A Review of the Mobile World Congress in Four Chapters

1.) If we had one overriding “theme” this year it was about the synergy of 5G connectivity, IoT, and virtual reality/augmented reality. Even “Smart Cities”. All of which I will cover in a few video interviews with the brainiacs at Nokia who kindly let us spend almost one-half day with them behind-the-scenes and who put much of this in perspective. Some points:

2.) As most of us have read, mobile represents the highest scale consumer tech worldwide. With more than 5 billion unique mobile subscribers at the end of 2017, mobile has a greater reach than any other technology. Yes, growth is slowing as an increasing number of developed mobile markets are approaching saturation; as such, according to a detailed GSMA report, it will take more than eight years to move from 5 to 6 billion mobile subscribers compared to the four years it took to move from 4 to 5 billion.

What I found most interesting was that mobile subscribers differ from mobile connections such that a unique user can have more than one SIM card. The number of connections excluding cellular IoT totalled 7.8 billion globally in 2017 and will reach 9.0 billion by 2025. There will therefore be three SIM cards for every two subscribers (a SIM ownership ratio of 1.5). In developing countries, the SIM ownership ratio is often higher, especially in Sub-Saharan Africa (1.7), where many consumers use multiple SIMs from different mobile operators, often through dual-SIM handsets, to make use of the best network coverage and call quality in certain locations.

3.) There is a continuing transition from the connected consumer to thedigital consumer. While more than 3 billion people use mobile internet globally (internet-connected consumers), their digital engagement — measured by the GSMA Global Mobile Engagement Index as well as numerous other companies that do mobile metrics — varies significantly between countries. By all accounts, South Korea and the Scandinavian countries taken as a block (but especially Finland and Sweden) have a relatively high mobile engagement scores. The U.S. (surprisingly) places a distant 3rd place.

But the world-wide trends remain the same: subscribers are using their phones on a regular basis to access not only internet-based messaging and social media but also entertainment content (such as movies, music, games and sports), e-commerce and other digitally delivered services and content (i.e. financial services, health, education, government services). A deeper look at the segmentation groups of smartphone users by their mobile engagement pattern reveals that the transition from the connected consumer to the digital consumer (those who regularly consume digital services and content via their smartphones) has been fastest in some of the most developed and tech-advanced markets — again, South Korea, Australia, across Scandinavia, and the U.S.

4.) Oh, and very big this year: no self-respecting, big stand-owner/company would be complete without: a drone, or some VR glasses, or a car, or something with 5G … or an expert on how to tackle security. And sometimes 3 or 4 of those things. Cover those bases!

5.) Google announced new languages for its Assistant and Flutter which are supposed to make our life easier. This touches upon the whole issue (too much for this post) on how a bunch of competitors, big and small, have come out with their own replacement Android frameworks. It’s like everyone who’s ever tried to do Android programming gave up and declared: “This is so bad that I will do my own startup to make it better.” And Google, not to be outdone by their competitors, responded by saying, “Oh yeah? Well you can’t compete with us, because we’re going to compete with us!” and they launched Flutter.

Android has been trying to hire mobile developers, which you’d think would be a straightforward task. But it turns out they’re the hottest commodity on the market right now. Everyone needs them, and there aren’t enough to go around. As Katherine Casey, a mobile dev friend, said: “It’s like trying to catch unicorns”. Why does everyone need mobile devs? Because the web is in a steady decline as the whole world moves to mobile. You probably remember Facebook going through its transition from web-first to mobile-first. Facebook went through an existential crisis when they realized that they had to become a mobile company or face oblivion. More on all of this in a subsequent post.

6.) Huawei announced the first 5G Chip that can do 2,3 Gbp.

7.)  Samsung unveiled its latest flagship smartphone, the Galaxy S9. Samsung sat out last year’s MWC, as it was still smarting from the embarrassment of its Note 7 phones unfortunate tendency to explode. Bummer for the marketing department, for sure. But based on reports this week the phone is not selling well. Beautiful to look at but … once again … massive software issues.

However, I must address all the chatter I heard which pretty much ran “if only Samsung just get out of its own way, built the hardware and let Google handle the software” because it misses the point:

8.) Obviously there was loads and loads more handset madness at MWC. I do not really cover that. I just ste al notes from my compadres at The Verge and CNET 🙂   Want more handset info? Just Google it.

9.) IBM unveiled a new movie with seven different endings (you make plot decisions as the movie progresses) which shows how  business apps combine the power of enterprise data, analytics, and artificial intelligence with user experience. It is rather cool. I devised three different endings. IBM’s intent is to redefine how employees interact, learn, connect, and perform. It  was directed by Brad Turner (the award winning director of the TV series  “24” ) and I had an opportunity to interview Brad and the IBM team and the Ogilvy & Mather advertising team that produced the film. Ogilvy & Mather are old chums from the Cannes Lions advertising conference.

10.) We heard this (in various forms) all over the event:  “In two years, a third of computing will be screenless, and two thirds of that will be processed by Alexa”. Well, a few points:

11.) Smart Cities. Every year MWC devotes more and more space to their “Smart City” area. This year time commitments permitted only a cursory stop.

But I did do a breakfast with Jackson Haywoord who advocates data-driven city building, especially as it pertains to active transportation. His point is that it’s hard to argue with numbers. Which is why he’s chosen to employ neuroscientific methods to examine the psychological impacts of urban living: “It’s extremely important that we equip our decision makers with the data and research they require to build healthy and safe cities. In the age of fake news, we’re in desperate need of evidence-based decision making”.

A few points we agreed on:

12.) Although officially released the week after MWC, the European Union team in Barcelona (principally from the unit on the Digital Economy and Society) discussed a report on how they are beefing up efforts to foster work around new technologies like blockchain.

The new initiative will see participants from EU-level and national authorities, as well as technology providers in Europe, seek to raise the “regulatory and supervisory capacity and knowledge” around new technologies, including blockchain. One rep I spoke with said “technologies like blockchain can be game changers for financial services and beyond. We need to build an enabling framework to let innovation flourish, while managing risks and protecting consumers.”

I scanned the official release/report this weekend and it identifies areas of possible development, including legal clarifications around smart contracts, initial coin offerings, and other jurisdictional issues raised by the technology. The plan seeks to foster “work around blockchain within in a regulated environment”.

And one bone to pick on blockchain. At the event several vendors (obviously not up-to-speed and/or provided poor speaking notes) kept saying blockchain represents something “akin to a new universal protocol, like TCP-IP or HTML were for the Internet”. 

On its face, such claims imply that this or that blockchain will serve as the basis for most of the world’s transactions and communications in the future. WRONG. This makes little sense when one considers how blockchains actually work. For one thing, blockchains themselves rely on protocols like TCP-IP, so it isn’t clear how they would ever serve as a replacement.

Furthermore, unlike base-level protocols, blockchains are “stateful” which means they store every valid communication that has ever been sent to them. As a result, well-designed blockchains need to consider the limitations of their users’ hardware and guard against spamming. This explains why Bitcoin Core, the Bitcoin software client, processes only 5-7 transactions per second, compared to Visa, which reliably processes 25,000 transactions per second.

And Bitcoin and blockchains are not a technology story, but a psychology story.

For more on blockchain at MWC, see my entry below on one company, fluency.ai in my “cool tech” section.

13.)  This had to be one of the funniest signs I saw at MWC:

I met with Deloitte Mobile and they told me that their studies show 10 percent of North Americans over age 18 will be engaged in four or more multiple, simultaneous advertisement-blocking behaviors in 2018; they call these people “adlergic”. There were multiple companies at MWC offering some form of ad blocking (such as software on computers and mobile devices, and streaming music and video services) and they are growing relatively quickly, while other forms of ad blocking (such as ad skipping with personal video recorders or changing channels on TV or radio) are stable or growing slowly.

But it appears that almost nobody blocks all ads. I am not referring to categories that are inherently impossible to block (for example, a highway billboard), but instead to the fact that according to Deloitte across the seven major ways of blocking ads, the percentage who block all seven was zero or nearly zero in all countries surveyed in 2017, and they predict that will be true again in 2018.

1.) The mobile industry is reducing the volume/need for large training data sets

Training an machine learning model can require up to millions of data elements. This can be a major barrier. Acquiring and labeling data to be used for training can be highly time-consuming and costly.

NOTE: numerous vendors and the GSMA had tutorial sessions on data science, all aimed for the layman. These sessions addressed algorithms, “black boxes”, data exploration, data wrangling, feature engineering, machine learning, neural networks, training data, etc.

As Jackie Hogan at GE Moble explained it, consider, as an example, a project that requires MRI images to be labeled with a diagnosis. It might cost over $30,000 to hire a radiologist to review and label 1,000 images at a rate of six images an hour. Privacy and confidentiality concerns can also make it difficult to obtain the data in the first place.

But a number of promising techniques are emerging that aim to reduce the amount of training data required for ML. One involves the use of synthetic data, generated algorithmically to mimic the characteristics of the real data. Deloitte Mobile tested a tool that was able to build an accurate model with only  a  fifth of the training data previously required; it synthesized the remaining 80 percent of data.

Synthetic training data can also open the door to the crowdsourcing of data science solutions. A number of organizations have engaged third parties to devise ML problem – solving models and are posting data sets appropriate for sharing that outside data scientists can work with. Researchers at MIT Media Lab used a real data set to create synthetic alternatives that could be used to crowdsource the development of predictive models without needing to disclose the original data set. In 11 out of  15 tests, the models developed from the synthetic data vault performed as well as those trained on real data.

Another technique that could reduce the need for training data is transfer learning. Deloitte Mobile: “With this approach, an ML model is pre-trained on one data set as a shortcut to learning a new data set in a similar domain, such as language translation or image recognition. Some ML tool vendors claim their use of transfer learning can cut the number of training examples customers need to provide by several orders of magnitude.”

Obviously, a big area that needs some depth, but a quick fire round:

The next generation of machine-learning chips, established and start-up hardware manufacturers are developing specialized hardware (such as GPUs, FPGAs and ASICs) to slash the time required to train ML models, by accelerating the calculations required and the transfer of data within the chip. These dedicated processors can help companies speed up ML training and execution manyfold, which in turn brings down the associated costs.

Microsoft research team using GPUs completed a system in one year to recognize certain conversational speech as capably as humans could. With CPUs, it would have taken five years.

Google stated that designing its own AI chip, a TPU, for neural networks execution and adding TPUs to CPU and GPU architecture helped the company save the cost of building a dozen extra data centers.

Early adopters of these specialized AI chips include major technology vendors and research institutions in data science and ML, but adoption is spreading to sectors such as retail, financial services and telecom. Accelerated training will become mainstream across all industries employing ML and multiplying the number of applications enterprises choose to undertake.

Last note: ML achievements get more impressive by the day. But ML models often suffer from a critical  aw: many are black boxes, meaning it is not possible to explain with confidence how they make their decisions. This makes them unsuitable or unpalatable for many applications, for reasons ranging from trust in the answers generated by a model, most obviously in regulatory compliance. But a  number of techniques have been created that help shine light into the black box of certain ML models, making them more interpretable and accurate. Folks from MIT Media Lab told me about a method of training a neural network that delivered accurate predictions and the rationales for those predictions.

2.) An AI breakthrough: transcribe your meetings in real time

The company is called AISense and the app is called Otter.ai. And they had a prominent place in Hall 2 (which also has the best end-of-the-confernce-day parties, by the way). They have taken a different approach to understanding audio data than Amazon Alexa, Google Assistant, and the other companies working on speech recognition. In the demo pitch they had set up our meeting in the app beforehand, so the software automatically recognized my teammate who could chime in with her own comments or as I interrupted with follow-up questions as we conducted a 3-person interview chat.

While Otter’s natural language processing wasn’t perfect by any means – punctuation is missing, words are misunderstood, speakers are sometimes misidentified – but it’s remarkably close, especially considering its speed and the fact that the app is free. They call their technology “Ambient Voice Intelligence” and they use the word ambient to indicate that this is working in the background. Given your brain can only remember 10-20% of the information from a meeting (there are scores of studies oaths) the aim is to help people capture that information and then search for it really fast. I had it uploaded to my Mac and one of my staffers was cleaning it up within 30 minutes of the meeting ending.

The search is the best feature. Once the recording is finished, the app’s machine learning automatically creates about 10 keywords so that you know what the meeting was about. And you can start searching the full text right away. Also useful is that once you hone in on a keyword, you can hit the play button to listen to the section of the audio where it occurred. And the next best feature of the app is that you can share recorded meetings. So, if you have a meeting and a colleague can’t attend, you can send them the transcript and audio afterward, so that they can find the stuff that’s relevant to them.

This is the holy grail for journalists who don’t want to do tedious, tedious transcription of important (and unimportant) interviews. Search in particular is really big. It’s on the App Store.

But even cooler ….

I try to average 2-3 book reads a quarter and I usually read those 2-3 books at the same, picking books that relate/overlap.  For example, over the summer break last year, I read (1)  World Without Mind, (2)  Deep Thinking: Where Machine Intelligence Ends and Human Creativity Begins, and (3)  A Mind at Play: How Claude Shannon Invented the Information Age. But while it’s important to read books, it is just as important to remember what you read and put it to good use. So I create searchable book notes which is essential for returning to ideas easily. I use Evernote only because it is searchable, easy to use across multiple devices, and you can create and save notes even when you’re not connected to the internet.

But Evernote does not convert audio recordings into text, nor does it allow you to search for a word mentioned inside the recording. So, now with AISense I am going to create my own system with Bear (another cloud-based note-taking app I actually prefer to Evernote) and create my own “voice intelligence” system.

3.) Along a similar vein, Synthesis — translate hard copy in real time

Google’s Translate product really started this back in 2015 but many companies are now in the stream to provide an instant translation feature for hard copy. One is  Synthesis — a start up — which lets you aim your phone camera at something written in a language you do not understand, say a sign, and it translates into your language with a (ridiculous) accuracy in almost real-time.

I think Google Translate is now up to 58 languages and Synthesis is 27 languages. The secret? Those convolutional neural networks popping up in almost every article in  Wired and the  MIT Technology Review.

4.) The BBC augmented reality app

Last week the BBC launched its new Civilisations series — it’s presented by Simon Schama, Mary Beard and David Olusoga and spans 31 countries and covers 5,000 works of art across human history — and they did so with a mobile phone app. The BBC was at MWC and the rep said it was BBC’s attempt to bring “art and culture from across the world to you with a BBC Augmented Reality app”. It brings together museums and galleries from across the UK through digital products and events, and the augmented reality app allows users to bring the treasures of these collections into their own homes. They gave us a screen shot:

Put very simply, AR, or augmented reality overlays virtual objects in the world around us that can be seen through a smartphone, tablet or headsets.  For some time the BBC had been talking to the UK’s culture sector about making the art and culture of the UK more accessible. For example, what if you can’t get to one of the big museums in London? Or, how can everyone see the hidden gems from the more remote locations? So … BANG! AR

5.)   Samsung’s new TVs are almost invisible

The TV was officially launched after MWC, but we caught a discussion about it at MWC.

Briefly: Samsung’s new QLED line of 4K TVs features a technology the company is calling “Ambient Mode”. Before you mount the TV, you’ll snap a picture of the wall it’s going to hang on – it doesn’t matter if it’s brick, wood, patterned wallpaper, or just a white wall – and then after it’s up, you can set that picture as the TV’s background:

The result is something that looks like a floating black rectangle mounted on a wall. Samsung even includes a digital version of the shadow this black rectangle would cast on the wall, as if there really wasn’t a large LED panel sitting in the middle of the thin metal strips.

There are five QLED models, with minor tweaks between them, ranging in size from 49 inches, up to an absolutely massive 88 inches.

Oh, important point: the televisions have a built-in timer so that the ambient setting will turn off after a while, in order to spare your electricity bill.

6.) How about vaccinations for cyberspace? A good security cure?

Vaccinations can prevent the spread of infectious disease. So maybe the same is true for cyberspace. The head of Head of Security for Telekom, Dirk Backofen, did a masterful presentation on the When WannaCry ransomware attack that hit in May 2017. No other previous attack had as great an impact. The truly insidious thing about this encryption Trojan? It spread worm-like from computer to computer, virtually eating its way through IT systems worldwide. As he noted “WannaCry exposed the global vulnerability of the IT infrastructure in a way never done before. We are years and years and years away from effective resilience against such attacks”.

Or to put it another way: globally connected IT structures are only as strong and robust as their weakest links.

He suggested a “broad-spectrum vaccine” against cyber attacks. There will still be successful attacks, of course, but their numbers will be reduced by better “vaccination coverage” according to him. He provided a very detailed analysis but let me just red-line his 8 points that make up the “vaccination” : (1) Security by Design; (2) Compulsory vulnerability scans; (3) APT protection for every business mailbox and the web; (4) DDoS protection in the terabit range; (5) Mobile security for companies; (6) Compulsory (managed) cyber defense for enterprise networks; (7) Treat industrial networks like critical infrastructures; and (8) Improve reaction times.

7.) To understand 5G systems you need to understand network slicing

Understanding telecommunications infrastructure … the organizations, personnel, procedures, facilities and networks employed to transmit and receive information by electrical or electronic means … is difficult stuff but there is no better place to learn it than at MWC. And with the advent of 5G there is more to learn. 5G systems are to be built in a way to enable logical network slices, which will allow telecom operators to provide networks on an as-a-service basis and meet the wide range of use cases that the 2020 timeframe is expected to demand. And so “network slicing” is the thing to learn.

Why? Because in a single 5G system, network slicing technology can provide connectivity for smart meters with a network slice that connects “internet of things” devices with a high availability and high reliability data-only service, with a given latency, data rate and security level. At the same time, the technology can provide another network slice with very high throughput, high data speeds and low latency for an augmented reality service. Different services require different latency, different data speeds so you need to “slice” the network accordingly. So with network slicing technology, a single physical network can be partitioned into multiple virtual networks allowing the operator to offer optimal support for different types of services for different types of customer segments.

There were several companies hawking their expertise in this area but the two that impressed me were Nokia Mobile and Cloudstreet, both of whom I video interviewed.

8.) Valossa … artificial intelligence that understands video

Yes, I pulled that title right from the Valossa website. But they have an incredible  full stack AI solution  designed for anybody working with video. It’s  application to the e-discovery market is glaringly obvious, and they have a technology far and away better than anything I have seen from any legal technology vendor. We are discussing a beta-test for an enormous video e-discovery project in Europe. We have a video interview with Mika Rautiainen, the company CEO, in process.

I had a treat discussing how you find that movie where Sean Connery wears red trousers. Or romantic comedies set in Hawaii. Valossa can do all that but it really does not come close to the incredible, sophisticated technology they have developed. Mika has the deepest experience I have ever seen in computer vision and artificial intelligence for media content recognition systems and his research team’s innovations in AI have been integrated into Valossa’s content analysis engine, as well as the company’s emotion and expression recognition engine. Based on the latest deep learning techniques, the Valossa AI intelligence engine recognizes people, visual context, speech topics, video categories and other video entities at an unprecedented level of detail, from scene to higher level understanding. More to come.

9.) Mobile monetization/economics 101

MWC is overloaded with vendors claiming their expertise at marketing ROI. But many tend to go a bit “Hugh Grant” when they’re confronted about return on investment (ROI), preferring to lean back on easy stats — vanity metrics — to prove their effectiveness: “Look at the number of website hits,” they’ll say. “You got followed by [INSERT INDUSTRY BOD HERE] on Twitter.” These are called vanity metrics for a reason. They look good on the surface but underneath, they’re empty. The reason marketing charlatans shy away from discussing ROI is that they can’t prove their own. It’s also the reason why good marketing directors run towards the issue with open arms. ROI can prove beyond doubt how effective their marketing efforts have been in a business.

The folks we spoke with at Accenture and at Adcash know their stuff. We’ll have a video with adcash in a few weeks that will take you through all of this point by point but in brief ….

These conversations were a “Masters Class” on how to calculate and demonstrate ROI, the breakdown of ROI on detailed marketing dashboards, etc.  Literally how to know if your budget is being poured into a leaking bucket. Granted, ROI can be complex, but it’s not the mystery some marketers like to make out.

Accenture took us through cost per acquisition (CPA) which is calculated by taking the total marketing budget and dividing it by the number of new customers acquired. A simple equation, but one that should determine every marketing decision you make. It should be applied to every channel and marketing campaign you work with and is crucial for long-term marketing planning. They also took us through customer LTV which is the abbreviation for lifetime value. LTV is the revenue you can expect to achieve from a customer in the time they’re with you. Example they gave: if your customer lifecycle is 18 months and the average customer spend is €1,000 per month, this leaves you with a lifetime value of €18,000. Simple. But also crucial when ascertaining marketing budgets and spend.

Adcash told us an existing marketing consultant should have devised a solid, documented marketing plan, and be working off it on a weekly basis. They should know where you are getting the most return. They should know what marketing channel or channels fulfil this criterion, and be prepared to demonstrate results on their marketing dashboard. Not “hey, we’ll use VR or social media”. Make them squirm.

10.) Evolution of the smartphone camera

One of my staffers, Silvia Di Prospero, spent one-half day just looking at smartphone cameras. She is a semi-pro photographer. She had the opportunity to test numerous sophisticated dual-lens smartphone cameras at LG, Samsung, Sony, etc. A few of her points:

One of the primary differentiators for all smartphones is the camera or, more precisely, the image displayed on a phone’s screen once an array of post-processing actions – often using proprietary hardware and software – have taken place. Improvements in photo quality over the next few years are likely, thanks to better, highly integrated software and hardware, with the role of software becoming increasingly significant. This should enable smartphones to get closer to the quality and characteristics of images captured by much larger, heavier and bulkier traditional cameras.

One type of photographic effect that had been unique to traditional SLRs was “bokeh,” where the subject is in focus and the background is blurred, highlighting the subject more prominently. On a standard smartphone, the entire image ordinarily would be in focus. The bokeh function in smartphones was first introduced in 2016, but it has since advanced markedly, and as of late 2017, most vendors’ flagship models supported this effect.

The first phones with a bokeh capability blend images from two lenses taking a photo at the same time. Software is then used to create a composite image, with the major challenge being to distinguish and separate precisely the subject from the background. Over time, the quality of the bokeh effect should steadily improve; ML enables more refined algorithms that become more accurate at splitting foreground from background. And more powerful and dedicated processors should enable these composite images to be generated ever faster.

Another innovation that is likely to become ubiquitous in smartphones over the next few years is optical image stabilization, which integrates hardware (the lens), sensors (the gyroscope) and processors to mitigate the impact of the device shaking when a photo is being taken. If the gyroscope detects the phone is moving (perhaps due to the press of a finger on a shutter), it adjusts the direction of the lens slightly to counteract the device’s movement. The result is no or reduced blur in a photo.

And, as she learned at the LG electronics stand, the smartphone’s camera is increasingly used in work contexts, again with software at the fore. A common administrative task in a work context is capturing information, which may be in the form of business cards, receipts or brochures, or key documents. Any camera can copy these images; software can remove excess content (such as the underlying table surface when photographing a business card) or shadows. One type of camera being integrated into some handsets used in a work environment is a back-facing IR camera, which detects heat (IR energy). A camera with integrated IR would be compact and portable, and could readily be carried by tradespeople, including in confined spaces or up ladders. Thermal imaging is being integrated into all iPhones and Samsung phones to be used in a wide range of contexts, such as tracing the source of leaks in a home, analyzing faulty motor engines or checking home appliances.

11.) Voxos – your smartphone, glasses & headphones in one device

Voxos is all about bone conduction technology. It was one of the start-ups on display (they have a Kickstarter launching soon) and they got a lot of buzz and their tiny booth was jam packed with attendees.

I had the opportunity to wear the glasses for a few hours, and we did a video interview with  Maja Köber, the company’s media director. It is a combination of headphones and glasses into one single device that enables users to play sound through the bones in their skull with a patented technology.

After I got home, I spoke with a doctor at the  French National Institute of Health and  Medical Research and he was impressed with the technology, and the intent.

I will have more when we publish the post and video. But in brief, here is how  sound through your skull works, instead of through your ears:

The integrated bone conduction technology transfers the sound to the head by vibrating the cheekbones. This enables users to access anything from music, GPS, phone calls to other information and data application on their device via Bluetooth, without ever losing track of their surroundings. In other words, Voxos smart glasses will enable users to operate different functions while keeping their eyes on the road.

As I said, they will go live on Kickstarter very soon and I am getting a free pair to try out for a longer period of time so I will have more.

12.) The Finish company Quuppa … the leading company for indoor positioning

As I have noted before, indoor positioning systems (IPS) are systems to locate objects or people inside a building using radio waves, magnetic fields, acoustic signals, beacons, or other sensory information … mostly collected by various mobile devices. 

This is incredibly important for the industrial Internet of Things. We as consumers get fed up with the consumer IoT side such as the growing inaccuracy of Google Maps and Apple Maps … and by the way, it is getting worse as many of us have surmised, and as detailed at several sessions here … but at the industrial level such inaccuracy is anathema.

The Finish company Quuppa is becoming the market leader in this field and they were one of the first videos we produced during the Congess. Here is that interview:

13.) Herta … a master at facial recognition technology

We all know the cameras are out there. In banks and government buildings. In trains stations and outside warehouses. In stadiums. At concert halls. On almost every street. They’re capturing images of people’s faces to help prevent nefarious activity. It is now a reality as facial recognition technology is considered crucial for video surveillance which is being deployed globally as a part of security strategy.

Headquartered in Barcelona but with global operations, Herta has become a dominant player. They provide the facial recognition technology for the Golden Globe awards show, and work with different police forces all over the world, as well as casinos in different countries. They have developed a unique “Bio Surveillance” lines of products, specially designed for security purposes and optimized for environments with large crowds.

I did a video interview with Javier Rodriguez, CEO of the company. He noted their current software is “faster, more accurate and able to compare 1 to 35 million faces in less than one second, working from a resolution as low as 24 pixels between eyes”. It can quickly identify VIP customers and exclude people that are blacklisted, and works as well with people on the move and partially hidden features. Another product line is geared towards marketing analytics and focuses on obtaining features such as age, gender and ethnicity.

Oh, and they have a cool technology for conferences. Walk up to the registration counter next to the designated bar, get your face recognized and place your drink order. Then, at your leisure, you can approach the bar. Your face will get identified on monitors behind the bar. Tipped off, the barkeep greets you by name and serves up your favorite drink.

Facial recognition technology (sort of) entered the main stream in 2010 when Facebook began to use it to help users tag photos, followed by Google which launched a photos app that helped users organize their pictures by automatically identifying family members and friends. Javier and I had a riveting discussion on the technology and where it is today which will be part of the video interview.

14.) Fujitsu Laboratories develops the first wearable, hands-free speech translation device

This was a technology developed for a specific purpose. Hospital workers often have their hands full with tasks such as patient care, plus they also often have to communicate with patients who speak other languages. With this in mind, Fujitsu Laboratories developed what it claims is “the world’s first wearable, hands-free speech translation device.” I certainly did not see anything like it at MWC, and my research after MWC did not turn up any similar devise. And a bit of deja vu. Last year I noted Fujitsu had developed a system in which stationary tablets with external microphones could identify different speakers, and translate their spoken language into another. Now, that technology has been shrunk down to a Wi-Fi-connected device that’s worn by the worker, like a name tag.

It has two integrated omni-directional microphones – one of them faces forward, to pick up the voice of the patient, while another faces up, to pick up the voice of the wearer. Onboard software automatically detects which of the two people is speaking, and a cloud-based server translates their words to synthesized speech in an output language of the user’s choice. This means, for instance, that before a conversation started, the user could stipulate that their own speech be translated to English, and that the patient’s speech be translated to Japanese.

Additionally, the device is claimed to be very good at filtering out background noise, and did so in the demo. According to the company hand-out, the technology has “achieved a speech detection accuracy of 95 percent in an environment with comparable noise levels to an examination room in a large hospital (about 60 decibels of noise) at a natural distance for a face-to-face conversation between a healthcare provider and a patient of about 80 cm [2.6 ft].”

More interesting to me was another Fujitsu product, a semantic-based machine translation system which aims at high quality simultaneous multilingual translation. The product is in beta and I was not allowed to take photos. But what they had done is set up 3 large screens … I could pick 3 non-English languages from a list of 27 languages to appear on those screens … and as I spoke English the three screens showed simultaneous translation in those languages. The system is based on “training” the database to learn certain words and phrases if you have a complex subject with a specialised language. It can also be trained to learn your vocal patterns.

I took turns with Matt Winer, an e-discovery attorney at the event, and we spoke on various legal subjects and the text flashed upon the screen in French, German and Spanish. Matt (a German speaker) switched to German and he said the system did an admirable job. For a system that was not trained, I would say it was 85-90% accurate. I could see this type of system being used at any number of conferences where you have presenters with various language fluency, more comfortable speaking in their native language but wishing to be understood without the use of translators.

15.) The shadowy world of encrypted phones

As I have noted in past coverage of MWC, the event is certainly the place to learn about mobile forensics. For two years running I have popped into one of the Intercede tutorials/labs which runs through almost everything relating to recovery of digital evidence or data from a mobile device. This usually refers to mobile phones; however, the programs also relate to any digital device that has both internal memory and communication ability, including PDA devices, GPS devices and tablet computers. Cellebrite, MSAB and Oxygen Forensics are pretty much the global leaders in this area.

Interesting this year was the larger-than-normal law enforcement contingent and I had an opportunity to chat with a few of them about the shadowy world of encrypted phones. There is a slew of companies that do this type of work (no, none of them had a stand). They build custom BlackBerry or Android devices (not Apple; too tough) that usually have the camera and microphone removed and have specialized secure messaging systems through private networks. These companies also typically provide server infrastructure to route the encrypted emails. These firms cater primarily to criminal organizations and … how shall I put this? … “non-state actors”. You may have read that this week the U.S. FBI and Interpol were able to shut down one such operation. Cellebrite seems to be the vendor-of-choice in cracking these phones.

16.) Web tools used to track Russian “influence ops” can also be used for e-discovery

Web tools used to track Russian “influence ops” can also be used for e-discovery.  As several vendors told me, the social media monitoring/analytical software used to track Russian ops (and that also do commercial analysis) is seeping more and more into the e-discovery world. Products like XI Social Discovery, Geofeedia, Dataminr, Dunami, and SocioSpyder (to name just a few) have been used by law enforcement, federal agencies, defense contractors, and the military. XI Social Discovery and Dataminr are already in the e-discovery space. Others are joining. The folks at Semantic Vision showed me this:

That is just one Russian troll amplified to show each and every social media node she or he hit on Facebook, LinkedIn, Reddit, and Twitter. The metrics and technology was incredible.

Many e-discovery technology providers have systems that can aggregate comprehensive social media content that can collect data from multiple social media accounts. Of course they must do so in a legally defensible manner and preserve the chain of custody. These systems save you vast amounts of time and money through the automated – and simultaneous – collection of data from multiple social media accounts.

But many times the “evidentiary value” is not critical. A law firm or corporation simply wants the “down and dirty” an employee’s misconduct, constructing a event timeline, tagging principal parties, or just helping the investigator explain what happened to upper management, to legal, and to law enforcement agencies.