mobile app
Auto Added by WPeMatico
Typewise taps $1M to build an offline next word prediction engine
Swiss keyboard startup Typewise has bagged a $1 million seed round to build out a typo-busting, ‘privacy-safe’ next word prediction engine designed to run entirely offline. No cloud connectivity, no data mining risk is the basic idea.
They also intend the tech to work on text inputs made on any device, be it a smartphone or desktop, a wearable, VR — or something weirder that Elon Musk might want to plug into your brain in future.
For now they’ve got a smartphone keyboard app that’s had around 250,000 downloads — with some 65,000 active users at this point.
The seed funding breaks down into $700K from more than a dozen local business angels; and $340K via the Swiss government through a mechanism (called “Innosuisse projects“), akin to a research grant, which is paying for the startup to employ machine learning experts at Zurich’s ETH research university to build out the core AI.
The team soft launched a smartphone keyboard app late last year, which includes some additional tweaks (such as an optional honeycomb layout they tout as more efficient; and the ability to edit next word predictions so the keyboard quickly groks your slang) to get users to start feeding in data to build out their AI.
Their main focus is on developing an offline next word prediction engine which could be licensed for use anywhere users are texting, not just on a mobile device.
“The goal is to develop a world-leading text prediction engine that runs completely on-device,” says co-founder David Eberle. “The smartphone keyboard really is a first use case. It’s great to test and develop our algorithms in a real-life setting with tens of thousands of users. The larger play is to bring word/sentence completion to any application that involves text entry, on mobiles or desktop (or in future also wearables/VR/Brain-Computer Interfaces).
“Currently it’s pretty much only Google working on this (see Gmail’s auto completion feature). Applications such as Microsoft Teams, Slack, Telegram, or even SAP, Oracle, Salesforce would want such productivity increase – and at that level privacy/data security matters a lot. Ultimately we envision that every “human-machine interface” is, at least on the text-input level, powered by Typewise.”
You’d be forgiven for thinking all this sounds a bit retro, given the earlier boom in smartphone AI keyboards — such as SwiftKey (now owned by Microsoft).
The founders have also pushed specific elements of their current keyboard app — such as the distinctive honeycomb layout — before, going down a crowdfunding route back in 2015, when they were calling the concept Wrio. But they reckon it’s now time to go all in — hence relaunching the business as Typewise and shooting to build a licensing business for offline next word prediction.
“We’ll use the funds to develop advanced text predictions… first launching it in the keyboard app and then bringing it to the desktop to start building partnerships with relevant software vendors,” says Eberle, noting they’re working on various enhancements to the keyboard app and also plan to spend on marketing to try to hit 1M active users next year.
“We have more ‘innovative stuff’ [incoming] on the UX side as well, e.g. interacting with auto correction (so the user can easily intervene when it does something wrong — in many countries users just turn it off on all keyboards because it gets annoying), gamifying the general typing experience (big opportunity for kids/teenagers, also making them more aware of what and how they type), etc.”
The competitive landscape around smartphone keyboard tech, largely dominated by tech giants, has left room for indie plays, is the thinking. Nor is Typewise the only startup thinking that way (Fleksy has similar ambitions, for one). However gaining traction vs such giants — and over long established typing methods — is the tricky bit.
Android maker Google has ploughed resource into its Gboard AI keyboard — larding it with features. While, on iOS, Apple’s interface for switching to a third party keyboard is infamously frustrating and finicky; the opposite of a seamless experience. Plus the native keyboard offers next word prediction baked in — and Apple has plenty of privacy credit. So why would a user bother switching is the problem there.
Competing for smartphone users’ fingers as an indie certainly isn’t easy. Alternative keyboard layouts and input mechanism are always a very tough sell as they disrupt people’s muscle memory and hit mobile users hard in their comfort and productivity zone. Unless the user is patient and/or stubborn enough to stick with a frustratingly different experience they’ll soon ditch for the keyboard devil they know. (‘Qwerty’ is an ancient typewriter layout turned typing habit we English speakers just can’t kick.)
Given all that, Typewise’s retooled focus on offline next word prediction to do white label b2b licensing makes more sense — assuming they can pull off the core tech.
And, again, they’re competing at a data disadvantage on that front vs more established tech giant keyboard players, even as they argue that’s also a market opportunity.
“Google and Microsoft (thanks to the acquisition of SwiftKey) have a solid technology in place and have started to offer text predictions outside of the keyboard; many of their competitors, however, will want to embed a proprietary (difficult to build) or independent technology, especially if their value proposition is focused on privacy/confidentiality,” Eberle argues.
“Would Telegram want to use Google’s text predictions? Would SAP want that their clients’ data goes through Microsoft’s prediction algorithms? That’s where we see our right to win: world-class text predictions that run on-device (privacy) and are made in Switzerland (independent environment, no security back doors, etc).”
Early impressions of Typewise’s next word prediction smarts (gleaned by via checking out its iOS app) are pretty low key (ha!). But it’s v1 of the AI — and Eberle talks bullishly of having “world class” developers working on it.
“The collaboration with ETH just started a few weeks ago and thus there are no significant improvements yet visible in the live app,” he tells TechCrunch. “As the collaboration runs until the end of 2021 (with the opportunity of extension) the vast majority of innovation is still to come.”
He also tells us Typewise is working with ETH’s Prof. Thomas Hofmann (chair of the Data Analytic Lab, formerly at Google), as well as having has two PhDs in NLP/ML and one MSc in ML contributing to the effort.
“We get exclusive rights to the [ETH] technology; they don’t hold equity but they get paid by the Swiss government on our behalf,” Eberle also notes.
Typewise says its smartphone app supports more than 35 languages. But its next word prediction AI can only handle English, German, French, Italian and Spanish at this point. The startup says more are being added.
Powered by WPeMatico
UK gives up on centralized coronavirus contacts-tracing app — will ‘likely’ switch to model backed by Apple and Google
The UK has given up building a centralized coronavirus contacts-tracing app and will instead switch to a decentralized app architecture, the BBC has reported. This suggests its any future app will be capable of plugging into the joint ‘exposure notification’ API which has been developed in recent weeks by Apple and Google.
The UK’s decision to abandon a bespoke app architecture comes more than a month after ministers had been reported to be eyeing such a switch. They went on to award a contract to an IT supplier to develop a decentralized tracing app in parallel as a backup — while continuing to test the centralized app, which is called NHS COVID-19.
At the same time, a number of European countries have now successfully launched contracts-tracing apps with a decentralized app architecture that’s able to plug into the ‘Gapple’ API — including Denmark, Germany, Italy, Latvia and Switzerland. Several more such apps remain in testing. While EU Member States just agreed on a technical framework to enable cross-border interoperability of apps based on the same architecture.
Germany — which launched the decentralized ‘Corona Warning App’ this week — announced its software had been downloaded 6.5M times in the first 24 hours. The country had initially appeared to favor a centralized approach but switched to a decentralized model back in April in the face of pushback from privacy and security experts.
The UK’s NHS COVID-19 app, meanwhile, has not progressed past field tests, after facing a plethora of technical barriers and privacy challenges — as a direct consequence of the government’s decision to opt for a proprietary system which uploads proximity data to a central server, rather than processing exposure notifications locally on device.
Apple and Google’s API, which is being used by all Europe’s decentralized apps, does not support centralized app architectures — meaning the UK app faced technical hurdles related to accessing Bluetooth in the background. The centralized choice also raised big questions around cross-border interoperability, as we’ve explained before. Questions had also been raised over the risk of mission creep and a lack of transparency and legal certainty over what would be done with people’s data.
So the UK’s move to abandon the approach and adopt a decentralized model is hardly surprising — although the time it’s taken the government to arrive at the obvious conclusion does raise some major questions over its competence at handling technology projects.
Michael Veale, a lecturer in digital rights and regulation at UCL — who has been involved in the development of the DP3T decentralized contacts-tracing standard, which influenced Apple and Google’s choice of API — welcomed the UK’s decision to ditch a centralized app architecture but questioned why the government has wasted so much time.
“This is a welcome, if a heavily and unnecessarily delayed, move by NHSX,” Veale told TechCrunch. “The Google -Apple system in a way is home-grown: Originating with research at a large consortium of universities led by Switzerland and including UCL in the UK. NHSX has no end of options and no reasonable excuse to not get the app out quickly now. Germany and Switzerland both have high quality open source code that can be easily adapted. The NHS England app will now be compatible with Northern Ireland, the Republic of Ireland, and also the many destinations for holidaymakers in and out of the UK.”
Perhaps unsurprisingly, UK ministers are now heavily de-emphasizing the importance of having an app in the fight against the coronavirus at all.
The Department for Health and Social Care’s, Lord Bethell, told the Science and Technology Committee yesterday the app will not now be ready until the winter. “We’re seeking to get something going for the winter, but it isn’t a priority for us,” he said.
Yet the centralized version of the NHS COVID-19 app has been in testing in a limited geographical pilot on the Isle of Wight since early May — and up until the middle of last month health minister, Matt Hancock, had said it would be rolled out nationally in mid May.
Of course that timeframe came and went without launch. And now the prospect of the UK having an app at all is being booted right into the back end of the year.
Compare and contrast that with government messaging at its daily coronavirus briefings back in May — when Hancock made “download the app” one of the key slogans — and the word ‘omnishambles‘ springs to mind…
NHSX relayed our request for comment on the switch to a decentralized system and the new timeframe for an app launch to the Department of Health and Social Care (DHSC) — but the department had not responded to us at the time of publication.
Earlier this week the BBC reported that a former Apple executive, Simon Thompson, was taking charge of the delayed app project — while the two lead managers, the NHSX’s Matthew Gould and Geraint Lewis — were reported to be stepping back.
Back in April, Gould told the Science and Technology Committee the app would “technically” be ready to launch in 2-3 weeks’ time, though he also said any national launch would depend on the preparedness of a wider government program of coronavirus testing and manual contacts tracing. He also emphasized the need for a major PR campaign to educate the public on downloading and using the app.
Government briefings to the press today have included suggestions that app testers on the Isle of Wight told it they were not comfortable receiving COVID-19 notifications via text message — and that the human touch of a phone call is preferred.
However none of the European countries that have already deployed contacts-tracing apps has promoted the software as a one-stop panacea for tackling COVID-19. Rather tracing apps are intended to supplement manual contacts-tracing methods — the latter involving the use of trained humans making phone calls to people who have been diagnosed with COVID-19 to ask who they might have been in contact with over the infectious period.
Even with major resource put into manual contacts-tracing, apps — which use Bluetooth signals to estimate proximity between smartphone users in order to calculate virus expose risk — could still play an important role by, for example, being able to trace strangers who are sat near an infected person on public transport.
Update: The DHSC has now issued a statement addressing reports of the switch of app architecture for the NHS COVID-19 app — in which it confirms, in between reams of blame-shifting spin, that it’s testing a new app that is able to plug into the Apple and Google API — and which it says it may go on to launch nationally, but without providing any time frame.
It also claims it’s working with Apple and Google to try to enhance how their technology estimates the distance between smartphone users.
“Through the systematic testing, a number of technical challenges were identified — including the reliability of detecting contacts on specific operating systems — which cannot be resolved in isolation with the app in its current form,” DHSC writes of the centralized NHS COVID-19 app.
“While it does not yet present a viable solution, at this stage an app based on the Google / Apple API appears most likely to address some of the specific limitations identified through our field testing. However, there is still more work to do on the Google / Apple solution which does not currently estimate distance in the way required.”
“Based on this, the focus of work will shift from the current app design and to work instead with Google and Apple to understand how using their solution can meet the specific needs of the public,” it adds.
We reached out to Apple and Google for comment. Apple declined to comment.
According to one source, the UK has been pressing for the tech giants’ API to include device model and RSSI info alongside the ephemeral IDs which devices that come into proximity exchange with each other — presumably to try to improve distance calculations via a better understanding of the specific hardware involved.
However introducing additional, fixed pieces of device-linked data would have the effect of undermining the privacy protections baked into the decentralized system — which uses ephemeral, rotating IDs in order to prevent third party tracking of app users. Any fixed data-points being exchanged would risk unpicking the whole anti-tracking approach.
Norway, another European country which opted for a centralized approach for coronavirus contacts tracing — but got an app launched in mid April — made the decision to suspend its operation this week, after an intervention by the national privacy watchdog. In that case the app was collecting both GPS and Bluetooth — posing a massive privacy risk. The watchdog warned the public health agency the tool was no longer a proportionate intervention — owing to what are now low levels of coronavirus risk in the country.
Powered by WPeMatico
You can now install the first beta of Android 11
After a series of developer previews, Google today released the first beta of Android 11, and with that, it is also making these pre-release versions available for over-the-air updates. This time around, the list of supported devices only includes the Pixel 2, 3, 3a and 4.
If you’re brave enough to try this early version (and I wouldn’t do so on your daily driver until a few more people have tested it), you can now enroll here. Like always, Google is also making OS images available for download and an updated emulator is available, too.
Google says the beta focuses on three key themes: people, controls and privacy.
Like in previous updates, Google once again worked on improving notifications — in this case, conversation notifications, which now appear in a dedicated section at the top of the pull-down shade. From there, you will be able to take actions right from inside the notification or ask the OS to remind you of this conversation at a later time. Also new is built-in support in the notification system for what are essentially chat bubbles, which messaging apps can now use to notify you even as you are working (or playing) in another app.
Another new feature is consolidated keyboard suggestions. With these, Autofill apps and Input Method Editors (think password managers and third-party keyboards), can now securely offer context-specific entries in the suggestion strip. Until now, enabling autofill for a password manager, for example, often involved delving into multiple settings and the whole experience often felt like a bit of a hack.
For those users who rely on voice to control their phones, Android now uses a new on-device system that aims to understand what is on the screen and then automatically generates labels and access points for voice commands.
As for controls, Google is now letting you long-press the power button to bring up controls for your smart home devices (though companies that want to appear in this new menu need to make use of Google’s new API for this). In one of the next beta releases, Google will also enable media controls that will make it easier to switch the output device for their audio and video content.
In terms of privacy, Google is adding one-time permissions so that an app only gets access to your microphone, camera or location once, as well as auto-resets for permissions when you haven’t used an app for a while.
A few months ago, Google said that developers would need to get a user’s approval to access background location. That caused a bit of a stir among developers and now Google will keep its current policies in place until 2021 to give developers more time to update their apps.
In addition to these user-facing features, Google is also launching a series of updates aimed at Android developers. You can read more about them here.
Powered by WPeMatico
Data startup Axiom secures $4M from Crane Venture Partners, emerges from stealth
Axiom, a startup that helps companies deal with their internal data, has secured a new $4 million seed round led by U.K.-based Crane Venture Partners, with participation from LocalGlobe, Fly VC and Mango Capital. Notable angel investors include former Xamarin founder and current GitHub CEO Nat Friedman and Heroku co-founder Adam Wiggins. The company is also emerging from a relative stealth mode to reveal that is has now raised $7 million in funding since it was founded in 2017.
The company says it is also launching with an enterprise-grade solution to manage and analyze machine data “at any scale, across any type of infrastructure.” Axiom gives DevOps teams a cloud-native, enterprise-grade solution to store and query their data all the time in one interface — without the overhead of maintaining and scaling data infrastructure.
DevOps teams have spent a great deal of time and money managing their infrastructure, but often without being able to own and analyze their machine data. Despite all the tools at hand, managing and analyzing critical data has been difficult, slow and resource-intensive, taking up far too much money and time for organizations. This is what Axiom is addressing with its platform to manage machine data and surface insights, more cheaply, they say, than other solutions.
Co-founder and CEO Neil Jagdish Patel told TechCrunch: “DevOps teams are stuck under the pressure of that, because it’s up to them to deliver a solution to that problem. And the solutions that existed are quite, well, they’re very complex. They’re very expensive to run and time-consuming. So with Axiom, our goal is to try and reduce the time to solve data problems, but also allow businesses to store more data to query at whenever they want.”
Why did they work with Crane? “We needed to figure out how enterprise sales work and how to take this product to market in a way that makes sense for the people who need it. We spoke to different investors, but when I sat down with Crane they just understood where we were. They have this razor-sharp focus on how they get you to market and how you make sure your sales process and marketing is a success. It’s been beneficial to us as were three engineers, so you need that,” said Patel.
Commenting, Scott Sage, founder and partner at Crane Venture Partners added: “Neil, Seif and Gord are a proven team that have created successful products that millions of developers use. We are proud to invest in Axiom to allow them to build a business helping DevOps teams turn logging challenges from a resource-intense problem to a business advantage.”
Axiom co-founders Neil Jagdish Patel, Seif Lotfy and Gord Allott previously created Xamarin Insights that enabled developers to monitor and analyse mobile app performance in real time for Xamarin, the open-source cross-platform app development framework. Xamarin was acquired by Microsoft for between $400 and $500 million in 2016. Before working at Xamarin, the co-founders also worked together at Canonical, the private commercial company behind the Ubuntu Project.
Powered by WPeMatico
UK eyeing switch to Apple-Google API for coronavirus contacts tracing — report
The UK may be rethinking its decision to shun Apple and Google’s API for its national coronavirus contacts tracing app, according to the Financial Times, which reported yesterday that the government is paying an IT supplier to investigate whether it can integrate the tech giants’ approach after all.
As we’ve reported before coronavirus contacts tracing apps are a new technology which aims to repurpose smartphones’ Bluetooth signals and device proximity to try to estimate individuals’ infection risk.
The UK’s forthcoming app, called NHS COVID-19, has faced controversy because it’s being designed to use a centralized app architecture. This means developers are having to come up with workarounds for platform limitations on background access to Bluetooth as the Apple-Google cross-platform API only works with decentralized systems.
The choice of a centralized app architecture has also raised concerns about the impact of such an unprecedented state data grab on citizens’ privacy and human rights, and the risk of state ‘mission creep‘.
The UK also looks increasingly isolated in its choice in Europe after the German government opted to switch to a decentralized model, joining several other European countries that have said they will opt for a p2p approach, including Estonia, Ireland and Switzerland.
In the region, France remains the other major backer of a centralized system for its forthcoming coronavirus contacts tracing app, StopCovid.
Apple and Google, meanwhile, are collaborating on a so-called “exposure notification” API for national coronavirus contacts tracing apps. The API is slated to launch this month and is designed to remove restrictions that could interfere with how contact events are logged. However it’s only available for apps that don’t hold users’ personal data on central servers and prohibits location tracking, with the pair emphasizing that their system is designed to put privacy at the core.
Yesterday the FT reported that NHSX, the digital transformation branch of UK’s National Health Service, has awarded a £3.8M contract to the London office of Zuhlke Engineering, a Switzerland-based IT development firm which was involved in developing the initial version of the NHS COVID-19 app.
The contract includes a requirement to “investigate the complexity, performance and feasibility of implementing native Apple and Google contact tracing APIs within the existing proximity mobile application and platform”, per the newspaper’s report.
The work is also described as a “two week timeboxed technical spike”, which the FT suggests means it’s still at a preliminary phase — thought it also notes the contract includes a deadline of mid-May.
The contracted work was due to begin yesterday, per the report.
We’ve reached out to Zuhlke for comment. Its website describes the company as “a strong solutions partner” that’s focused on projects related to digital product delivery; cloud migration; scaling digital platforms; and the Internet of Things.
We also put questions arising from the FT report to NHSX.
At the time of writing the unit had not responded but yesterday a spokesperson told the newspaper: “We’ve been working with Apple and Google throughout the app’s development and it’s quite right and normal to continue to refine the app.”
The specific technical issue that appears to be causing concern relates to a workaround the developers have devised to try to circumvent platform limitations on Bluetooth that’s intended to wake up phones when the app itself is not being actively used in order that the proximity handshakes can still be carried out (and contacts events properly logged).
Thing is, if any of the devices fail to wake up and emit their identifiers so other nearby devices can log their presence there will be gaps in the data. Which, in plainer language, means the app might miss some close encounters between users — and therefore fail to notify some people of potential infection risk.
Recent reports have suggested the NHSX workaround has a particular problem with iPhones not being able to wake up other iPhones. And while Google’s Android OS is the more dominant platform in the UK (running on circa ~60% of smartphones, per Kantar) there will still be plenty of instances of two or more iPhone users passing near each other. So if their apps fail to wake up they won’t exchange data and those encounters won’t be logged.
On this, the FT quotes one person familiar with the NHS testing process who told it the app was able to work in the background in most cases, except when two iPhones were locked and left unused for around 30 minutes, and without any Android devices coming within 60m of the devices. The source also told it that bringing an Android device running the app close to the iPhone would “wake up” its Bluetooth connection.
Clearly, the government having to tell everyone in the UK to use an Android smartphone not an iPhone wouldn’t be a particularly palatable political message.
This is effectively a form of Android Herd Immunity: for the good of Britain, vaccinate your friends by giving them Androids!
— Michael Veale (@mikarv) May 5, 2020
One source with information about the NHSX testing process told us the unit has this week been asking IT suppliers for facilities or input on testing environments with “50-100 Bluetooth devices of mixed origin”, to help with challenges in testing the Bluetooth exchanges — which raises questions about how extensively this core functionality has been tested up to now. (Again, we’ve put questions to the NHSX about testing and will update this report with any response.)
Work on planning and developing the NHS COVID-19 app began March 7, according to evidence given to a UK parliamentary committee by the NHSX CEO’s, Matthew Gould, last month.
Gould has also previously suggested that the app could be “technically” ready to launch in as little as two or three weeks time from now. While a limited geographical trial of the app kicked off this week in the Isle of Wight. Prior to that, an alpha version of the app was tested at an RAF base involving staff carrying out simulations of people going shopping, per a BBC report last month.
Gould faced questions over the choice of centralized vs decentralized app architecture from the human rights committee earlier this week. He suggested then that the government is not “locked” to the choice — telling the committee: “We are constantly reassessing which approach is the right one — and if it becomes clear that the balance of advantage lies in a different approach then we will take that different approach. We’re not irredeemably wedded to one approach; if we need to shift then we will… It’s a very pragmatic decision about what approach is likely to get the results that we need to get.”
However it’s unclear how quickly such a major change to app architecture could be implemented, given centralized vs decentralized systems work in very different ways.
Additionally, such a big shift — more than two months into the NHSX’s project — seems, at such a late stage, as if it would be more closely characterized as a rebuild, rather than a little finessing (as suggested by the NHSX spokesperson’s remark to the FT vis-a-vis ‘refining’ the app).
In related news today, Reuters reports that Colombia has pulled its own coronavirus contacts tracing app after experiencing glitches and inaccuracies. The app had used alternative technology to power contacts logging via Bluetooth and wi-fi. A government official told the news agency it aims to rebuild the system and may now use the Apple-Google API.
Australia has also reported Bluetooth related problems with its national coronavirus app. And has also been reported to be moving towards adopting the Apple-Google API.
While, Singapore, the first country to launch a Bluetooth app for coronavirus contacts tracing, was also the first to run into technical hitches related to platform limits on background access — likely contributing to low download rates for the app (reportedly below 20%).
Powered by WPeMatico
A cryptocurrency stealing app found on Google Play was downloaded over a thousand times
Researchers have found two apps masquerading as cryptocurrency apps on Android’s app store, Google Play.
One of them was largely a dud. The second was designed to steal cryptocurrency, the researchers said.
Security firm ESET said one of the two fake Android apps impersonated Trezor, a hardware cryptocurrency wallet. The good news is that the app couldn’t be used to steal cryptocurrency stored by Trezor. But the researchers found the app was connected to a second Android app that could have been used to scam funds out of unsuspecting victims.
Lukas Stefanko, a security researcher at ESET — who has a long history of finding dodgy Android apps — said the fake Trezor app “appeared trustworthy at first glance” but was using a fake developer name to impersonate the company.
The fake app was designed to trick users into turning over a victim’s login credentials. Uploaded to Google Play on May 1, the app quickly ranked as the second-most popular search result when searching for “Trezor” behind the legitimate app, said Stefanko. Users on Reddit also found the fake app and reported it as recently as two weeks ago.
According to Stefanko, the server where user credentials were sent was linked to a website linked to another fake wallet, purportedly to store cryptocurrency, and also listed on Google Play since February 25.
“The app claims it lets its users create wallets for various cryptocurrencies,” said Stefanko. “However, its actual purpose is to trick users into transferring cryptocurrency into the attackers’ wallets – a classic case of what we’ve named wallet address scams in our previous research into cryptocurrency-targeting malware.”
Both apps were collectively downloaded more than a thousand times. After ESET contacted Google, the apps were pulled offline the next day.
Read more:
- Half a million Android users tricked into downloading malware from Google Play
- Security researchers find over a dozen iPhone apps linked to Golduck malware
- A powerful spyware app now targets iPhone owners
- Google warns app developers of three malicious SDKs being used for ad fraud
- Apple tells app developers to disclose or remove screen recording code
- Apple restores Google’s internal iOS apps after certificate misuse punishment
Powered by WPeMatico
Skedulo raises $28M for its mobile workforce management service
Skedulo, a service that helps businesses manage their mobile employees, today announced that it has raised a $28 million Series B funding round led by M12, Microsoft’s venture fund. Existing investors Blackbird and Castanoa Ventures also participated in this round.
The company’s service offers businesses all the necessary tools to manage their mobile employees, including their schedules. A lot of small businesses still use basic spreadsheets and email to do this, but that’s obviously not the most efficient way to match the right employee to the right job, for example.
“Workforce management has traditionally been focused on employees that are sitting at a desk for the majority of their day,” Skedulo CEO and co-founder Matt Fairhurst told me. “The overwhelming majority — 80 percent — of workers will be deskless by 2020 and so far, there has been no one that has addressed the needs of this growing population at scale. We’re excited to help enterprises confront these challenges head-on so they can compete and lean into rapidly changing customer and employee expectations.”
At the core of Skedulo, which offers both a mobile app and web-based interface, is the company’s so-called “Mastermind” engine that helps businesses automatically match the right employee to a job based on the priorities the company has specified. The company plans to use the new funding to enhance this tool through new machine learning capabilities. Skedulo will also soon offer new analytics tools and integrations with third-party services like HR and financial management tools, as well as payroll systems.
The company also plans to use the new funding to double its headcount, which includes hiring at least 60 new employees in its Australian offices in Brisbane and Sydney.
As part of this round, Priya Saiprasad, principal of M12, will join Skedulo’s board of directors. “We found a strong sense of aligned purpose with Priya Saiprasad and the team at M12 — and their desire to invest in companies that help reduce cycles in a person’s working day,” Fairhurst said. “Fundamentally, Skedulo is a productivity company. We help companies, the back-office and mobile workforce, reduce the number of cycles it takes to get work done. This gives them time back to focus on the work that matters most.”
Powered by WPeMatico
Sam’s Club to test new Scan & Go system that uses computer vision instead of barcodes
In October, Walmart-owned Sam’s Club opened a test store in Dallas where it planned to trial new technology, including mobile checkout, an Amazon Go-like camera system, in-store navigation, electronic shelf labels and more. This morning, the retailer announced it will now begin testing a revamped Scan & Go service as well, which leverages computer vision and machine learning to make mobile scanning easier and faster.
The current Scan & Go system, launched two years ago, requires Sam’s Club shoppers to locate the barcode on the item they’re buying and scan it using the Sam’s Club mobile app. The app allows shoppers to account for items they’re buying as they place them in their shopping cart, then pay in the app instead of standing in line at checkout.
However convenient, the system itself can still be frustrating at times because you’ll need to actually find the barcode on the item — often turning the item over from one side to the other to find the sticker or tag. This process can be difficult for heavier items, and frustrating when the barcoded label or tag has fallen off.
It also can end up taking several seconds to complete — which adds up when you’re filling a cart with groceries during a big stocking-up trip.
The new scanning technology will instead use computer vision and ML (machine learning) to recognize products without scanning the barcode, cutting the time it takes for the app to identify the product in question, the retailer explains.
In a video demo, Sam’s Club showed how it might take a typical shopper 9.3 seconds to scan a pack of water using the old system, versus 3.4 seconds using the newer technology.
Of course, the times will vary based on the shopper’s skill, the item being scanned and how well the technology performs, among other factors. A large package of water is a more extreme example, but one that demonstrates well the potential of the system… if it works.
The idea with the newly opened Dallas test store is to put new technology into practice quickly in a real-world environment, to see what performs well and what doesn’t, while also gathering customer feedback. Dallas was chosen as the location for the store because of the tech talent and recruiting potential in the area, and because it’s a short trip from Walmart’s Bentonville, Arkansas headquarters, the company said earlier.
Sam’s Club says it has filed a patent related to the new scanning technology, and will begin testing it this spring at the Dallas area “Sam’s Club Now” store. It will later expand the technology to the tools used by employees, too.
Powered by WPeMatico
Many popular iPhone apps secretly record your screen without asking
Many major companies, like Air Canada, Hollister and Expedia, are recording every tap and swipe you make on their iPhone apps. In most cases you won’t even realize it. And they don’t need to ask for permission.
You can assume that most apps are collecting data on you. Some even monetize your data without your knowledge. But TechCrunch has found several popular iPhone apps, from hoteliers, travel sites, airlines, cell phone carriers, banks and financiers, that don’t ask or make it clear — if at all — that they know exactly how you’re using their apps.
Worse, even though these apps are meant to mask certain fields, some inadvertently expose sensitive data.
Apps like Abercrombie & Fitch, Hotels.com and Singapore Airlines also use Glassbox, a customer experience analytics firm, one of a handful of companies that allows developers to embed “session replay” technology into their apps. These session replays let app developers record the screen and play them back to see how its users interacted with the app to figure out if something didn’t work or if there was an error. Every tap, button push and keyboard entry is recorded — effectively screenshotted — and sent back to the app developers.
Or, as Glassbox said in a recent tweet: “Imagine if your website or mobile app could see exactly what your customers do in real time, and why they did it?”
The App Analyst, a mobile expert who writes about his analyses of popular apps on his eponymous blog, recently found Air Canada’s iPhone app wasn’t properly masking the session replays when they were sent, exposing passport numbers and credit card data in each replay session. Just weeks earlier, Air Canada said its app had a data breach, exposing 20,000 profiles.
“This gives Air Canada employees — and anyone else capable of accessing the screenshot database — to see unencrypted credit card and password information,” he told TechCrunch.
In the case of Air Canada’s app, although the fields are masked, the masking didn’t always stick (Image: The App Analyst/supplied)
We asked The App Analyst to look at a sample of apps that Glassbox had listed on its website as customers. Using Charles Proxy, a man-in-the-middle tool used to intercept the data sent from the app, the researcher could examine what data was going out of the device.
Not every app was leaking masked data; none of the apps we examined said they were recording a user’s screen — let alone sending them back to each company or directly to Glassbox’s cloud.
That could be a problem if any one of Glassbox’s customers aren’t properly masking data, he said in an email. “Since this data is often sent back to Glassbox servers I wouldn’t be shocked if they have already had instances of them capturing sensitive banking information and passwords,” he said.
The App Analyst said that while Hollister and Abercrombie & Fitch sent their session replays to Glassbox, others like Expedia and Hotels.com opted to capture and send session replay data back to a server on their own domain. He said that the data was “mostly obfuscated,” but did see in some cases email addresses and postal codes. The researcher said Singapore Airlines also collected session replay data but sent it back to Glassbox’s cloud.
Without analyzing the data for each app, it’s impossible to know if an app is recording a user’s screens of how you’re using the app. We didn’t even find it in the small print of their privacy policies.
Apps that are submitted to Apple’s App Store must have a privacy policy, but none of the apps we reviewed make it clear in their policies that they record a user’s screen. Glassbox doesn’t require any special permission from Apple or from the user, so there’s no way a user would know.
Expedia’s policy makes no mention of recording your screen, nor does Hotels.com’s policy. And in Air Canada’s case, we couldn’t spot a single line in its iOS terms and conditions or privacy policy that suggests the iPhone app sends screen data back to the airline. And in Singapore Airlines’ privacy policy, there’s no mention, either.
We asked all of the companies to point us to exactly where in its privacy policies it permits each app to capture what a user does on their phone.
Only Abercombie responded, confirming that Glassbox “helps support a seamless shopping experience, enabling us to identify and address any issues customers might encounter in their digital experience.” The spokesperson pointing to Abercrombie’s privacy policy makes no mention of session replays, neither does its sister-brand Hollister’s policy.
“I think users should take an active role in how they share their data, and the first step to this is having companies be forthright in sharing how they collect their users data and who they share it with,” said The App Analyst.
When asked, Glassbox said it doesn’t enforce its customers to mention its usage in their privacy policy.
“Glassbox has a unique capability to reconstruct the mobile application view in a visual format, which is another view of analytics, Glassbox SDK can interact with our customers native app only and technically cannot break the boundary of the app,” the spokesperson said, such as when the system keyboard covers part of the native app, “Glassbox does not have access to it,” the spokesperson said.
Glassbox is one of many session replay services on the market. Appsee actively markets its “user recording” technology that lets developers “see your app through your user’s eyes,” while UXCam says it lets developers “watch recordings of your users’ sessions, including all their gestures and triggered events.” Most went under the radar until Mixpanel sparked anger for mistakenly harvesting passwords after masking safeguards failed.
It’s not an industry that’s likely to go away any time soon — companies rely on this kind of session replay data to understand why things break, which can be costly in high-revenue situations.
But for the fact that the app developers don’t publicize it just goes to show how creepy even they know it is.
Got a tip? You can send tips securely over Signal and WhatsApp to +1 646-755–8849. You can also send PGP email with the fingerprint: 4D0E 92F2 E36A EC51 DAAE 5D97 CB8C 15FA EB6C EEA5.
Powered by WPeMatico
Google brings offline neural machine translations for 59 languages to its Translate app
Currently, when the Google Translate apps for iOS and Android has access to the internet, its translations are far superior to those it produces when it’s offline. That’s because the offline translations are phrase-based, meaning they use an older machine translation technique than the machine learning-powered systems in the cloud that the app has access to when it’s online. But that’s changing today. Google is now rolling out offline Neural Machine Translation (NMT) support for 59 languages in the Translate apps.
Today, only a small number of users will see the updated offline translations, but it will roll out to all users within the next few weeks.
The list of supported languages consists of a wide range of languages. Because I don’t want to play favorites, here is the full list: Afrikaans, Albanian, Arabic, Belarusian, Bengali, Bulgarian, Catalan, Chinese, Croatian, Czech, Danish, Dutch, English, Esperanto, Estonian, Filipino, Finnish, French, Galician, Georgian, German, Greek, Gujarati, Haitian, Creole, Hebrew, Hindi, Hungarian, Icelandic, Indonesian, Irish, Italian, Japanese, Jannada, Korean, Latvian, Lithuanian, Macedonian, Malay, Maltese, Marathi, Norwegian, Persian, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Swahili, Swedish, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu, Vietnamese and Welsh.
In the past, running these deep learning models on a mobile device wasn’t really an option since mobile phones didn’t have the right hardware to efficiently run them. Now, thanks to both advances in hardware and software, that’s less of an issue and Google, Microsoft and others have also found ways to compress these models to a manageable size. In Google’s case, that’s about 30 to 40 megabytes per language.
It’s worth noting that Microsoft also announced a similar feature for its Translator app earlier this year. It uses a very similar technique, but for the time being, it only supports about a dozen languages.
Powered by WPeMatico