‘Observation’ is a tense, atmospheric puzzler where you play a modern HAL 9000

When you watch 2001: A Space Odyssey, do you find yourself criticizing HAL 9000’s machinations and thinking, “I could do better than that!” If so, Observation may be right up your alley. In it you play a space station AI called SAM that is called upon by the humans on board to help resolve a deadly mystery — though you may be a part of it yourself.

The game takes place in the near future on board the titular space station, a sort of expanded version of the ISS. You are booted up by astronaut Emma Fisher after an unspecified event that seems to have damaged the station. You, as the Systems Administration and Maintenance AI, are tasked with helping her out as she first tries to simply survive the immediate aftermath, then starts to investigate what happened.

To do so you perform various tasks such a digital agent would do, such as unlocking and opening hatches, checking for system errors, collecting information from damaged laptops, and so on. It’s mostly done through the many cameras mounted throughout the station, between which you can usually move freely and change the angle so you can get at this hatch or that scrap of paper on the wall.

But from the beginning it’s clear that this is not a simple case of a micrometeorite or some other common space anomaly. I won’t spoil any of the surprises, but suffice it to say that like in 2001, the mystery runs deeper than that, and SAM itself is implicated.

Observation is a puzzle game that plays out in real time, though you are rarely presented with a task that needs to be completed in a rush — your commands are rarely an urgent “Open the pod bay doors, SAM!” and more “Something’s wrong with the cooling system, so this hatch won’t open, can you look into it?”

And so you search using your cameras for, say, the server that controls that system, or the scrap of paper that has its schematic so you can reboot it. These solutions are usually just a matter of being, well, observant, but occasionally can be frustrating gadget hunts where you don’t know what you’re looking for among the busy background of a working space station and the detritus of the disaster.

If you’re having trouble with something, chances are you’re overthinking it. I had to look up the solution to one situation, and it turns out I had simply overlooked some interactive objects because they looked so much like background. (For the record, it turns out you can turn stuff on and off at power outlets.)

When you have to operate something, like an airlock, there is usually a little minigame to complete in which you must figure out which series of buttons to hit or hold — nothing too taxing, just a way to make it so you aren’t just pressing the Action Button all the time. The controls can be a bit clunky, such as one that had me hold down s to do one thing, then press and hold w at the same time. Do they not understand the same finger does both those things? Fortunately you can remap controls and although mouse movement is a bit stiff, there’s no need for twitchy response time.

Although the puzzles are a bit simplistic, it’s a pleasure navigating the station because it is so beautifully realized. The creators clearly did a ton of research and Observation, that is to say the station, is a convincing 21st century operation — cameras and laptops are stashed everywhere, sticky notes from the Russian and Chinese denizens, luggage and experiments tucked away or half finished.

It’s also all viewed through a combination of post-processing effects that make it all feel like you really are viewing it through a security camera system. These effects are a bit inconsistent — at one time you’ll hear what sounds like the whine of an 80s drive or system spinning up; others reflect a sort of Windows 98SE aesthetic; your own interface looks like something out of Terminator. It isn’t cohesive, exactly, but the truth is neither are the systems onboard the ISS and other space hardware. And it’s a nice touch that lets the developer differentiate each part of the station and the different devices you connect to.

The modeling of the main character, Emma, is also excellent, though lapsing a bit into uncanny valley territory due to some clunky animations here and there. Maybe it’s just the microgravity. But one thing that can’t be faulted is the voice acting — Emma’s actor is brilliant, and other voices you encounter are also well done. Considering the amount of dialogue in the game this could have been a dealbreaker, but instead it’s a pleasure to hear. Ambient audio is likewise lovely — wear headphones.

The atmosphere is oppressive and tense, but not exactly scary; Don’t expect a xenomorph to bust out of any vents, but also don’t expect Space Station Simulator 2019. This is a serious, adult (though not explicit or violent) sci-fi narrative and, from what I’ve played, a smart and interesting one.

I haven’t finished the game (which was sent to me in advance for review… but I’ve been in an intense love/hate relationship with Mordhau), but based on what I’ve played I can easily recommend Observation to anyone with a mind to take on mildly difficult puzzles and experience a well-presented story in a carefully crafted environment. Space buffs will also enjoy. At under $25 right now (less with this week’s sale going on) I’d say it’s a no-brainer.

Observation released earlier this week on the Epic Games and PlayStation stores.

DefinedCrowd offers mobile apps to empower its AI-annotating masses

DefinedCrowd, the Startup Battlefield alumnus that produces and refines data for AI-training purposes, has just debuted iOS and Android apps for its army of human annotators. It should help speed up a process that the company already touts as one of the fastest in the industry.

It’s no secret that AI relies almost totally on data that has been hand-annotated by humans, pointing out objects in photos, analyzing the meaning of sentences or expressions, and so on. Doing this work has become a sort of cottage industry, with many annotators doing it part time or between other jobs.

There’s a limit, however, to what you can do if the interface you must use to do it is only available on certain platforms. Just as others occasionally answer an email or look over a presentation while riding the bus or getting lunch, it’s nice to be able to do work on mobile — essential, really, at this point.

To that end DefinedCrowd has made its own app, which shares the Neevo branding of the company’s annotation community, that lets its annotators work whenever they want, tackling image or speech annotation tasks on the go. It’s available on iOS and Android starting today.

It’s a natural evolution of the market, CEO Daniela Braga told me. There’s a huge demand for this kind of annotation work, and it makes no sense to restrict the schedules or platforms of the people doing it. She suggested everyone in the annotation space would have apps soon, just as every productivity or messaging service does. And why not?

DefinedCrowd’s next-gen platform solves the AI data acquisition problem

The company is growing quickly, going from a handful of employees to over a hundred, spread over its offices in Lisbon, Porto, Seattle, and Tokyo. The market, likewise, is exploding as more and more companies find that AI is not just applicable to what they do, but not out of their reach.

AT&T outclassed Verizon in hurricane response, and it wasn’t close, union says

A Florida man sets up a sign that says,

Enlarge / PANAMA CITY, Fla. – OCTOBER 19: Mark Mauldin hangs a sign near the front of his property expressing his dissatisfaction with his Verizon cell phone service following Hurricane Michael, which slammed into the Florida Panhandle on October 10. (credit: Getty Images | Scott Olson )

filing with the Federal Communications Commission, which recently found that carriers’ mistakes prolonged outages caused by the hurricane. Many customers had to go without cellular service for more than a week.

It’s not surprising for a union to argue that union workers are preferable to contractors, of course. But it seems clear that AT&T did a better job than Verizon after the storm. In the days following the October 2018 hurricane, Florida Governor Rick Scott slammed Verizon for its poor hurricane response while praising AT&T for quickly restoring service.

Read 15 remaining paragraphs | Comments

Big revenues, huge valuations and major losses: charting the era of the unicorn IPO

We can make charts galore about the tech IPO market. Yet none of them diminish the profound sense that we are in uncharted territory.

Never before have so many companies with such high revenues gone public at such lofty valuations, all while sustaining such massive losses. If you’re a “growth matters most” investor, these are exciting times in IPO-land. If you’re the old-fashioned value type who prefers profits, it may be best to sit out this cycle.

Believers in putting market dominance before profits got their biggest IPO opportunity perhaps ever last week, with Uber’s much-awaited dud of a market debut. With a market cap hovering around $64 billion, Uber is far below the $120 billion it was initially rumored to target. Nonetheless, one could convincingly argue it’s still a rich valuation for a company that just posted a Q1 loss of around $1 billion on $3 billion in revenue.

So how do Uber’s revenues, losses and valuation stack up amidst the recent crop of unicorn IPOs? To put things in context, we assembled a list of 15 tech unicorns that went public over the past three quarters. We compared their valuations, along with revenues and losses for 2018 (in most cases the most recently available data), in the chart below:

 

Put these companies altogether in a pot, and they’d make one enormous, money-losing super-unicorn, with more than $25 billion in annual revenue coupled to more than $6 billion in losses. It’ll be interesting to revisit this list in a few quarters to see if that pattern changes, and profits become more commonplace.

History

It’s easy to draw comparisons to the decades-old dot-com bubble, but this time things are different. During the dot-com bubble, I remember penning this lead sentence:

“If the era of the Internet IPO had a theme song, it might be this: There’s no business like no business.”

That notion made sense for bubble-era companies, which commonly went public a few years after inception, before amassing meaningful revenues.

That tune won’t work this time around. If the era of the unicorn IPO had a theme song, it wouldn’t be nearly as catchy. Maybe something like: “There’s no business like lots of business and lots of losses too.”

I won’t be buying tickets to that musical. But when it comes to buying IPO shares, the unicorn proposition is a bit more appealing than the 2000 cycle. After all, it’s reasonably plausible for a company with dominant market share to tweak its margins over time. It’s a lot harder to grow revenues from nothing to hundreds of millions or billions, particularly if investors grow averse to funding continued losses.

Of course, the dot-com bubble and the unicorn IPO era do share a common theme: Investors are betting on an optimistic vision of future potential. If expectations don’t pan out, expect share prices to follow suit.

>20,000 Linksys routers leak historic record of every device ever connected

>20,000 Linksys routers leak historic record of every device ever connected

(credit: US Navy)

(credit: Troy Mursch)

Independent researcher Troy Mursch said the leak is the result of a flaw in almost three dozen models of Linksys routers. It took about 25 minutes for the Binary Edge search engine of Internet-connected devices to find 21,401 vulnerable devices on Friday. A scan earlier in the week found 25,617. They were leaking a total of 756,565 unique MAC addresses. Exploiting the flaw requires only a few lines of code that harvest every MAC address, device name, and operating system that has ever connected to each of them.

Read 9 remaining paragraphs | Comments

Microsoft open-sources a crucial algorithm behind its Bing Search services

Microsoft today announced that it has open-sourced a key piece of what makes its Bing search services able to quickly return search results to its users. By making this technology open, the company hopes that developers will be able to build similar experiences for their users in other domains where users search through vast data troves, including in retail, though in this age of abundant data, chances are developers will find plenty of other enterprise and consumer use cases, too.

The piece of software the company open-sourced today is a library Microsoft developed to make better use of all the data it collected and AI models it built for Bing .

“Only a few years ago, web search was simple. Users typed a few words and waded through pages of results,” the company notes in today’s announcement. “Today, those same users may instead snap a picture on a phone and drop it into a search box or use an intelligent assistant to ask a question without physically touching a device at all. They may also type a question and expect an actual reply, not a list of pages with likely answers.”

With the Space Partition Tree and Graph (SPTAG) algorithm that is at the core of the open-sourced Python library, Microsoft is able to search through billions of pieces of information in milliseconds.

Vector search itself isn’t a new idea, of course. What Microsoft has done, though, is apply this concept to working with deep learning models. First, the team takes a pre-trained model and encodes that data into vectors, where every vector represents a word or pixel. Using the new SPTAG library, it then generates a vector index. As queries come in, the deep learning model translates that text or image into a vector and the library finds the most related vectors in that index.

“With Bing search, the vectorizing effort has extended to over 150 billion pieces of data indexed by the search engine to bring improvement over traditional keyword matching,” Microsoft says. “These include single words, characters, web page snippets, full queries and other media. Once a user searches, Bing can scan the indexed vectors and deliver the best match.”

The library is now available under the MIT license and provides all of the tools to build and search these distributed vector indexes. You can find more details about how to get started with using this library — as well as application samples — here.

The radio navigation planes use to land safely is insecure and can be hacked

A plane in the researchers' demonstration attack as spoofed ILS signals induce a pilot to land to the right of the runway.

Enlarge / A plane in the researchers’ demonstration attack as spoofed ILS signals induce a pilot to land to the right of the runway. (credit: Sathaye et al.)

software defined radio, the researchers can spoof airport signals in a way that causes a pilot’s navigation instruments to falsely indicate a plane is off course. Normal training will call for the pilot to adjust the plane’s descent rate or alignment accordingly and create a potential accident as a result.

Read 36 remaining paragraphs | Comments

Week-in-Review: Google impersonates Apple and Bezos eyes the moon

After Mark Zuckerberg’s privacy mea culpa at F8 last week, Google got its turn at I/O to promise consumers that their data wasn’t going anywhere that they didn’t want it to go. In short: they aimed to take a page from Apple.

For Google, a clear strategy at the event was essentially highlighting how it wasn’t collecting user data in certain circumstances; those circumstances seemed to be largely focused on whenever the data wasn’t all that useful to Google to begin with.

Here’s everything Google announced at the I/O 2019 Keynote

Google received applause for “privacy commitments” on its new Next Home Max like not cloud-uploading a 3D mesh of a user’s face that’s used for tailoring information, as though doing it in the cloud would make any sense for the company to do. Keeping information on-device was still the exception to the rule at I/O this week, the cloud is still where Google keeps its sharpest wits.

Shoot me tips or feedback
on Twitter @lucasmtny or email
lucas@techcrunch.com

Are expectations so low that we expect Google to keeps everything that we enter on its services forever? While the company has introduced opt-out auto-delete features for information like location data and web history, Google still writes the rules for you handle your own data, specifying that you still allow the data to stay on Google servers for either 3 or 18 months, periods of time that still allow Google to hold onto the most relevant of your info in order to surface information.

Privacy is a product of tradeoffs when you’re online, but companies like Google often seem to communicate that trading information is an inevitability of getting a tailored experience.

Look at a product from Apple like News+, one would imagine that the only way a service could understand what you like to read is by handing that user information to the service owner and sending those suggestions back down. Apple instead handles this by sending users a package of articles and by using on-device processing, the company is able to suggest your next article without publishers or Apple knowing what it is that you’re reading across the network of sites.

Apple is of course reliant on a business model that’s focused on selling hardware not advertising, and thus they’re in a bit more understandable of a position when it comes to eschewing personal data collection in most circumstances.

The company made some partial progress towards righting privacy wrongs, but the biggest winner in the company’s privacy rebrand was meant to be Google.

My colleague Josh had a less cynical view on Google’s promises, though we both share the opinion that Facebook doesn’t deserve much trust in its new privacy “mission,” here’s his piece that runs counter to about everything above:

Facebook talked privacy, Google actually built it

Trends of the week

Here are a few big news items from big companies, with green links to all the sweet, sweet added context.

  • Bezos’ moon moonshot
    Amazon’s CEO is psyched about the moon but he’s not planning to put HQ2 on it, he wants his rocket company Blue Origin to plant its newly-unveiled lunar lander on the surface. At a dedicated event, Bezos discussed some of the company’s plans to turn the moon into a future home for humanity. “It’s time to go back to the Moon. This time to stay,” Bezos said.
  • No five-star rating for Uber opening
    After several weeks of headlines surrounding Lyft’s disappointing public debut, Uber got its turn Friday. The result was none too pretty, after opening at the lowest end of its range, the company still dove in first day trading ending the day with a share price of $41.50. The company has an uphill road to profitability, but as it looks to cut costs, some Uber drivers showcased at a protest that they were already feeling the squeeze.
  • Elon’s tweets land him in more trouble
    Elon’s tweet about the cave-diver, calling him a “pedo guy” is going to trial after all. You can peep the legal documents alongside out full story here.
  • I have you now
    This week, I wrote a feature on a tiny Czech game studio that’s built the most popular VR game on the market. It involves light sabers and EDM and a music-mixing CEO who had plenty to say about banking $20 million in revenue and opting not to raise any outside cash along the way.

AP Photo/Jeff Chiu

GAFA Gaffes

How did the top tech companies screw up this week? This clearly needs its own section, in order of awfulness:

  1. Googlers want some acknowledgment from the top:
    [Google employees demand Larry Page address walkout and retaliation]
  2. Protestors take to the skies during Google’s privacy keynote:
    [Protestors fly banner-towing plane over Google I/O]

Extra Crunch

Our premium subscription service continues to churn out some great pieces. We had a fascinating piece go up this week where my colleague Eric chatted with some of Silicon Valley’s most prolific investors about where the puck is moving on media investments. Here’s one tidbit from Sequoia’s Stephanie Zhan:

Where top VCs are investing in media, entertainment & gaming

At Sequoia, we see incredible potential for the world of gaming and entertainment. Among others, we are excited about: The next virtual third place where consumers are hanging out with friends. In the past, your local Starbucks or AIM would have been your go-to place to see and be seen. Today…

Here are some of our other top reads this week for premium subscribers, check out the read butting heads with growing discontent for office Slack usage…

  • Against the Slacklash
  • AWS remains in firm control of the cloud infrastructure market
  • Keyword research in 2019: modern tactics for growing targeted search traffic

Want more TechCrunch newsletters? Sign up here.