‘Frozen’ CG snow and crash-test cadavers offer hints for 60-year-old Russian mystery deaths

The Dyatlov Pass incident is the mother of all cold cases: nine people found dead in 1959, deep in the Ural mountains, under circumstances no one has ever been able to satisfactorily explain. But new research uses simulation techniques from multiple eras to advance what is perhaps the least implausible story of this tragic mystery.

The paper, published yesterday in Nature Communications Earth and Environment, was accompanied by a highly readable summary in National Geographic, which is very much worth your time. (Even if the headline is the dreaded “Has science solved…?”)

Essentially the mystery is this: The eight students and their ski instructor had pitched their tent on a slope that seemed safe — if not perfectly so then comparatively considering the surroundings at Kholat Syakhl, or “Dead Mountain” — but were later found spread out around the area in various stages of disrobing and destruction. The carnage seemed beyond what an avalanche would produce, and anyway there seemed to be no evidence or likelihood of one in the first place.

For more than 60 years this has been a source of speculation and conspiracy, especially since there was the appearance of a cover-up by the Soviet government at the time. Even Russia revisiting the event in 2019 didn’t seem to produce a convincing explanation.

Enter Alexander Puzrin and Johan Gaume, from Switzerland’s ETH Zürich and EPFL, respectively, two highly prestigious and advanced technical institutes. Curious about the incident for their own reasons, they began looking into how to work out once and for all what happened. An interesting personal detail:

The scientific investigation came with an added benefit from Puzrin’s wife, who is Russian. “When I told her that I was working on the Dyatlov mystery, for the first time she looked at me with real respect,” he says.

One hardly knows what to say!

At all events the researchers put together a new hypothesis based on a few ideas.

First, the slope was not as shallow as it appeared — it was near the minimum for an avalanche to occur, and the snow was characterized as having a base layer conducive to slippage of snow on top. Freezing winds could have added mass and set off a slide under the cut-out in which the group put their tent.

Second, Gaume visited the creators of the movie “Frozen,” which featured highly realistic snow simulation. He met with Disney’s snow simulation specialist and got permission to use and modify the code — but in this case, to see what an avalanche striking sleeping students would do to them. Their simulations showed that it wouldn’t take much — a block of icy snow the size of a large car — to cause the devastation witnessed by the rescue party.

Diagram showing simulation of snow as it might impact a person sleeping in a tent at Dyatlov.

Image Credits: Gaume, Puzrin / Nature

Third, they used research performed by GM that broke the ribs of a hundred cadavers — for the purposes of tuning seatbelts. They proposed that because the Russian students would have been sleeping on their skis, it was fairly similar to how certain cadavers with rigid supports reacted to impacts. Thus the horrific injuries instead of the usual asphyxiation produced by being submerged in a drift that usually happens to victims of avalanches.

It’s all still speculation on top of speculation, but the important part is that by combining these various, reasonably objective measures, Puzrin and Gaume show that it’s possible that an avalanche was responsible for the Dyatlov Pass incident, however rare the combination of circumstances must have been.

They freely admit that many may not accept this explanation — “It’s too normal,” said Gaume — and will continue to pursue the conspiracies and fantasy scenarios the incident has spawned for half a century. But for others it may offer some solace: a reason to believe that these poor nine souls were just in the wrong place at the wrong time.

Reap big benefits when you attend both TC Early Stage 2021 events

Where can new founders and budding entrepreneurs turn for expert advice to navigate the formative phases of building a startup? Head to TechCrunch Early Stage — a two-day virtual bootcamp that gives early founders (pre-seed through Series A) access to the leading experts across a range of essential entrepreneurial skills.

We’re talking dozens of workshops addressing operations, fundraising, pitch deck pointers, term sheet tips, product-market fit, brand building, growth marketing, recruiting, taming your tech stack and a lot more.

That’s a lot of ground to cover, amirite? That’s exactly why we’re hosting two Early Stage events this year. Each one offers different content, a separate slate of speakers and unique perspectives. Both feature highly interactive Q&As with the experts. Get answers to all your burning questions!

Note: In a hurry? Scroll down to the bottom (or click here) to get the 411 on pass types, early-bird pricing and deadlines.

The first TC-ES, on April 1 & 2, covers topics ranging from fundraising and operations to product lifecycle and recruiting — for starters. The second, on July 8 & 9, spans marketplace positioning, growth marketing, content development and even more on fundraising.

Sure, you can go to a single Early Stage event, but savvy startuppers (that’s you, right?) will sign on for a double dose of knowledge and attend both. That’s not just marketing talk — the benefits are real.

Buy a dual-event pass (for a tidy discount, by the way), and you’ll not only learn more, but you’ll also have more time to absorb and implement critical advice that can lead to your success. Seriously, which topics and tips can you afford to miss? Don’t let the FOMO haunt you.

You’ll also have plenty of opportunity to connect and network with other early founders, later-stage founders and other smart members of the startup — oh, what’s the word? — community. Build your contacts, find indispensable support and mentorship. Remember, we all go further together.

We introduced TC Early Stage last year to rave reviews — like this one from Chloe Leaaetoa, founder of Socicraft:

You learn from industry leaders and seasoned founders — people who’ve already been there and done that. They were genuine and honest about industry expectations. Plus, they shared first-hand accounts, which made them more relatable.

And this one from Ashley Barrington, founder of MarketPearl:

I recommend going to Early Stage. The virtual aspect helps in terms of scheduling, it offers community-building through networking, and it gives early stage founders a framework for navigating the startup ecosystem. This is the stage where founders need more support, especially if they haven’t done this before.

Go all-in and attend TC Early Stage 2021 in April and again in July. Increase your knowledge, sharpen your skills, avoid pitfalls and expand your network. Those are mighty big benefits, and you deserve every one of them.

Is your company interested in sponsoring or exhibiting at Early Stage 2021 – Operations & Fundraising? Contact our sponsorship sales team by filling out this form.

Customer advisory boards are a gold mine for startup brand champions

Yousuf Khan
Contributor

Yousuf Khan is a partner at Ridge Ventures. Prior to joining Ridge, he was the first CIO of Automation Anywhere, CIO and Vice President of Customer Success at cloud-based AI platform Moveworks, as well as CIO of Pure Storage, Qualys and Hult International Business School.

As a 20-year CIO and advisor to multiple startups, I sat on many customer advisory boards (CABs) and saw how they were formed. Some companies have highly functioning CABs, others merely serve as feedback loops. Any startup striving to connect directly with their customers would benefit from establishing one.

Here are some considerations to make certain your customer advisory board is a success.

Why CABs matter

For those unfamiliar, a customer advisory board is a group of customers who come together to share their experiences, insights and advice with an organization. First and foremost, the CAB functions to recognize and include the voice of the customer, an essential part of your company’s journey since customers interact closer than anyone with your product or service.

It’s best to designate early adopters to be on the board — those who took a chance on you and have been on the frontlines as your company evolved — as well as some newer customers.

While establishing this group signals appreciation and respect for your customers, it also provides an opportunity for you to formalize and structure the feedback you are requesting from them. You can seek validation for product ideas or guidance on roadmap development, test out marketing messaging and even tap into market intelligence.

The greatest benefit of a CAB, however, is the creation of champions for your brand. These loyal partners will ultimately offer testimonials, references and referrals. Key to this partnership is a shared sense of playing a small part in building the future of your company.

The greatest benefit of a CAB is the creation of champions for your brand.

Assembling your CAB superteam

The best route to assembling your CAB is to start with a very small group and expand slowly. There’s quite a bit of nuance in the selection of who to include. Do you go after the executive who sponsored you, the one who saw a vision and thought your solution would fit?

Perhaps. But that individual may not be using your product every day, be involved deeply in its operational aspects and/or have their finger on the pulse of the end-user’s experience.

Three-dimensional search engine Physna wants to be the Google of the physical world

In June of 1999, Sequoia Capital and Kleiner Perkins invested $25 million into an early-stage company developing a new search engine called Google, paving the way for a revolution in how knowledge online was organized and shared.

Now, Sequoia Capital is placing another bet on a different kind of search engine, one for physical objects in three dimensions, just as the introduction of three-dimensional sensing technologies on consumer phones are poised to create a revolution in spatial computing.

At least, that’s the bet that Sequoia Capital’s Shaun Maguire is making on the Columbus, Ohio-based startup* Physna.

Maguire and Sequoia are leading a $20 million bet into the company alongside Drive Capital, the Columbus, Ohio-based venture firm founded by two former Sequoia partners, Mark Kvamme and Chris Olsen.

“There’s been this open problem in mathematics, which is how you do three-dimensional search. How do you define a metric that gives you other similar three-dimensional objects. This has a long history in mathematics,” Maguire said. “When I first met [Physna founder] Paul Powers, he had already come up with a wildly novel distance metric to compare different three-dimensional objects. If you have one distance metric, you can find other objects that are a distance away. His thinking underlying that is so unbelievably creative. If I were to put it in the language of modern mathematics… it just involves a lot of really advanced ideas that actually also works.”

Powers’ idea — and Physna’s technology — was a long time coming.

A lawyer by training and an entrepreneur at heart, Powers came to the problem of three-dimensional search through his old day job as an intellectual property lawyer.

Powers chose IP law because he thought it was the most interesting way to operate at the intersection of technology and law — and would provide good grounding for whatever company the serial entrepreneur would eventually launch next. While practicing, Powers hit upon a big problem, while some intellectual property theft around software and services was easy to catch, it was harder to identify when actual products or parts were being stolen as trade secrets. “We were always able to find 2D intellectual property theft,” Powers said, but catching IP theft in three dimensions was elusive.”

From its launch in 2015 through 2019, Powers worked with co-founder and chief technology officer Glenn Warner Jr. on developing the product, which was initially intended to protect product designs from theft. Tragically, just as the company was getting ready to unveil its transformation into the three-dimensional search engine it had become, Warner died.

Powers soldiered on, rebuilding the company and its executive team with the help of Dennis DeMeyere, who joined the company in 2020 after a stint in Google’s office of the chief technology officer and technical director for Google Cloud.

“When I moved, I jumped on a plane with two checked bags and moved into a hotel, until I could rent a fully furnished home,” DeMeyere told Protocol last year.

Other heavy hitters were also drawn to the Cincinnati-based company thanks in no small part to Olsen and Kvamme’s Silicon Valley connections. They include GitHub’s chief technology officer, Jason Warner, who has a seat on the company’s board of directors alongside Drive Capital’s co-founder Kvamme, who serves as the chairman.

In Physna, Kvamme, Maguire and Warner see a combination of GitHub and Google — especially after the launch last year of the company’s consumer-facing site, Thangs.

That site allows users to search for three-dimensional objects by a description or by uploading a model or image. As Mike Murphy at Protocol noted, it’s a bit like Thingiverse, Yeggi or other sites used by 3D-printing hobbyists. What the site can also do is show users the collaborative history of each model and the model’s component parts — if it involves different objects.

Hence the GitHub and Google combination. And users can set up profiles to store their own models or collaborate and comment on public models.

What caught Maguire’s eye about the company was the way users were gravitating to the free site. “There were tens of thousands of people using it every day,” he said. It’s a replica of the way many successful companies try a freemium or professional consumer hybrid approach to selling products. “They have a free version and people are using it all the time and loving it. That is a foundation that they can build from,” said Maguire.

And Maguire thinks that the spatial computing wave is coming sooner than anyone may realize. “The new iPhone has lidar on it… This is the first consumer device that comes shipped with a 3D scanner with lidar and I think three dimensional is about to explode.”

Eventually, Physna could be a technology hub where users can scan three-dimensional objects into their phones and have a representational model for reproduction either as a virtual object or as something that can be converted into a file for 3D printing.

Right now, hundreds of businesses have approached the company with different requests for how to apply its technology, according to Powers.

One new feature will allow you to take a picture of something and not only show you what that is or where it goes. Even if that is into a part of the assembly. We shatter a vase and with the vase shards we can show you how the pieces fit back together,” Powers said.

Typical contracts for the company’s software range from $25,000 to $50,000 for enterprise customers, but the software that powers Physna’s product is more than just a single application, according to Powers.

“We’re not just a product. We’re a fundamental technology,” said Powers. “There is a gap between the physical and the digital.”

For Sequoia and Drive Capital, Physna’s software is the technology to bridge that gap.

You can now give Facebook’s Oversight Board feedback on the decision to suspend Trump

Facebook’s “Supreme Court” is now accepting comments on one of its earliest and likely most consequential cases. The Facebook Oversight Board announced Friday that it would begin accepting public feedback on Facebook’s suspension of former President Trump.

Mark Zuckerberg announced Trump’s suspension on January 7, after the then-president of the United States incited his followers to riot at the nation’s Capitol, an event that resulted in a number of deaths and imperiled the peaceful transition of power.

In a post calling for feedback, the Oversight Board describes the two posts that led to Trump’s suspension. One is a version of the video the president shared the day of the Capitol riot in which he sympathizes with rioters and validates their claim that the “election was stolen from us.” In the second post, Trump reiterates those views, falsely bemoaning a “sacred landslide election victory” that was “unceremoniously & viciously stripped away.”

The board says the point of the public comment process is to incorporate “diverse perspectives” from third parties who wish to share research that might inform their decisions, though it seems a lot more likely the board will wind up with a tidal wave of subjective and probably not particularly useful political takes. Nonetheless, the comment process will be open for 10 days and comments will be collected in an appendix for each case. The board will issue a decision on Trump’s Facebook fate within 90 days of January 21, though the verdict could come sooner.

The Oversight Board specifically invites public comments that consider:

Whether Facebook’s decision to suspend President Trump’s accounts for an indefinite period complied with the company’s responsibilities to respect freedom of expression and human rights, if alternative measures should have been taken, and what measures should be taken for these accounts going forward.

How Facebook should assess off-Facebook context in enforcing its Community Standards, particularly where Facebook seeks to determine whether content may incite violence.

How Facebook should treat the expression of political candidates, office holders, and former office holders, considering their varying positions of power, the importance of political opposition, and the public’s right to information.

The accessibility of Facebook’s rules for account-level enforcement (e.g. disabling accounts or account functions) and appeals against that enforcement.

Considerations for the consistent global enforcement of Facebook’s content policies against political leaders, whether at the content-level (e.g. content removal) or account-level (e.g. disabling account functions), including the relevance of Facebook’s “newsworthiness” exemption and Facebook’s human rights responsibilities.

The Oversight Board’s post gets very granular on the Trump suspension, critiquing Facebook for lack of specificity when the company didn’t state exactly which part of its community standards were violated. Between this and the five recent cases, the board appears to view its role as a technical one, in which it examines each case against Facebook’s existing ruleset and then makes recommendations for future policy rather than working backward from its own broader recommendations.

The Facebook Oversight Board announced its first cluster of decisions this week, overturning the company’s own choice to remove potentially objectionable content in four of five cases. None of those cases pertained to content relevant to Trump’s account suspension, but they prove that the Oversight Board isn’t afraid to go against the company’s own thinking — at least when it comes to what gets taken down.

Nerdy, a tutoring marketplace startup, is going public via SPAC

A gig economy-powered consumer edtech platform is heading to the New York Stock Exchange.

Edtech startup Nerdy, which owns the popular tutoring business Varsity Tutors, is seeking to become a public company through a special purpose acquisition vehicle, otherwise known as a SPAC.

Nerdy will merge with TPG Pace Tech Opportunities (NYSE: PACE), a publicly traded SPAC since 2015. The transaction is expected to close in the second quarter of this year.

The deal will value Nerdy at $1.7 billion. Through the transaction, the business plans to raise up to $750 million in cash, including $150 million in PIPE financing aggregated by Franklin Templeton, Healthcare of Ontario Pension Plan, Koch Industries and Learn Capital.

Nerdy’s flagship business, Varsity Tutors, is a two-sided marketplace that matches tutors to students in large, small or 1:1 group environments. The learning platform covers more than 3,000 subjects. Like other edtech companies, Varsity Tutors uses artificial intelligence and data analytics to better match experts to learners. Additionally, in August, Varsity Tutors launched a homeschooling offering meant to replace traditional school. It onboarded 120 full-time educators, who came from public schools and charter schools, with competitive salaries.

Financial performance

TechCrunch reviewed the Nerdy-SPAC investor presentation, which can be read here.

Nerdy is among consumer edtech businesses that saw rapid growth and opportunity due to the demands of remote learning brought about by the coronavirus pandemic. In the second half of 2020, Nerdy’s annualized revenue surpassed $120 million. In the last quarter of 2020, the company saw its online revenue grow 87%, online paid active learners grow 59% and paid online sessions grow 169%, compared to the same time period last year, the business reports.

Drilling into its realized results instead of its more-favorable annualized performance from its third and fourth quarters of 2020, Nerdy saw estimated revenues of $106 million in the year, up just 16% from its 2019 result.

That growth rate is slower than what it managed in 2019, some 26% growth, and is around half of what it anticipates for 2021, namely 31% growth. But Nerdy has even stronger projections for 2022, a year in which it expects to drive revenues of $198 million, up 43% from its 2021 expectation of $138 million.

Whether the company can hit those goals remains to be seen; SPAC-led debuts allow for the company being taken public in the transaction to forecast more than companies that follow traditional IPO paths are allowed.

The company’s growth also failed to stem its losses. Nerdy is not yet profitable. Its 2020 estimates list an anticipated net loss of $23 million, which is more than it lost in 2019 but less than its 2018 deficit. Based on last year’s growth, Nerdy estimates that its net loss will slim to $8 million in 2021, and will achieve profitability by 2023.

How did Nerdy fail to reduce its losses last year as its revenues expanded? The company’s costs showed modest gains and losses, apart from its sales and marketing line item. That particular realm of expense rose from $38 million in 2019 to an estimated $44 million in 2020.

In contrast, while Nerdy’s net losses were largely static in 2020, its estimated net margin did improve from -24% in 2019 to an estimated -22% in 2020. It has a ways to go to reach the black, though its financials do indicate that the company thinks that net income is only a few years away.

To reach profitability, Nerdy anticipates it will require 2023 revenues of $267 million, growth from 2022 of 35% and gross margins five points stronger than its 67% result it estimated it reached last year.

A closer look at Nerdy’s business brings up a common question amid the SPAC boom: Is the reverse-merger being used to bring companies with lackluster near-term growth stories to the public market that otherwise couldn’t have? So far, a number of edtech startups have taken the SPAC route, including Skillsoft, Meten International and now Nerdy.

After edtech had a strong 2020, sector investors say to expect more exits as startups cross the $100 million ARR mark. Deborah Quazzo, managing partner of GSV, told TechCrunch in December that “what’s happening in edtech is that capital markets are liquidating.” The ability to move fluidly between privately held and publicly held companies is a characteristic of tech sectors with deep capital markets, which is different from edtech’s “old days, where the options to exit were very narrow.”

Internet of Cars: A driver-side primer on IoT implementation

Mathew Yarger
Contributor

Mathew is Head of Mobility and Automotive at the IOTA Foundation, where he develops strategy and solutions around the use of data with distributed ledger technology-based innovations. These solutions focus on enabling flexible and privacy-centric data and value transfer within the public sector for use in smart cities, critical infrastructure, environmental and energy systems, and mobility-focused data industries.

Billions of devices are currently connected to the Internet of Things (IoT), and researchers are predicting tremendous growth in the coming decade.

One of the most exciting, challenging and potentially lucrative areas of the IoT is the automotive sector. The car is a major component of most people’s daily lives, and a “smart” car could do a lot to save people time and money.

At the same time, the “Internet of Cars” carries with it dystopian visions of increased ad noise and security threats. It’s worth considering for a moment what these scenarios look like — good and bad — and how consumers can educate themselves to ensure that the cars of the future are driving in the right direction.

The car is a major component of most people’s daily lives, and a “smart” car could do a lot to save people time and money.

The promises and problems of connected cars

Imagine if your car was able to call your mechanic when the engine was showing signs of trouble. Imagine if the mechanic could read a data report from your engine and order the required parts ahead of time. Imagine if the data on those parts could be aggregated to warn of the need for any mass recalls? What if your car could communicate with other cars around it in a traffic jam, and the cars could all work together to space out and ease congestion?

What if your car could pay automatically at parking garages and drive-throughs? Anyone that owns a car is familiar with all these pain points, and the prospect of a new system that erases these spots of friction would be a welcome development.

But how can we ensure that all of this new data from our smart cars will be handled in a secure and private way? It seems likely, as car manufacturers work quickly to bring their products online, that tech giants will be the first partners to help implement the Internet of Cars. This might be cause for concern amongst consumers who are growing tired of their data being sold or hacked. The big tech companies aren’t inherently evil, but their basic business models are structured in such a way that consumer privacy and security are not the main priorities.

It’s not hard to imagine how the Internet of Cars could move in a much darker direction: Advertisements with real-time location data updating constantly on your windshield, personal data such as your driving habits stored on central servers, and a myriad of new vulnerabilities for hackers to exploit. How do we bring cars online so that friction in our lives is smoothed down without introducing a unique set of new problems?

Data security must be the foundation of the IoC

Of course Big Tech companies will be eager to offer connectivity for drivers, but it’s most likely going to come at the price of giving personal data over to Big Tech servers. This brings with it, as always, two major problems. The first is that centralized data represents a honeypot for hackers. No matter the strength of the security system, hackers realize that once they break through, they have access to the whole pot. The second problem is that the value of all that data is simply too lucrative for the owner to ignore. The data will always be sold, regardless of all the lip service promising to make it anonymous.

The IoT represents a new layer of IT integration in our lives; it will be at least as much of a game-changer as the internet was originally. Even with the advancement of the mobile internet brought about by smartphones, internet implementations have, until now, basically been carried out through clunky interfaces like screens, keyboards and mouses. The IoT is going to bring a new level of sophistication to how and where we interface online, but this also means a new level of intrusion into our physical reality. In the case of cars, we can be rightly wary that this new development might be problematic, but it doesn’t have to be.

Distributed ledger technology (DLT) represents a path forward for the Internet of Cars, because it builds data security and privacy into the foundations of any connected devices. Any model of DLT includes some basic concepts such that data is carried on a decentralized network of computers and servers. It also means that data is stored permanently, and that new entries of any data are subject to a mathematical verification. DLT is a fundamentally different way to handle massive amounts of data. DLTs have proven to be extremely resilient to attacks, and the data on these networks is nearly impossible to collect and sell.

Picking the right tool for the job

There are millions of internet-connected cars already on the road, albeit mostly with crude subscription services for music and weather apps. With further advances, connection will be much more encompassing, with the average connected car having up to 200 sensors installed, each recording a point of data, minute by minute. The numbers quickly become staggering, and in emergency situations, the need for data agility is apparent. Picture driving on a highway in medium traffic.

If someone’s tire blows out half a mile ahead, this information could be quickly conveyed to surrounding cars, warning of the potential for emergency braking. Any DLT solution would have to include a very nimble verification process for all these new packets of information to be brought into and carried by the network.

Additionally, because of the computational complexity involved, almost all DLTs today charge a fee for each new transaction brought into the network. In fact, the fee is an integral part of the structure of many of these computational models. This is obviously not going to be workable in a system like urban traffic that would be generating billions of “transactions” every day. The truth is that decentralized data networks were never designed to handle these kinds of massive use-case scenarios. Blockchain, for example, is very elegant at censorship-resistance in a network, and this has proven valuable in certain financial use cases.

But a DLT that expects a little money every time a car’s air conditioning reports its output is simply unusable for that application. Any DLT that’s going to give us a high level of security and real-time connectivity will also have to be feeless.

Security, speed and ease of adaptability through a no-fee structure are the three critical points for any network backing up the Internet of Cars. DLTs are clearly the most secure option, but they must also provide scalability and a feeless structure.

The example of being able to pay automatically for a parking garage visit might seem like a trite convenience. In actuality, if we can implement these types of small transactions properly from the beginning, then the hurdles we will jump in solving the complexity and volume of the car traffic data environment will go a long way to creating a safe, consumer-friendly Internet of Things in general.

When thinking about a completely connected physical environment, the alternatives to scalable, fee-less DLT are frankly scary.

Firehawk Aerospace extends seed funding to $2.5 million with $1.2 million from Harlow Capital

Rocket fuel technology startup Firehawk Aerospace has added $1.2 million to its existing seed financing, bringing the full amount invested in the round to $2.5 million. The new tranche comes from Harlow Capital Management, a Dallas-based firm run by Colby Harlow, who will join Firehawk’s board of directors as part of the deal.

Firewhawk, which was a finalist in our first-ever all-virtual Startup Battlefield at TC Disrupt last September, has developed a new kind of hybrid rocket fuel that greatly enhances rocket launch safety, cost and transportation using additive manufacturing (basically, the grown-up version of 3D printing). Hybrid rocket fuel (which combines aspects of both liquid and solid propellants used previously) isn’t new, but past technology has been unable to compete on cost and efficacy relative to existing nonhybrid alternatives.

The startup’s Chief Scientist Ron Jones was able to get around these limitations with two new approaches: Using a fuel with a hard polymer structure and producing it using additive manufacturing instead of casting via molds with a liquid that hardens.

Firehawk now intends to use its seed funding to test its technology in operational conditions and at the kind of scale required for commercialization, and to build out its partnerships and client list. The startup also intends to grow its R&D and manufacturing operations in both Texas and Oklahoma.

Okta SaaS report finds Office 365 wins the cloud — sort of

Each year Okta processes millions of SaaS logons via its authentication system. It kindly aggregates that data to find the most popular apps and publishes an annual report. This year it found that the most popular tool by far was Microsoft Office 365.

It’s worth noting that while app usage popularity varied by region, Office 365 was number one with a bullet across the board, whether globally or when the report broke it down by geographic area. That wasn’t true of any other product in this report, so Office 365 has extensive usage across the world (at least among companies that use Okta).

But as with everything cloud, it’s not a simple matter to say that because lots of people signed onto Office 365, Microsoft is the clear winner in a broader sense. In reality, the cloud is a complex marketplace, and just because people use one tool doesn’t preclude them from using tools that compete directly with it.

As a case in point, consider that the report found that 36% of Microsoft 365 customers were also using Google Workspace (formerly known as G Suite), which offers a similar set of office productivity tools. Further, Okta found that 42% of Office 365 customers were using Zoom and 32% were using Slack.

This is pretty remarkable when you consider that Office 365 bundles Teams with similar functionality for free. What’s more, so does Google with Google Hangouts, so people use the tool they want when they want, and sometimes it seems they use competing versions of the same tool. The report also found that of those Office 365 users, 44% are using Salesforce, 41% AWS, 15% Smartsheet and 14% Tableau (which is owned by Salesforce). Microsoft has products in all those categories.

Microsoft is clearly a big company with a lot of products, but the report blows a hole in the idea that because people like Office 365, they are going to be big fans of other Microsoft products, or that they can count on any kind of brand loyalty across the range of products or even exclusivity within the same product category.

All of this, and much of the other data in this report makes tremendously interesting reading as far as it goes. It’s not a definitive window on the state of SaaS. It’s a definitive reading on the state of Okta customers’ use of SaaS, on the Okta Integration Network (OIN), a point the company readily acknowledges in the report’s methodology section.

“As you read this report, keep in mind that this data is representative of Okta’s customers, the applications and integrations we connect to through the OIN, and the ways in which users access these tools through our service,” the report stated.

But it is a way to look at the state of SaaS taking advantage of the 9400 Okta customers using the network and the 6,500 integrations to the world’s most popular SaaS tools. That gives the company a unique view into the world of SaaS. What you can conclude is that the cloud is complicated, and it’s not a zero-sum game by any means. In fact, being a winner in one area is not a guarantee of winning across the board.