or Data Science stream.

What else?

We’ve just gone through the major dataset types — but this is absolutely not an exhaustive list.

To generate new insights, we encourage the use of other datasets that are commonly not used by geologists in their exploration programs.

Here’s a quick hitlist of public datasets to get started with:

• The South Australian Resources Information Gateway (SARIG)
• National Computational Infrastructure (NCI) Geoscientific Datasets
• NCI Remote Sensing
• Geoscience Australia data and publications portal
• Commonwealth Scientific and Industrial Research Organisation (CSIRO) general data
• CSIRO Hydrogeochemistry data
• Other CSIRO Geochemistry data
• NationalMap

So you want to dig a hole: How geologists use data to make decisions was originally published in Unearthed Community on Medium, where people are continuing the conversation by highlighting and responding to this story.

Multi-disciplinary approaches have been shown time and time again to produce the best solutions for difficult problems, in a variety of instances.

We’re encouraging Explorers to test this out for minerals exploration!

In medicine, multi-disciplinary teams have been shown to improve outcomes for patients with chronic or complex health challenges.

In reality television, the challenges are always won quickest when the Beauties and the Geeks combine their knowledge — and the Survivors don’t stab each other in the back.

In nature, egrets and cattle team up; the birds get a free ride and meals of tasty insects, the cattle get picked clean of fleas and ticks and have an early warning system for approaching predators.

This is relevant!

See, the Explorer Challenge combines two disciplines each with advanced technical knowledge; data science, and geoscience. This is the kind of project where multi-disciplinary teams have a clear opportunity to excel; the breadth and depth of knowledge doesn’t sit neatly in any domain, and it would take some kind of wizard to know everything that’s going to be relevant.

Though we have poked the geoscience bear somewhat with that Sheldon Cooper meme, amongst other allusions, we 100% rate the importance of geoscience for the challenge. Having sound geological assumptions will be crucial for model development.

Likewise, we think geologists will benefit from the skills of data scientists: yes, we know you use modelling software and can handle the maths, but if geologists could solve this problem alone, we wouldn’t need to hold this challenge at all.

There are more benefits though. Not only do musti-disciplinary teams give you a much greater breadth of technical knowledge, and a multitude of ways that said knowledge can be combined, but they also offer different perspectives and thinking and working styles.

Whilst different thinking and working styles also exist within single domains, there tend to be dominant styles of thinking and ways approaching problems found in every field. It can be argued that this is a natural, justified state — that different personalities just suit different careers better; but also, it’s a state that’s compounded by both conscious and unconscious bias, hiring and promoting similar kinds of people as past staff and leaders.

Regardless of the cause though, combining fields helps gather a diversity in views, which can generate previously untested ideas.

Our resident geologist-turned-data-scientist described the two dominant thinking styles as “splitting” and “mashing”, while speaking at the Perth Machine Learning Group meetup.

“Speaking as someone who has sat on both sides of the fence… when it comes down to it, data scientists and geologists think in fundamentally different ways,” he said.

“Machine learning is really good at doing splitting — categorising things, pulling out trends… Geologists are really good at taking information and mashing it back together… the reason why mashing is important is that it’s not always clear from the data alone what it is we’re actually looking for.”

“We think that the winning teams are going to be the ones that take a combination of approach.”

So, how can you form a team?

If you’re ready to try this multi-disciplinary thing, but need to make some new connections to do so, head on over to the #explorer channel on Slack (once you create an account on our platform, you can join our Slack workspace).

You can post a detailed description of yourself and your ideal teammates, worthy of eHarmony, or just stick up a simple “seeking geos, apply within.”

Slack has private chat functionality, so you can take conversations off the public channel once you find some potential collaborators.

Any tips for teamwork?

We’d love to hear any tips for working successfully in multi-disciplinary technical teams! Comment below or shoot us an email if you’d like to share!

By Your Powers Combined: the potential of multi-disciplinary teamwork was originally published in Unearthed Community on Medium, where people are continuing the conversation by highlighting and responding to this story.

1. E.H. Gombrich, ‘The Story of Art’

E.H. Gombrich’s defining work ‘The Story of Art’ is the ultimate compendium for all designers. His magnificent work covers the world history of art (and its associated theories), from 30,000BC to the Modernist era in the 20th century in a form that is as relevant today as it was upon its publication in 1950. Of course, it is reflective of the time and there is a distinct lack of female, LGBTQ, and minority artists; but it is crucial to read first in order to critique the conventional canon of the History of Art. As such, it remains the bedrock source of knowledge for everyone starting a serious journey into the arts. Completing the tome in its entirety is the equivalent of a BA in Art History, though it’s also easy to dip into specific chapters as needed.

2. Stanley Kubrick

A filmmaker who needs no introduction. His understanding of colour and symmetry, paired with the 1960s-70s complex relationship with the future makes for an aesthetic which all perfectionists from all eras can relate to. When thinking about product design for new technologies, it’s so important to realise that each generations’ philosophy about the distant future is heavily impacted by both the politics and prevalent artists of the time. To us, the electric red bubble chairs aboard the space ships in 2001: A Space Odyssey seem dated, even kitsch. However, the internal designs of the ships are still lightyears away from what the International Space Station could be. I often reflect upon how the politics of my time affect my work, and which set designs in Kubrick’s films still seem to be cutting edge half a century later.

3. Frank Lloyd Wright — Falling Water

The entire body of Frank Lloyd Wright’s architectural designs and how they both interacted with, and defined Modernist design theory, is worth an article on its own. However, the most appropriate of his buildings which impacts technology product design is Falling Water (1936–38). Up until this point, buildings had to fight against nature in order to function. Constructing a building, whether public, government, corporate, or private, was a battle to thrust traditional forms onto an unforgiving plot of land. This work completely re-evaluated the purpose of architecture. Rather than trying to fight against the landscape, he proposed integrating the natural defining features of a plot of land into the building itself. Coupled with new engineering breakthroughs in reinforced concrete and cantilevering, the final result is an ode to ‘leaning into it’. Specifically for wearables design, it’s important not to fight against the limitations and forms of the human body; and in terms of implementing a solution it’s important to use the geo-social context of the problem as an anchor for design, rather than something to fix.

5. Guillermo Del Toro’s Creatures

Guillermo Del Toro is the encapsulation of not ‘over teching’ something. His fantastical creatures are by the far the most believable to come out of the history of cinema. The reason why is that he has a deep understanding of the way that CGI works and its limitations. Up until recently, every CGI frame that you’ve seen has been ‘hand drawn’ by an artist (over the last 5 years, there has been an introduction of machine learning (ML) techniques into the sector, but that has its own set of problems). Yes, that means computer renders, but those renders still have to be created by someone with a degree in sculpture, and animated frame-by-frame. The closer you manage to get a shot to where you want the final image to be, the better the end result is going to be because CGI artists produce much higher quality work if you can give them something to ‘trace’ over, rather than leaving them to guess the intricacies of your inner visions. Therefore, Del Toro creates as much of his creatures as possible in person, utilising a mixture of traditional and cutting edge animatronics techniques, only dusting the top with CGI for finishing touches.

I regularly use this methodology when approaching new problems that cutting edge technology can’t fix. Recently I was awarded Komatsu’s title of one of the ‘Top 4 Tech Innovators of 2018’ for a deliberately low-tech solution to an old problem of mine bucket tooth loss. The problem is thus: steel bucket teeth get erratically ground down on site and end up going into crushers, causing major delays on site and fatalities for people retrieving the lost teeth. Contemporary RFID tags would normally be the obvious solution to this problem, but the radio signals can’t get past the thick steel of the teeth (likewise for all digital signals). So, how could we communicate the loss of a tooth on site without using digital signals? My team and I went right the way back through the history of human communication and landed on smoke signals. We created a proof of concept for a coloured bio-degradable smoke bomb system, embedded within the tooth, which would go off as soon as a tooth became dislocated from the bucket. In addition to creating an obvious signal to the site that a tooth had fallen from the bucket, the smoke also stained the tooth a bright colour, making it easier to locate either by hand, or via an image recognition system en route to the crusher.

6. Paul Franklin

You may not have heard of Paul Franklin, but you’ll already know his most famous creations. Franklin is the founder of Double Negative: one of the most prolific visual effects companies in Hollywood (based in London) and the team behind all of Christopher Nolan’s groundbreaking aesthetics. He’s won Oscars for his work on Inception and Interstellar. Franklin teaches us two very important lessons for design: firstly, how roboticists should approach new problems, and secondly how art can be used as a medium to solve scientific problems.

Franklin and I share a love of Brutalist architecture and abstract robots. TARS from Interstellar looks nothing like how the majority of movie robots look like, and this is exactly the right approach for robotics design. When solving a new problem, the best roboticists don’t think ‘how would a human do this task?’ they think ‘what are the core elements of this task, and what is the most efficient way to achieve those elements?’. For instance, when designing a robot to pick oranges from trees, the worst approach to go for is ‘let’s make a 5 axis robot arm to do this, because that is the best way a human would do it.’ Instead, a more effective design would be a wall of rotating knives and a vacuum connected via snake arm to a delivery truck.

Art and science don’t seem like easy bedfellows in the 21st century, but Franklin has shown through his black hole simulation on Interstellar that art can be a medium to solve real-world scientific problems. Both Franklin and Nolan are known for their obsessive attention to detail, and their work on Interstellar has changed the boundaries of what filmmakers can achieve. They worked with American theoretical physicist and Nobel laureate Dr Kip Thorne to create scientifically accurate models of the extra-terrestrial phenomena that drive the narrative of Interstellar. So accurate, in fact, that they discovered new ways that light gets bent in black holes through the code they developed called DNGR (Double Negative Gravitational Renderer). This led to the academic paper, ‘Gravitational Lensing by Spinning Black Holes in Astrophysics, and in the Movie Interstellar’, which you can read via the University of Cornell’s library portal.

I was lucky enough to have Franklin as my mentor during my time as a professional artist, and he was one of the main supporters of my transition from the arts into technology. Like everyone, I can’t wait to see what he does next.

Alex Moss is CEO & Head Designer of Canaria: a NASA-endorsed technology company developing medical-grade wearables for use in astronaut training and the resources sector to prevent catastrophic events.

www.canariatechnologies.com

My Top 6 Technology Design Influences was originally published in Unearthed Community on Medium, where people are continuing the conversation by highlighting and responding to this story.

Fitbit is in trouble. Current projections for Fitbit’s 2017 revenue come it at $1.6 billion[1] , a downturn of $1 billion compared to their 2016 sales of $2.6 billion[2]. Despite having a 23% market share of the wearables space[3], Fitbit had to lay off 6% of its workforce earlier this year[4]. Competitors have arisen in the form of Xiaomi, Garmin, and Samsung, but a clear trend of smartwatches stealing this market is evident[5]. Sales look set to continue in the fitness wearables space, albeit at a much slower pace than before. So, what has gone wrong with the first generation of wearables? It’s not just the human nature of attraction to fads and the inevitable boredom that follows. The failure of the first generation of wearables may be traced to 5 things: the industry wasting the potential of their technology on shallow problems, an ineffective use of the data gathered, issues surrounding data privacy, a lack of trust, and a lack of behavioural change from their use.

1. Consumer applications are a waste of wearables’ potential

This is the main reason why the first generation of wearables has failed. On the spectrum of world issues, they’re just not dealing with substantive problems. B to B uses and healthcare are the sectors that wearables should be focusing on, where there are the largest gains to be had. To give some context, here are some non-consumer applications of wearables:

-Epilepsy management
-The tracking of vulnerable elderly dementia sufferers by concerned family members
-HAZMAT use
-Predictive fatigue control for long-distance drivers to slash roadside fatalities
-Toxic fume reporting in polluted cities
-Accident reporting for cyclists and motorcyclists, linking them to the emergency services the second a crash takes place
-Stroke detection and prevention
-Anti-sepsis monitoring for surgical patients

In contrast, what we’ve been given are wristbands which make us feel bad about our sporadic gym attendance. This is simply not good enough. The invention of wearables has created an entirely new scientific field of biometric analysis which wasn’t possible 10 years ago, so why have we been left with fitness gadgets?

The answer lies in a simple analysis of effort/benefit ratio within the field of wearables. Consumer goods are a lot easier to get to market than goods which require high levels of regulation such as CE and FDA approval, which take years of R & D to develop. This allows a business to have a high initial windfall of revenue, but an inevitable petering out of profit once the consumer trend they’re catering to has finished running its course. Aside from the sheer effort of engineering required to create a product which is capable of addressing a complex problem, other stakeholders need to be involved in order for the solution to be a success. For instance, in order for the aforementioned use case of predictive fatigue control for long-distance drivers to be implemented, there would have to be serious changes in government legislation about safety requirements on the road, as well as an overhaul of public awareness about the issue. In short, there isn’t a willingness to tackle the really tough problems out there.

2. Lack of trust

The first generation of wearables has opted for an immediate windfall of profit, but at the cost of trustworthiness. Recent comparisons between various wearables for tracking physical activity showed large variations in accuracy between different devices, with error margins of up to 25%[6] [7]. This is because they do not have FDA or CE approvals which require more rigorous development processes, nor any desire to go through the cost or effort of gaining these. Without these approvals, they cannot be used for medical or research purposes. This wouldn’t be such a problem if consumers didn’t use these wearables for self-medicating purposes[8], which the equipment is inherently unfit for.

Added to the issue of untrustworthiness is the unwillingness of consumer wearable companies to address users’ concerns about the safety of their devices in casual day to day use. There are recurring concerns about the impact of electromagnetic radiation on users[9]. Anecdotal evidence seems to support that a small number of users may be hyper-sensitive to wearables[10]. However, this in itself is not a problem for wearables as a whole. The problem comes from the confusion that arises in users when this issue is not addressed head-on by consumer wearables companies. All that would take to resolve this problem would be a few trials into the hypersensitivity of users, with the result of ‘Warning: some users may experience mild pain from the use of this product. Please test for 1 week to establish sensitivity and stop use if pain continues’ on the back of packaging. This would be no different from nut allergy warnings which have now become the norm on foodstuffs.

3. I’ve got the data- now what?

Raw data is not the same as an end-to-end solution. Providing a user with data is only a part of any decision-making process. It seems that neither company nor user knows what the bio-metric data produced should actually be used for. It’s great that you know what your average heart rate is, but what are you actually going to do with that information? It’s not being displayed within the context of solving a problem where knowing your average heart rate would be a crucial metric. It’s just plonked on your smartphone on a cool-looking user interface to provide the illusion of productivity. What this produces in reality is a facsimile of a solution which has no problem to solve.

In addition to the fundamental problem of directionless data, even if there was a well-defined issue to solve, the quality of data provided by consumer wearables is not sophisticated enough to make progress towards a solution[11]. The feedback systems being utilised are based on reductive statistics, such as average weekly heart rate and general level of activity. Until individualised, sophisticated data sets are provided to users within an intelligent sense-making system, the actual problems that users can solve with wearables are going to be next to nothing.

4. Concerns about data privacy

Given that there are almost no laws which prevent the selling of data to 3rd parties, consumers’ concerns about the privacy of their data are absolutely justified. Of course, wearables companies may say that they’re not selling data onto other companies, but there is no real way to prove this due to a lack of legal infrastructures to tackle this very real, and very serious problem. Over the last few years we have experienced a medley of scandals about data privacy breaches, from Windows 10 secretly sharing personal information with Microsoft[12], to Whatsapp sharing conversations with parent company Facebook without engaged user consent[13], and, of course, Fitbit and Jawbone (RIP) openly sharing or selling user data to 3rd parties[14] [15].

Although this is an industry-wide problem in the technology sector, the implications of wearables sharing biometric data with 3rd parties are far more catastrophic than your sexts being reviewed by the U.S.A’s Internal Revenue Service, as revealed by the 2013 Wikileaks scandal.

This is because the 3rd party sales process can cause a snowball effect. Let’s say that HealthBands Inc decides to sell your non-anonymised biometric data to BigInsurance Plt. BigInsurance Plt decides to use this data to change their insurance tariffs. Great news if you’re super fit and healthy, bad news if you have a high likelihood of acquiring a lifestyle-related disease- you’re now ineligible for health insurance. BigInsurance Plt now decides to sell this data on to MegaCorp Inc who’s interested in doing a major background check on their current cohort of sales applicants. With this new large data set, not only are they able to match some of their applicants to their original biometric data record, they also establish that a few of their applicants have a background of depression[16]. MegaCorp Inc deems these applicants too high risk to hire, and immediately rejects their job applications without giving a reason why.

Over on the other side of town, Johnny-No-Good decides to buy a bulk of data from a few different sources: insurance companies, big data analytics companies, supermarkets, and porn sites. He can now correlate all of these different data sets in order to identify you more thoroughly- going so far as to put phone number to name, and name to bank account details. He then sells this information over the Darknet to identity thieves for a neat sum of cash.

And just like that- your consumer wearable device has caused you to not afford treatment for a serious illness, be deemed unfit for employment, and become a victim to identity theft. This may seem far-fetched right now, but without proper laws in place to either regulate or prevent the selling of data to 3rd parties, this is the future that we are all facing.

This future is already starting: there are an array of health apps which base their entire business model around the unauthorised selling of their user’s data to insurance companies. Most notable recently is EpicHealth — a British med-tech company who brandishes this source of revenue with pride to investors whilst hiding behind a smokescreen of good publicity. Companies like these are directly causing a deterioration of living standards for everyone, and they will not stop unless governments step in and force them to stop.

5. Consumer wearables only track behaviour, they don’t change it

Over half of all Fitbit buyers stop using them within the year, with one third of them stopping use within the first six months[17] . A recurring theme with the first generation of consumer wearables is that the results don’t meet user expectations. The mistake here has been to confuse the tool with the skill required to use it. A recent excellent example of this is the implementation of Hitachi automated vehicles in Australian mine sites.

The expectation when Hitachi started implementing their automated vehicles in different mine sites was that there would be tangible, unified results across the sites. It makes sense to assume this; there is absolutely no difference between any of the vehicles nor their base settings, so why wouldn’t you see blanket improvements in productivity? The results were surprising to everyone involved. Every mine site reported completely different results. The slight differences in site layout or mineral being mined couldn’t have been enough to push the results to these extremes, especially with industry-standard operational standards in effect. So what was the differential factor? It was the humans. A new technology isn’t enough to solve a problem, a business culture change has to take place in order for it to work. The same applies on a personal scale.

What the first generation of consumer wearables hasn’t realised is that they are not technology companies. They are actually personal fitness businesses who have wound up selling technology gadgets in lieu of actual health improvement systems. So is it any wonder that users are disappointed?

Alex Moss is CEO & Head Designer of Canaria: a NASA-endorsed technology company developing medical-grade wearables for use in the resources sector to prevent catastrophic events.

www.canariatechnologies.com

[1] https://investorplace.com/2017/05/dont-chase-fitbit-inc-fit-stock-during-its-post-earnings-pop/#.Wcm86KiCw2w

[2] https://investor.fitbit.com/press/press-releases/press-release-details/2016/Fitbit-Reports-712M-Q415-and-186B-FY15-Revenue-Guides-to-24-to-25B-Revenue-in-FY16/default.aspx

[3] https://www.idc.com/getdoc.jsp?containerId=prUS41996116

[4] https://www.bloomberg.com/news/articles/2017-01-30/fitbit-plummets-after-report-of-job-cuts-weak-fourth-quarter

[5] https://thejournal.com/articles/2017/09/01/smartwatches-overtake-fitness-trackers.aspx

[6] Lee JM, Kim Y, Welk GJ. Validity of consumer-based physical activity monitors. Medicine and science in sports and exercise. 2014;46(9):1840–8. doi: 10.1249/MSS.0000000000000287 [PubMed]

[7] Case MA; Burwick HA; Volpp KG; Patel M. Accuracy of Smartphone Applications and Wearable Devices for Tracking Physical Activity Data. JAMA. 2015;313(6):10–11. [PubMed]

[8] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4737495/

[9] https://community.fitbit.com/t5/Surge/Fitbit-and-electromagnetic-radiation/td-p/1412524

[10] https://www.mumsnet.com/Talk/exercise/2639550-Are-fitbits-safe-for-constant-use

[11] http://jamanetwork.com/journals/jama/article-abstract/2089651

[12] http://www.independent.co.uk/life-style/gadgets-and-tech/news/windows-10-sends-personal-data-to-microsoft-even-if-users-tell-it-not-to-10453549.html

[13] http://www.techadvisor.co.uk/feature/internet/how-secure-is-whatsapp-whatsapp-security-encryption-explained-3637780/

[14] Fitbit’s Ts & Cs: https://www.fitbit.com/eu/legal/privacy this is nicely summed up by Kelsey Munro in her Sydney Morning Herald article, ‘Data collection: Wearable fitness device information tracking your life’, “Fitbit’s policy says it will sell your data with your name removed “to interested audiences” and notes that the US, where all Fitbit data is held, may not have the same privacy protections as your own country”

[15] Jawbone’s Ts & Cs: https://jawbone.com/legal/terms Jawbone will “share” de-identified data with 3rd parties

[16] They have reached this assumption, which itself may or may not be correct, by noticing patterns of inactivity in these applicants’ biometric data sets.

[17] http://jamanetwork.com/journals/jama/article-abstract/2089651

5 Reasons Why the First Generation of Wearables has Failed was originally published in Unearthed Community on Medium, where people are continuing the conversation by highlighting and responding to this story.

“Vivacious and over-the-top”, with “shock-value” and “heels that are too high” are all criticisms that Founding CEO and Head Designer of Canaria Alex Moss has taken on the chin and used to brand herself as a woman who can meld fashion, AI, art and design into a functional form.

We had the pleasure of getting to know Alex when Canaria took part in the Unearthed Accelerator program last year and we are delighted to welcome Alex as our second Community Contributor. Over the coming week she will be sharing articles, interviews, videos and other content that is relevant to her work improving the health and safety of the global industrial workforce through Canaria’s predictive vital signs monitoring system.

Alex has extensive experience in business management and design. She has been a creative director for ad campaigns for global fashion brands Philip Treacy, Le Roni and Doc Martens, and featured in Italian Vogue for her collaborative projects with photographer Kurtiss Lloyd.

Alex was founding editor-in-chief for M2 Magazine and founding CEO of avant guard art dealership and curation company, The Court.

She has a background in Design Psychology from the Courtauld Institute and created the initial concept for Canaria with Dr Robert Finean in 2016 as part of the NASA Space Apps Challenge, where they won Global Best Use of Hardware 2016. Alex has been profiled by worldwide media outlets including BBC Culture and the Evening Standard for her work on Canaria.

A skilled public speaker (winner CSIRO Ribbit ‘Best Pitch Health Informatics 2017’), she’s recently been dubbed one of the ‘Young Space Leaders of the Future’ by the 2018 Global Aerospace Summit in UAE.

To find out more about Alex Moss and what drives her, read her profile interview in Electric Lady, a platform that amplifies the strength of women in music, politics, science, sport and beyond: https://www.electriclady.world/interviews-1/2018/6/23/alex-moss

You can follow the progress of Canaria via Alex’s Instagram account: @alexsorinamoss

Melding fashion, AI, art and design into a functional form was originally published in Unearthed Community on Medium, where people are continuing the conversation by highlighting and responding to this story.

They have a clean, ethical and progressive business model targeting new territories that gives investors access to exploration projects at their inception before the stage at which they are normally offered to the public.

Unearthed Marketing Communications Lead Hannah Frankish recently caught up with Greenfields seed investor and In-House Advisor Ahmad Saleem to find out more about their exciting first exploration territory, the Greenland Frontier Project, their startup journey, his insights into innovation and entrepreneurship in the mining industry, and their equity crowdfunding campaign. Here’s what Ahmad had to say:

How did Greenfields Exploration first come about? What was your motivation to embark on the incubator approach to frontier exploration?

There are two parts to the story of how our company came about. The first relates to how Greenland came about. It all started with Jonathan Bell (our Managing Director and Founder) in 2016. It was the bottom of the commodity cycle, and Jon had been looking at ways to start his own company for a while by this point. But he was struggling to find a suitable flagship project to build the company around. It just so happened that the Greenland Government came to Perth for an open day to talk about the potential in their neck of the woods. Jon was at this presentation and really liked what he saw. It was a frontier region of the world with the right geology, but had largely been ignored by other explorers, meaning competition would be low, but potential could be very high. It is rare for a small exploration company to stumble upon a project like this. So, it really was a no-brainer opportunity to start in Greenland.

The second part of the story relates to how the business model came about. Having worked with, followed and invested in several mineral exploration companies, Jon noticed that a lot of junior exploration companies were struggling to find new deposits because of some fundamental restrictions. The main issue was around investment. Investors seemed to only want to invest in exploration companies that looked likely to provide a return in the short term. So, this was forcing companies to look more towards mature areas even though there was a chance of a big discovery somewhere else. The result is that a lot of money is being spent ‘whittling away at the same piece of wood’, with lower chances of discovery and ultimately yielding smaller, lower-quality results. This was probably causing another derivative problem: forcing the few successful explorers to become miners.

To be good at exploration requires a completely different skill set than being good at developing and running a mine.

Therefore, this would most likely not benefit these small companies in the long term. All of these company behaviours made sense within the overall industry trend of mineral exploration discovery rates dropping drastically in the last decade or so. Jon realised that there was an opportunity here: for a company built around early stage exploration that focused on how to do exploration better without worrying about becoming a miner.

This is where the incubator business model came about. In a nutshell, we aim to be the Airbnb of early stage exploration. We want to benefit from earning potential, while not becoming overburdened with asset maintenance.

We will find frontier regions of the world to explore, take on the risk of exploring these areas, develop exploration projects, and when it comes time to aggressively test these areas, we will partner with another company that is better suited to this type of exploration, retain an interest in the project and move on to finding the next project to work on.

We are not completely unique in using this business model. There are a few Canadian companies that are using it, and you could argue that Australia’s most successful explorer, Mark Creasy, operates using this model. What sets us apart is that we are marrying the incubator model with a strong intention of using the power of the crowd. We are using crowdfunding to get up and running, by passing the established and restricted funding avenues that prefer the norm rather than the innovative opportunities. Greenfields also aims to run crowdsource competitions. We are cognisant that we are a small company and do not want to be limited by what we know or can do. So, we wanted to build a business model around using the expertise that sits out there in the wider industry and community.

How did the Greenfields team meet? What were your career backgrounds before getting involved?

The main characteristic that binds us together is that we are all motivated by trying to do something different in this industry. As Founder, Jon was instrumental in putting the vision behind the company together. Lindsay Dick is a lawyer but wanted to get more involved in the startup space. And Lindsay met Jon through his brother, who is a geologist. Mat (Chairperson Mathew Longworth) previously worked with Jon and having been involved with a few mineral exploration companies before, liked the opportunity of working for a company that wanted to take the road less travelled. My involvement started as a seed investor early on, as I really wanted to back Jon with his ideas and what he was trying to do. I then became further involved by helping the Greenfields team plan and execute the field programs that need to be carried out in Greenland. In summary, we are all here for the right reasons, which makes it easy to get along.

Has launching Greenfields changed the way you think about innovation and entrepreneurship in the mining industry? Do you think startups should “dig” the resources sector?

Yes. There is no denying that it has been a challenge on many fronts. Even though we had extensive validation about our geological concepts by some of the smartest people in the game, we found that funding for new ideas and projects is incredibly hard. If we had stuck to the status quo, simply pegged a small area in one of the usual well-worn locations, and yelled a hot commodity, financing would have been much easier. Although funding a small brownfield project is less of a challenge, the trade-off is that the probability of a major discovery is much lower.

There is a quote by Adam Grant, a psychologist at Wharton, who said that “currently we do not have a lack of novel ideas. We have a lack of people willing to back the right novel ideas.” And that is what we have found; a real aversion to backing new and novel ideas.

Consequently, there does seem to be a disconnect between the risks people are willing to take and the rewards that they are asking for. No one wants to fund the behaviour that will lead to fundamental value generation in this industry. Whereas, you look outside the industry, like the risk tolerance in the ICO universe, or the entrepreneurial spirit of Silicon Valley, and you can see that there is an appetite out there for true entrepreneurial endeavours, especially with a potentially disruptive business model. So, we realised very early on that we had to embrace other industries and people as much as, or more, than our own.

Can you describe the opportunity that Greenfields Exploration is chasing and the size of the prize?

I’d get thrown in jail if I stated dollar figures — and quite rightly so. But at the end of the day, exploration is an inherently risky enterprise, so we can’t make any promises whatsoever. That said, we are a junior targeting an opportunity that would be fit for a major mining company.

This is the benefit of working in a frontier region. You have the opportunity of discovering a giant deposit, something that could be world class.

To give you an example, our Greenland project is prospective for copper deposits that you see in the Central African Copperbelt and the European Kupferschiefer belt in Poland and Germany. A lot of the world’s copper supply comes from these two areas. We have similar geology on our project, according to our technical analyses and independent reviews. The project is also located in a similar location to the world’s greatest nickel deposit. But while this potential is much more conceptual, it is something that gets our blood flowing as well.

That does not mean we will definitely find something — but the potential is there, and it is largely untested. And that is what excites us. We don’t know how many places on Earth tick these boxes, but it may be less than the number of fingers on a hand. The nice thing about exploring for the big stuff is that all sorts of smaller targets will fall out of the tree. We know there is potential for nickel, zinc, lead, tungsten gold and more, but for the moment, we are focusing on the elephants in the room: copper cobalt and nickel.

How is Greenfields Exploration a superior solution to the status quo?

Our business model is built around aiming to do one thing and do it well: be a frontier explorer. We are not interested in developing a mine.

We plan to do low-cost, easily executable exploration work that substantially lifts the knowledge bar and the understanding of an area, before joint-venturing or spinning-out the projects into different companies.

This allows us to retain exposure to a mine (and potential cash flow) and build a portfolio of minority interests. But it also helps other companies by technically de-risking projects or areas to explore. Within Greenfields, we will diversify the risk through a portfolio holding, while having dedicated teams focus on each project. This is opposed to the classic model where you have one hero project in a mature location, and the others in the company languish. This approach minimises the downside risk, while maintaining exposure to the upside. We think it is a safer, smart and more efficient way to explore.

How does Greenfields envisage handing over projects to mining companies?

Conventional joint-ventures over the entire project, or selective mineral rights. This will primarily involve in-ground expenditure, and depending on the situation, additional cash or shares.

Our primary focus is to retain a minority interest in the asset. To make sure we don’t get held hostage to a buyers’ market, we may also spin-out the assets into a separate company and allow that company to focus on a different skill set.

At the end of the day, each partner and asset will dictate the nature of the transaction, so we’re quite flexible on how things may proceed. We are open to trying different ways of getting a deal done.

How engaged do you think the mining industry is with local startups and entrepreneurs in WA? How do you think the engagement could be improved?

For true entrepreneurial pursuits, I think the engagement is low. For ideas or solutions providing incremental change, it is palatable. To be fair, most industries are resistant to new concepts or places, as it is a behaviour that is inherent in most humans.

Seeing the appearance of shared work spaces such as CORE Innovation Hub is a step in the right direction. Perhaps a way the government can help is by extending the government Early Stage Innovation Company tax incentives to early-stage explorers.

At the moment our seed investors do not have the tax offset benefits that they could get in other innovation sectors. Similarly, the state government tips in dollar for dollar into programs such as the Exploration Incentive Scheme that largely go towards supporting drilling. However, drilling is a very precise method that is appropriate for relatively mature projects. Having funds for geophysics and other sensing methods, or techniques used as reconnaissance methods as opposed to direct testing through drilling, would help a lot as well.

It is well-known that founding a startup is no easy feat. What are the major challenges that you have faced or foresee facing? How will you overcome them? Any advice for resources sector startups?

Crowdfunding! The beauty here is that if you have a solid vision of a good opportunity, you can by-pass the traditional gate-keepers. In an odd quirk of the market, what often happens is that funds sit on the sidelines to see if the retail (mum and dad) investors like the story, before moving in and mopping up the shares.

Crowdfunding helps demonstrate this appetite to the funds, something that previously was not possible in Australia.

Greenfields Exploration has recently launched an equity crowdfunding campaign. Why have you chosen to do this?

We think that the type of person that gets involved in crowdfunding has the risk tolerance for progressive, alternative companies like our own.

Conventional markets may be conservative, but ICO investors are extremely risk-tolerant. We see crowdfunding as a good middle ground.

Of course, being first helps with getting attention. The downside is that the attention is on the funding novelty, potentially to the detriment of the commercial underpinnings.

Are there any other exciting developments on the horizon for Greenfields that you can share?

We are doing so much it is hard to get it all out in one message! On the immediate horizon, we are getting ready for the work that we plan to do in Greenland in a few months. We are also talking to other startups, both within mining and outside, to see whether they can help us explore better or utilise our datasets better. But probably the most pertinent, and the reason we are here talking to Unearthed, is that we plan to run some crowd competitions in the near future. As we mentioned earlier, we do not want to be limited by what we know.

It just makes sense to get as much input from others outside of our company to help interpret our data and define exploration targets for us. It is our point of difference and a cornerstone to how we want to operate.

We also hope to develop in-house expertise on not only running crowd competitions but knowing what to do with the results. We hope this will become another competitive advantage for us.

Thank you to Ahmad Saleem for taking the time to share the Greenfields Exploration startup journey and details of their progressive business model and crowdfunding campaign.

We wish the Greenfields team every success in their future adventures giving investors access to exploration projects at their inception.

To find out more about Greenfields Exploration, visit: https://www.gexpl.com, or if you are interested in investing in the Greenfields Exploration crowd-sourced funding campaign, visit: https://billfolda.com/offer-detail/105-greenfields-exploration-limited.

For more information about upcoming Unearthed events and competitions, visit: https://portal.unearthed.solutions/.

Breaking new ground by crowd funding mining exploration was originally published in Unearthed Community on Medium, where people are continuing the conversation by highlighting and responding to this story.

Originally published by Ross Chesnick, Director of Commercial Strategy, Slingshot Aerospace

In a packed co-working space just outside of downtown Las Vegas, an audience of mining professionals, interspersed with members of the startup and tech community, came together to shape the future of an industry. During a breakout session, we all cracked a smile in response to a mining manager’s statement. “Mining has been one of the slowest industries to embrace new technology. However, there is a bit of an opportunity now, and the iron is hotter than it has been in the past.” This comment was the perfect verbiage to describe the event that hosted it, BattleBorn. As a product of the combined efforts of Barrick, Unearthed, Switch and Cisco, BattleBorn was created to stimulate discussions on the future of mining.

Through a series of hackathons, exhibitions and speaking sessions, BattleBorn created an environment where startups and new technologies had a unique opportunity to garner feedback with the decision makers of the mining community. Speaking from the perspective of a startup, this was an invaluable moment where we could not only shake the hands of industry thought leaders, but also to identify the paint points and tangible hurdles for any new mining technology. For us at Slingshot Aerospace, this interaction and information have been the impetus to further pursue this vertical.

About Slingshot Aerospace

Slingshot Aerospace, a Techstars 2017 alumni, is a signal processing analytics and AI service provider. To elaborate, we ingest and analyze aerial and satellite imagery to unlock deep intelligence within the data. This intelligence allows us to auto-detect and monitor changes to infrastructure, flooding, vegetation, and a variety of terrestrial anomalies over time. In putting these capabilities to the test, and what was one of our proudest moments as a company, Slingshot supported Team Rubicon and BAE in disaster response efforts for both Hurricanes Harvey and Irma. The output of Slingshot’s analysis provided a comprehensive understanding of the extent of the flooding and its severity, ultimately assisting search and rescue operations of first responders. Slingshot also used its algorithms to quickly identify forward relief and triage staging areas close to non-flooded roadways and optimal hospital routes.

This application in disaster response was just one instance that leveraged the scale, speed and predictive power of our cloud native ecosystem. In just this past year, we’ve entered into contractual agreements with some of the biggest names in defense and are in current talks with stakeholders in several other verticals, including mining, real estate, financial services and civil/county offices.

Leveraging Startups to Address Pain Points

As my colleague and I stood behind the Slingshot booth at the Startup Exchange, the final day of BattleBorn, we were eager to network with the mining professional attendees. As they visited our space, the more insight they fed us. Some of it was feedback on our tech, but much of it, and more importantly, was identifying their pain points. Time and time again, mine managers and innovation directors pointed to the recurring problems underlying the risks associated with tailings dams. Given the environmental, social and economic hazards posed by the potential failure of these storage structures, we learned that the onus is on the mining manager to ensure that such a rupture never occurs during the full lifecycle of a site. While in-situ monitoring solutions do exist to address this concern, which require periodic on-site visits to the dams, mining professionals at the event emphasized the growing need for scalable remote sensing capabilities. This includes leveraging satellite imagery, with varying revisit rate requirements, and computer vision capabilities that can automatically spot anomalies to these earth-filled embankments. As we were told, any predictive metric on a potential dam failure would be hugely valuable. This key takeaway wasn’t just an industry insight, but a market opportunity for us at Slingshot, aligning perfectly with our core capabilities. Attending BattleBorn was paramount to the discovery of this business application.

Looking Forward

In stark parallel to the mine manager’s earlier quote, the mining industry may not be known as an early adopter of advanced tech, but events like BattleBorn are pushing this paradigm. Such a shift is widening the window for innovative technology providers, like Slingshot Aerospace, to enter the market. We look forward to our continued engagement with companies like Barrick, and any other mining operation looking to embrace this new wave of tech savvy, responsible mining.

Interested in learning more about Slingshot Aerospace? Feel free to visit us at www.slingshotaerospace.com or email us atross@slingshotaerospace.com for more information.

Through the Lens of a Startup: BattleBorn and the Paradigm of Innovation in Mining was originally published in Unearthed Community on Medium, where people are continuing the conversation by highlighting and responding to this story.