Author: azeeadmin

IDC’s latest tablet numbers tell a familiar story. Shipments are down yet again, a trend that shows no sign of abating. There is, however, one clear bright spot in all of this, bucking the trend.

Quarter after quarter, convertibles have continued to grow, as users have demanded more productivity than is traditionally possible with slates. It’s no surprise, then, that Apple, Microsoft, Google and Samsung are all vying for mindshare. The convertible segment may still be small, but at least it’s moving in the right direction.

Samsung’s shifted approaches a number of times over the years. Say what you will about the company’s approach to devices, but at least the company’s everything-and-the-kitchen-sink model has allowed them to stay limber for such a hulking hardware behemoth. The Galaxy Tab S4 simultaneously finds the company building on existing technology, while attempting to reinvent the wheel in the process.

The Tab S4 isn’t a radical departure from its predecessor, save for one key thing: DeX. The brand, which previously referred to smartphone docking stations for Galaxy handsets, is now the name of the custom desktop version of Android Samsung built. The change clearly hasn’t fully taken, however — Samsung’s official DeX site has yet to reflect the move as of this writing. But, then, the company’s clearly been busy these past few weeks.

Porting it over to the tablet means the company can offer a device that’s capable of doing double duty as both a standard slate tablet and a makeshift laptop. It’s also a roundabout of way of not relying on Windows 10 on the Galaxy line. After all, the ability to switch modes has long been the operating system’s raison d’être, so this really feels like a thumb in the eye of Microsoft. Given the recent well-received launch of the Surface Go, however, I suspect the company’s not really sweating the slight.

I didn’t get a great justification from Samsung as to the motivation behind the move ahead of launch. I suspect the whole thing boils down to control — something you’re afforded a lot more of when you work with Android vs. Windows. There’s also the fact that Windows 10 doesn’t have nearly the selection of apps you get with Android — definitely a major factor if you’re working with the more locked down/tablet-friendly 10 S.

DeX in effect

The result is something of a mixed bag. I will say that the DeX layout looks better on the 10.5-inch display, rather than the big, curved DeX docking monitor, where the Android icons feel entirely out of place.

For the most part, the DeX version of Android operates the same as the standard tablet version, allbeit with smaller icons. The desktop features three by default: My Files, Gallery and Settings. With the latter, you can adjust the DeX settings, so the OS automatically switches to the mode when you plug the tablet into the keyboard dock (though not the other way around, oddly).

I’d prefer it if the DeX settings were more easily accessible in the drop down menu, but one gets the impression that Samsung’s still working out some of the kinks on this one. Also, while it’s possible to get it to auto switch, the tablet screen always switches off when removed from the dock. A slight nuisance, but not the end of the world.

The biggest nuisance is, predictably, the same issue you run into on ChromeOS ever since Google allowed Play Store access. There simply aren’t that many apps optimized for the mode. You can access them through the Apps for Samsung Desk header located in the familiar Android apps menu. The company didn’t disclose the number of compatible apps for good reason. There are like 16. On the upside, the company teamed with Microsoft for Word/PowerPoint/Excel — all important inclusions when you’re pushing productivity on your shiny new device.

There are some other big names here, like Amazon WorksSpaces and The New York Times. There’s also Deezer, which seems to be eager to get on any operating system that will have them, bless their hearts. There’s also TripAdvisor and Craigslist, neither of which I would honestly put on my list of must-need desktop apps.

The main issue with the unoptimized apps is scalability — and the fact that they’re not designed to work well with the mouse input (which is, admittedly, optional). They do run, but opening them to full screen size requires restarting the app every time. One assumes, however, that more companies will get on board if the new Tab proves a hit for the company.

I’m not really convinced that DeX was the right choice over Android here, but at the very least, it gives the company room to grow on that side of things. It also offers an important differentiator over the iPad Pro, which has introduced some desktop-like functionality, but still relies on standard iOS. Apple’s been wary to blur the lines between desktop and laptop too much, relinquishing the upper hand to Samsung here.

Just in case

Here’s a bit of a shocking admission: I actually kind of like the keyboard case. I didn’t expect to. I don’t really ever like keyboard cases. They’re one of the bigger barriers between me and the possibility of ever taking convertibles too seriously as a potential laptop replacement. The keys on the S4 are plasticky, sure, but I prefer them to the standard keyboard case found on products like the Surface Go, which basically feels like typing on felt.

That said, there’s really no getting around how cramped the keyboard is, when designed for a 10-inch tablet. I considered pulling the Tab out for meetings multiple times this week, but ultimately thought better of it. I’m sure I’d be able to ramp up my speed given enough time with the system, but it ultimately wasn’t worth the potential of missing some important points, as my fingers struggled to keep up.

The lack of a trackpad feels like a glaring omission. Granted, it’s not exactly standard on keyboard cases these days, but Microsoft certainly demonstrated that it’s possible on a small scale with the Surface Go. And given the prominence of desktop mode here, one expects to be using the cursor as much as touch here. You can always use the S-Pen or buy a bluetooth mouse, but it seems a shame to have to add that cost on top of the keyboard case’s $149 asking price.

Bonus points, however, for including an S-Pen slot in the keyboard case for those of us who would lose our heads if they weren’t attached. The magnetic Microsoft pen that sticks to the side of the Surface is still the one to beat, but the holster does the trick, too.

All work, no play?

I remain skeptical as ever that convertibles will serve as a sufficient replacement for both laptop and tablet. I’m certainly not in a rush to put my MacBook out to pasture. That said, the devices provide a nice supplemental function for those who don’t want to port their heavier PCs around.

The $649 Tab S4 certainly has solid specs, with 4GB of RAM and up to 256GB of storage (augmented via microSD). The 7,300mAh battery can’t compete with the iPad Pro’s 8,134, but it should get you through a day’s use, no problem. The screen is solid for movie playback, and ditto for those AKG-tuned speakers.

I liked the keyboard more than I thought I would, as well, though it’s a bit cramped and using a keyboard without a trackpad feels like relying on a phantom limb. The DeX desktop, meanwhile, is an interesting addition to the tightening convertible tablet race, though it’s got a ways to go before it feels as fully fleshed out Windows 10.

Facebook is diving deeper into in-house game development with the launch of its own version of Snapchat’s multiplayer augmented reality video chat games. Today, Facebook Messenger globally launches its first two AR video chat games that you can play with up to six people.

“Don’t Smile” is like a staring contest that detects if you grin, and then users AR to contort your it’s an exaggerated Joker’s smirk while awarding your opponent the win. “Asteroids Attack” sees you move your face around the navigate a space ship, avoiding rocks and grabbing laser beam powerups. Soon, Facebook also plans to launch “Beach Bump” for passing an AR ball back and forth, and a “Kitten Craze” cat matching game. To play the games, you start a video chat, hit the Star button to open the filter menu, and then select one of the games. You can snap and share screenshots to your chat thread while you play.

The games are effectively a way to pass the time while you video chat, rather than something you’d ever play on your own. They could be a hit with parents and grandparents who are away and want to spend time with a kid…who isn’t exactly the best conversationalist.

Facebook tells me it built these games itself using the AR Studio tool it launched last year to let developers create their own AR face filters. When asked if game development would be available to everyone through AR studio, a spokesperson told me “Not today, but we’ve seen sucessful short-session AR games developed by the creator community and are always looking out for ways to bring the best AR content to the FB family of apps.”

For now, there will be no ads, sponsored branding, or in-app purchases in Messenger’s video chat games. But those all offer opportunities for Facebook and potentially outside developers to earn money. Facebook could easily show an ad interstitial between game rounds, let brands build games to promote movie releases or product launches, or let you buy powerups to beat friends or cosmetically upgrade your in-game face.

The games feel less polished than the launch titles for Snapchat’s Snappables gaming platform that launched in April. Snapchat focused on taking over your whole screen with augmented reality, transporting you into space or a disco dance hall. Facebook’s games merely overlay a few graphics on the world around you. But Facebook’s games are more purposefully designed for split-screen multiplayer. Snapchat is reportedly building its own third-party game development platform, but it seems Facebook wanted to get the drop on it.

The AR video chat games live separately from the Messenger Instant Games platform the company launched last year. These include arcade classics and new mobile titles that users can play by themselves and challenge friends over high-scores. Facebook now allows developers of Instant Games to monetize with in-app purchases and ads, foreshadowing what could come to AR video chat games.

Facebook has rarely developed its own games. It did build a few mini-games like an arcade pop-a-shot style basketball game and a soccer game to show off what the Messenger Instant Games platform could become. But typically it’s stuck to letting outside developers lead. Here, it may be trying to set examples of what developers should build before actually spawning a platform around video chat games.

Now with over 1.3 billion users, Facebook Messenger is seeking more ways to keep people engaged. Having already devoured many people’s one-on-one utility chats, it’s fun group chats, video calling, and gaming that could get people spending more time in the app.

It’s been a long and trip-filled wait but mixed reality headgear maker Magic Leap will finally, finally be shipping its first piece of hardware this summer.

We were still waiting on the price-tag — but it’s just been officially revealed: The developer-focused Magic Leap One ‘creator edition’ headset will set you back at least $2,295.

So a considerable chunk of change — albeit this bit of kit is not intended as a mass market consumer device (although Magic Leap’s founder frothed about it being “at the border of practical for everybody” in an interview with the Verge) but rather an AR headset for developers to create content that could excite future consumers.

A ‘Pro’ version of the kit — with an extra hub cable and some kind of rapid replacement service if the kit breaks — costs an additional $495, according to CNET. While certain (possibly necessary) extras such as prescription lenses also cost more. So it’s pushing towards 3x iPhone Xes at that point.

The augmented reality startup, which has raised at least $2.3 billion, according to Crunchbase, attracting a string of high profile investors including Google, Alibaba, Andreessen Horowitz and others, is only offering its first piece of reality bending eyewear to “creators in cities across the contiguous U.S.”.

Potential buyers are asked to input their zip code via its website to check if it will agree to take their money but it adds that “the list is growing daily”.

We tried the TC SF office zip and — unsurprisingly — got an affirmative of delivery there. But any folks in, for example, Hawaii wanting to spend big to space out are out of luck for now…

CNET reports that the headset is only available in six U.S. cities at this stage: Chicago, Los Angeles, Miami, New York, San Francisco (Bay Area), and Seattle — with Magic Leap saying that “many” more will be added in fall.

The company specifies it will “hand deliver” the package to buyers — and “personally get you set up”. So evidently it wants to try to make sure its first flush of expensive hardware doesn’t get sucked down the toilet of dashed developer expectations.

It describes the computing paradigm it’s seeking to shift, i.e. with the help of enthused developers and content creators, as “spatial computing” — but it really needs a whole crowd of technically and creatively minded people to step with it if it’s going to successfully deliver that.

Founders for Change, an initiative from All Raise, is hosting its first-ever meetup for its group of founders focused on diversity and inclusion in tech. Since launching in March, Founders for Change has grown to more than 800 founders from companies like Pandora, Patreon, Shippo, PagerDuty, Udacity and others.

“It’s one of those things where the movement really resonated with a number of founders,” Freestyle Capital Partner and AllRaise FfC co-lead Jenny Lefcourt told TechCrunch. “It was an online movement and everywhere for a period of time. People continued to sign up but it was just online. This is the beginning.”

That’s to say that this is likely not the last meetup for founders. The aim of this particular meetup is to share best practices, learnings and offer tools for founders to better implement diversity and inclusion at their respective startups.

“That’s the impetus for the event on Monday,” Lefcourt said. “It’s the first gathering of so many of these founders on this subject.”

Throughout the day, there will be breakouts and sessions about building belonging, how, why and when to build a diverse board, why it’s important to implement diversity and inclusion efforts early on and more. TK says she’s expecting about 100 founders to be in attendance.

“We were pretty amazed that this many founders are going to give such a big chunk of the day and evening to diversity and inclusion,” Lefcourt said. “It makes it clear how important they all recognize it is.”

All Raise, the nonprofit organization that’s behind diversity and inclusion efforts like Female Founder Office Hours (FFOH) and Founders for Change (FfC), wants to increase the amount of funding female founders receive. Currently, female founders receive just 15 percent of all venture funding, according to All Raise. Within the next five years, All Raise wants to increase that to 25 percent, while also doubling the percentage of women in VC partner roles in the next 10 years.

When Anki invited us by the office last week to check out their latest creation, we weren’t sure what to expect. In the eight years since the company was founded by a trio of Carnegie Mellon graduates, it’s offered up a handful of distinctly different smart toys, from its Drive smart cars to 2016’s Cozmo robot.

The truth is that the company’s latest creation looks an awful lot like its last. In fact, the only immediately apparent change is a dark coat of paint, akin to what the iMac received when it graduated to the iMac Pro. In fact, that’s probably a pretty solid analogy beyond just the color change, so let’s stick with that.

Simply put, Vector is Cozmo for adults. In many ways, the new ‘bot is built on the lessons learned from Cozmo, coupled with more advanced internals. Vector has ~700 parts — double the number of its predecessor, while its brain is a much more advanced Snapdragon processor. Vector’s face also features a higher res display, making it capable of expressing subtler emotion redesigned by that staff of ex-Pixar and Dreamworks animators the company employs.

So, what does all of this add up to, exactly? The company certainly has some grand ambitions. Anki believes its well positioned to offer users the gateway to the next generation of home robots. Both R2-D2 and Rosie from the Jetsons are casually name-checked in the company’s press release, naturally.

The truth about Vector is more modest, though the product does seem like a much more mainstream solution than the recently departed Kuri. For one thing, it will run less than half the price at $250. That’s still $70 more than Cozmo, mind. Unlike Cozmo, however, Vector doesn’t require a constant smartphone connection — just WiFi — which is part of why having a more advanced on-board computation system is important.

Along with that, Vector also gets an HD camera with a 120-degree field of view, so it can double as a roaming security device (that functionality is coming next year) and four microphones, which allow it to Alexa-style commands with a “Hey Vector.” Like Cozmo, the robot will also initiate conversation when you make eye contact, so you don’t have to summon it ever time. Instead of opting for Alexa or Google Assistant, however, the company used third-party knowledge graphs to build its own system, so Vector doesn’t break character by launching another assistant.

Like Cozmo, I suspect the company’s got a lot planned for the robot by way of software updates. For the time being, however, I haven’t seen a lot that convinces me that Vector’s much more than a slightly more advanced Cozmo — which could make the pricey a robot toy a hard sell for serious adults. Broader ambitions should include a robot that can traverse more than limited range of a desktop, in order to better patrol the home.

iRobot’s upcoming smart home plans for the Roomba serve as a potential way forward for home robotics, though Anki believes that it has the upper hand here, by investing a good deal in personality from the outset. Perhaps people really do need that human to robot connection to really entrust their lives to such a device, though a compete lack of personality certainly hasn’t hurt Amazon’s Echo or Google Home. 

But Anki gets the benefit of the doubt here. This is all very early stages, and our early skepticism about Cozmo being a niche product was put to rest when the company managed to sell 1.5 million globally. Launching as a Kickstarter campaign should also help Anki assess early interest in the product and scale from there.

I remain skeptical as ever that Vector is the mainstream home robot of the future — it seems more of a stepping stone, really. And certainly Anki is being transparent about its job of constantly building on the learnings of past models. At the very least, the company’s got the interest, revenue and the funding to avoid the recent pitfalls of Sphero and Kuri.

MessageBird, the Amsterdam-based cloud communications platform backed by Accel in the U.S. and Europe’s Atomico, is unveiling a new product today that aims to make it easier for enterprises to communicate with customers across various channels of their choosing.

Dubbed “Programmable Conversations” (yes, really!), the product takes the form of a single API that unifies customer interactions across multiple channels into a single conversation thread. Out of the box these include WhatsApp, WeChat, Facebook Messenger, Line, Telegram, SMS and voice interactions. The idea is that by providing a consolidated view of a customer’s entire communication history with an enterprise, customer support agents and other customer-facing staff will have the firepower to stay on top of their game in terms of the customer service they provide.

Or, put another way, more communication channels inevitably lead to fragmented conversations, which, especially when multiple support staff are involved, can lead to a degradation of service. Programmable Conversations is an attempt to help solve this problem.

In a call with MessageBird founder and CEO Robert Vis, he told me that more broadly enterprises — and fast-growing startups — no longer have the luxury of dictating how and through what channels customers converse with them. Traditionally, customer service would be delivered via a dedicated phone number, but the plethora of established and emerging online messaging and communications channels has radically increased the number of options customers have and expect.

However, this creates a headache for businesses as each channel needs developer time to be integrated into an existing CRM or business process and additional staffing to service conversations across multiple channels.

It is this heavy lifting that MessageBird’s Programmable Conversations takes care off — keeping conversations in sync across multiple channels, for example, isn’t technically simple — thus cutting down on not just initial implementation time and cost, but also continued maintenance and upkeep.

Vis also explained that Programmable Conversations is designed to enable comms for enterprises that are global — including scale-ups with global ambitions from the get-go — in terms of the territories, carrier integrations and messaging platforms the company supports.

“Delivering communications experiences that improve customer satisfaction and loyalty has to be a focus of businesses today,” adds the MessageBird CEO in a statement. “Consumers today want to connect with businesses in the same way they do with their friends and family – on their own time, via their preferred channel with all the context of previous conversations. With Programmable Conversations enterprises can now easily build a modern communications experience while reducing the burden of their often over-tasked developers”.

We caught wind pretty early on that Anki’s Cozmo was selling briskly when the adorable little robot temporarily sold out around Christmas 2016. Along with the announcement of the company’s latest robot, the startup released Cozmo sales figures for the first time, and things are, indeed, looking pretty good for the Wall-E-inspired ‘bot.

Turns out the company has sold more than 1.5 million Cozmos to date. Not too shabby for a $180 robot toy that admittedly launched with somewhat limited functionality (with the promise to add more in time) Anki also says that the device was the best-selling toy in its class for the whole of 2017, which seems to officially put to rest any concern that the pricey robot was going to be too much of a niche device to warrant serious consideration. 

It also gives the company solid footing to launch a robot like the pricier Vector, even as companies like Sphero and Kuri have stumbled. Along with all of that, the company also generated just under $100 million in revenue last year. Anki currently employees 203 and has raised north of $200 million, with a Series D arriving in 2016, no doubt spurred on by some of the early attention the company go when Apple featured its Drive cars at a keynote way back in 2013.

While some of the largest technology companies in the world are racing to figure out the next generation of machine learning-focused chips that will support devices — whether that’s data centers or edge devices — there’s a whole class of startups that are racing to get there first.

That includes Cerebras Systems, one of the startups that has raised a significant amount of capital, which is looking to continue targeting next-generation machine learning operations with the hiring of Dhiraj Mallick as its Vice President of Engineering and Business Development. Prior to joining Cerebras, Mallick served as the VP of architecture and CTO of Intel’s data center group. That group generated more than $5.5 billion in the second quarter this year, up from nearly $4.4 billion in the second quarter of 2017, and has generated more than $10 billion in revenue in the first half of this year. Prior to Intel, Mallick spent time at AMD and SeaMicro.

That latter part is going to be a big part of the puzzle, as Google looks to lock in customers in its cloud platform with tools like the Tensor Processing Unit, the third generation of which was announced at Google I/O earlier this year. Data centers are able to handle some of the heavy lifting when it comes to training the models that handle machine learning processes like image recognition as they don’t necessarily have to worry about space (or partly heat, in the case of the TPU running with liquid cooling) constraints. Google is betting on that with the TPU, optimizing its hardware for its TensorFlow machine learning framework and trying to build a whole developer ecosystem that it can lock into its hardware with that and its new edge-focused TPU for inference.

Cerebras Systems is one of a class of startups that want to figure out what the next generation of machine hardware looks like, and most of them have raised tens of millions of dollars. It’s one of the startups that has been working on its technology for a considerable amount of time. Others include Mythic, SambaNova, Graphcore, and more than a dozen others that are all looking at different pieces of the machine learning ecosystem. But the end goal for all of them is to capture part of the machine learning process — whether that’s inference on the device or training in a server somewhere — and optimize a piece of hardware for just that.

And while Google looks to lock in developers into its TensorFlow ecosystem with the TPU, that there are a number of different frameworks for machine learning may actually open the door for some startups like the ones mentioned above. There are frameworks like PyTorch and Caffe2, and having a kind of third-party piece of equipment that works across a number of different developer frameworks may end up being attractive to some companies. Nvidia has been one of the largest beneficiaries here of the emergence of GPUs as a go-to piece of hardware for machine learning, but these startups all bet on room for a new piece of hardware that’s even better at those specialized operations.

Fast-growing Chinese media startup ByteDance is looking to raise as much as $3 billion to continue growth for its empire of mobile-based entertainment apps, which include news aggregator Toutiao and video platform Tiktok.

The Beijing-based startup is in early-stage talks with investors to raise $2.5 billion to $3 billion, according to a source with knowledge of the plans. That investment round could value ByteDance as high as $75 billion, although the source stressed that the valuation is a target and it might not be reached.

It’s audacious, but if that lofty goal is reached then ByteDance would become the world’s highest-valued startup ahead of the likes of Didi Chuxing ($56 billion) and Uber ($62 billion). Only Ant Financial has raised at a higher valuation, but the company is an affiliate of Alibaba and therefore not your average ‘startup.’

The Wall Street Journal first broke news of the ByteDance investment plan.

But there’s more: Earlier this week, the Financial Times cited sources who indicate that ByteDance is keen to go public in Hong Kong with an IPO slated to happen next year.

ByteDance is best-known for Toutiao, its news aggregator app that claims 120 million daily users, while it also operates a short-video platform called Douyin. The latter is known as TikTok overseas and it counts 500 million active users. TikTok recently merged with Musical.ly, the app that’s popular in the U.S. and was acquired by ByteDance for $1 billion, in an effort aimed at combining both userbases to create an app with global popularity.

The firm also operates international versions of Toutiao, including TopBuzz and NewsRepublic while it is an investor in streaming app Live.me.

The company’s growth has been mercurial but it has also come with problems as the company entered China’s tech spotlight and became a truly mainstream service in China.

ByteDance had its knuckles wrapped by authorities at the beginning of the year after it was deemed to have inadequately policed content on its platform. Then in April, its ‘Neihan Duanzi’ joke app was shuttered following a government order while Toutiao was temporarily removed from app stores. It returns days later after the company had grown its content team to 10,000 staff and admitted that some content it had hosted “did not accord with core socialist values and was not a good guide for public opinion.”

Analysis of open source information carried out by the investigative website Bellingcat suggests drones that had been repurposed as flying bombs were indeed used in an attack on the president of Venezuela at the weekend.

The Venezuelan government claimed three days ago that an attempt had been made to assassination president Maduro using two drones loaded with explosives. The president had been giving a speech at the time which was being broadcast live on television when the incident occurred.

Initial video from a state-owned television network showed the reaction of Maduro, those around him and a parade of soldiers at the event to what appeared to be two blasts somewhere off camera. But the footage did not include shots of any drones or explosions.

News organization AP also reported that firefighters at scene had shed doubt on the drone attack claim — suggesting there had instead been a gas explosion in a nearby flat.

Since then more footage has emerged, including videos purporting to show a drone exploding and a drone tumbling alongside a building.

Bellingcat has carried out an analysis of publicly available information related to the attack, including syncing timings of the state broadcast of Maduro’s speech, and using frame-by-frame analysis combined with photos and satellite imagery of Caracus to try to pinpoint locations of additional footage that has emerged to determine whether the drone attack claim stands up.

The Venezuelan government has claimed the drones used were DJI Matrice 600s, each carrying approximately 1kg of C4 plastic explosive and, when detonated, capable of causing damage at a radius of around 50 meters.

DJI Matrice 600 drones are a commercial model, normally used for industrial work — with a U.S. price tag of around $5,000 apiece, suggesting the attack could have cost little over $10k to carry out — with 1kg of plastic explosive available commercially (for demolition purposes) at a cost of around $30.

Bellingcat says its analysis supports the government’s claim that the drone model used was a DJI Matrice 600, noting that the drones involved in the event each had six rotors. It also points to a photo of drone wreckage which appears to show the distinctive silver rotor tip of the model, although it also notes the drones appear to have had their legs removed.

Venezuela’s interior minister, Nestor Reverol, also claimed the government thwarted the attack using “special techniques and [radio] signal inhibitors”, which “disoriented” the drone that detonated closest to the presidential stand — a capability Bellingcat notes the Venezuelan security services are reported to have.

The second drone was said by Reverol to have “lost control” and crashed into a nearby building.

Bellingcat says it is possible to geolocate the video of the falling drone to the same location as the fire in the apartment that firefighters had claimed was caused by a gas canister explosion. It adds that images taken of this location during the fire show a hole in the wall of the apartment in the vicinity of where the drone would have crashed.

“It is a very likely possibility that the downed drone subsequently detonated, creating the hole in the wall of this apartment, igniting a fire, and causing the sound of the second explosion which can be heard in Video 2 [of the state TV broadcast of Maduro’s speech],” it further suggests.

Here’s its conclusion:

From the open sources of information available, it appears that an attack took place using two DBIEDs while Maduro was giving a speech. Both the drones appear visually similar to DJI Matrice 600s, with at least one displaying features that are consistent with this model. These drones appear to have been loaded with explosive and flown towards the parade.

The first drone detonated somewhere above or near the parade, the most likely cause of the casualties announced by the Venezuelan government and pictured on social media. The second drone crashed and exploded approximately 14 seconds later and 400 meters away from the stage, and is the most likely cause of the fire which the Venezuelan firefighters described.

It also considers the claim of attribution by a group on social media, calling itself “Soldados de Franelas” (aka ‘T-Shirt Soldiers’ — a reference to a technique used by protestors wrapping a t-shirt around their head to cover their face and protect their identity), suggesting it’s not clear from the group’s Twitter messages that they are “unequivocally claiming responsibility for the event”, owing to use of passive language, and to a claim that the drones were shot down by government snipers — which it says “does not appear to be supported by the open source information available”.