On Tuesday, transportation network company, Lyft announced they have released an autonomous vehicle open-source data set; one which the company believes to be the largest of its kind.
In an effort to fast track the development of vehicles able to drive themselves, the set is freely available in the existing nuScenes format which was initially developed by Aptiv.
With this release, Lyft aim “to empower the community, stimulate further development, and share our insights into future opportunities from the perspective of an advanced industrial autonomous vehicles program.”
Looking into the data set more closely, it is indeed vast. The company has amassed data gathered from a fleet of autonomous vehicles in a bounded geographic area. Each car was equipped with an impressive array of info gathering tech; 40-beam roof lidar and 40-beam bumper lidars, wide-field-of-view cameras, and a long-focal-length camera mounted slightly pointing up to detect traffic lights.
The collection itself was compiled by Lyft’s Level 5 team. This stellar group comprises of over 300 engineers, applied researchers, product managers, operations managers, and more; all working toward building a self-driving system for ride-sharing.
Coupled with this latest open source release, Lyft also announced an engineering challenge for developers interested in building machine learning models using the samples it contains. The lucky winner will enjoy $25,000 in cash prizes. Lyft will also fly out the top contestants to December’s NeurIPS conference.
Lyft have been working hard to expand the debate within the autonomous vehicle space for some time. In March last year, Lyft began tests of autonomous cars on public highways; just one year after it had completed tests on private roads and ran employee pilots. Additionally, in May the company partnered with Waymo to enable Waymo customers to hail driverless Waymo cars from the Lyft app. It also has an ongoing collaboration with self-driving car startup Aptiv, which makes a small fleet of autonomous vehicles available to Lyft customers in Las Vegas.
There does however appear to be a trend emerging! In June, Waymo also revealed a high-quality multimodal sensor data set for autonomous driving at the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). The Waymo model features 3,000 driving scenes totaling 16.7 hours of video data, 600,000 frames, and approximately 25 million 3D bounding boxes and 22 million 2D bounding boxes.
Although these two sets are the most recent, there are others influencing the conversation. Other open source libraries include ones from Mapillary, KITTI and the Cityscapes dataset developed and maintained by luxury vehicle manufacturer, Daimler.
Lyft have high expectations for the influencing properties of their set. In a statement launched at the release had this to say;
“Autonomous vehicles are expected to dramatically redefine the future of transportation. When fully realized, this technology promises to unlock a myriad of societal, environmental, and economic benefits”.
It is undeniable that to be effective, autonomous vehicles need both informed engineers and a fully integrated smart road network. Open-source data such as that created by Lyft and its peers will only propel the progression of the development of these self-driven vehicles. Maybe we could see fully autonomous vehicles on roads across the globe sooner than we think?
LinkedIn is migrating to Microsoft Azure.
Professional networking site, LinkedIn have announced this week that they are swapping out its data centers for the public cloud.
Perhaps this isn’t so surprising. Some commentators have suggested that this development was always on the cards following Microsoft’s 2016 acquisition of LinkedIn. This assertion has been disputed by Mohak Shroff, SVP of engineering at LinkedIn however. When questioned about the biggest technological transformation in the company’s history, he remarked that although “the integration has really been a remarkable success … Microsoft has really allowed us independence in decision making”. He went on to say that Microsoft have “offered us whatever resources we have needed, offered to work with us if we make the decision to move, but the decision has always been ours.”
The transition though is set to be slow and carefully controlled. LinkedIn plans to migrate its 645 million members over a three year period so as not to compromise service, user accessibility and site reliability. Shroff confirmed. “It will be a gradual migration. We’ll see increasing workloads on Azure over time, with a pretty significant inflection point, about a year and a half, two years out from now. And then kind of an accelerated migration post that.”
The decision to move to the public cloud was made with the intention of benefiting its members in a number of ways. The number of Azure regions becoming available to LinkedIn is growing, and as a result, the site looks set to become more proximate to its members.
Not only this. The malleable and scalable nature of the public cloud allows for moveable levels of service. On the whole, LinkedIn has a fairly consistent pattern of traffic, however, there are times when it surges; especially with offline processing. The brains behind LinkedIn are expecting to enjoy the strength of Azure’s data analytics, AI, and ML capabilities, plus its ability to leverage Graphics Processing Units.
Over time, it is also thought that the move will facilitate LinkedIn’s ability to focus on building product, securing the site, and leveraging at-scale systems on the Azure stack. The hope is that the move to Azure will free its engineers to build customer-facing functionality.
The recruitment industry relies heavily on LinkedIn, and of course, our day to day at Ignite Digital HQ is no exception. As heavy users of the platform, It will be interesting to see over the coming years if we are indeed able to harness the benefits outlined by Shroff in his publicised statements.
Smart Cities impact workplace trends.
A recent report published by the International Data Corporation indicates that an estimated two-thirds of global cities will be investing up to $135 billion in smart city technology by 2021. Such high levels of investment are not surprising as the world becomes increasingly dependent on technology for efficiency and success.
Cities all across the globe are emerging as “smart cities”. From these hubs of innovation, workplace trends are having to evolve to keep pace.
Many 21st century workplaces now offer a BYOD policy. Bring Your Own Device protocol allows employees to use their own personal laptop, tablet, wearable tech or smartphone for example.
According to research, 67% of employees say they now use personal devices at work, while 87% of businesses rely on their workers’ access to mobile business apps. As remote working has become both more common and highly valued, this allows employees to take their favourite at-home tech and integrate it better, allowing them to work remotely in a more efficient manner.
However, BYOD is not without its challenges and has itself been a catalyst for another trend revealed by the study. As this and the wireless technology trend enjoy its boom, so too does the need for advanced security tech. More data is being stored on the cloud. If that data isn’t secured correctly, companies risk losing or leaking a lot of sensitive information.
In line with this, a recent review by SHRM found that nearly half of organisations surveyed (46%) are now using biometric authentication tech to protect data on smartphones. Smart cities offer flexible offices spaces with a choice of shared
and private workspaces. This “hot-desk” formula requires both excellent network and round-the-clock on-premises standard security. This tandem strategy is being implemented across these smart cities, and are helping companies to operate in a safer and more secure environment.
Awareness of workplace wellness is a trend which has also emerged as a feature of Smart City working. Although, this may seem a world away from the tech scene; notorious for its high output “996” environment … it is tech itself that employees and employers are turning to in order to reduce burnout and addiction to overwork. According to a Global Wellness Institute Survey, 70-80% of companies believe that wellness programmes reduce absenteeism and boost productivity.
In line with these findings, companies are looking to make positive changes to improve the wellbeing of their employees. Wearable devices, such as Fitbits allow employees to monitor their stats; ensuring they hit their steps, control their heart rates and even ensure they get enough sleep. Really savvy smart companies are also making use of this data. They are collating it to identify potential risks in the workplace, both in terms of overworked individuals and health and safety.
Alibaba to help Salesforce localise and sell in China.
The 20-year-old leader in Customer Relationship Management (CRM), Salesforce are teaming up with Alibaba in order to make headway in the Asian market.
The partners both have something to gain from this new relationship. Alibaba does not have a Salesforce equivalent, servicing the small and medium businesses selling through its market place or using its cloud computing services. This new alliance will fill that gap.
Alibaba will be the exclusive provider of Salesforce to enterprise customers in mainland China, Hong Kong, Macau, and Taiwan; while Salesforce will become the exclusive enterprise CRM software suite sold by Alibaba.
For Salesforce, the collaboration will also promote sales avenues in China. Through Alibaba’s cloud infrastructure and data platform, Salesforce will be able to offer localised solutions and better serve its multinational customers.
Ken Shen, vice president of Alibaba’s cloud intelligence commented on the motivation behind the partnership. He cited that more and more of their multinational customers were asking for support wherever they do business around the world. The team-up with Salesforce would facilitate that and ultimately should improve revenue from these markets.
Overall, only 10% of Salesforce’ revenue stream came from Asia in the early part of the year. By comparison, 70% came from the Americas and 20% from the European market.
Besides gaining international custom, the partnership will also enable Salesforce to store its China-based data at Alibaba Cloud. This is beneficial due to the restrictions in China with regard to how overseas companies must store and use customer data. Chinese regulation requires such enterprises to work with domestic firms when processing and storing data sourced from Chinese users.
Alibaba’s cloud-based ambitions also look set to be realised by its Salesforce alliance. Cloud growth has been a key vertical at Alibaba. Securing such a heavyweight partner will only help underline its dominance as China’s biggest cloud service provider.
Thankfully, our haul of fantastic contractors placed from Ignite Digital HQ give us, or our clients, very few problems. The same cannot be said for one contract programmer based in Pennsylvania who was working on a project for the US arm of German tech firm, Siemens.
Dave Tinley had enjoyed a placement at the tech conglomerate creating custom automated spreadsheets used by the firm to manage orders for electrical equipment.
Closer investigations into his work revealed that the orders were not the only thing being controlled by Mr Tinley; so too were the demands on his service!
It has been alleged that this rogue contractor had used his skills to install a “Logic-Bomb” into the spreadsheet; one that would force his client to become dependent on his ability to fix the recurrent problem!
‘Logic Bombs’ are bits of malicious code that disrupt the program when specific conditions are met, like a specific time on a specific date, for example. In this case, Tinley had timed the ‘bomb’ to go off every few years causing the spreadsheet to experience glitches like error messages and size changes to on-screen buttons. He fixed the system by pushing back the date the spreadsheets would stop working again, the government said.
When inevitably, the system glitched, Siemens would be forced to call back Mr Tinley to fix the issue.
He appeared to be getting away with it. His duplicity was only discovered when he was forced to give his former colleagues the passwords that protected the system’s code whilst he was on holiday.
Was it all worth it? Probably not. Mr Tinley was forced to plead guilty to a charge of intentional damage to a protected computer; a crime that could cost him up to 10 years in prison and up to a $250,000 fine.
It doesn’t look great for this errant contractor…it is not unusual for programmers who plant logic bombs to serve jail time. In 2008, a system administrator was sentenced to 30 months over his failed logic bomb. He pleaded guilty to planting the bomb, which was designed to delete company data after he left the position. Additionally, in 2018, an Atlanta judge sentenced a database programmer to two years in prison after he pleaded guilty to planting a logic bomb in the US Army’s payroll databases. This bomb really DID go off – literally. It deleted data that prevented US Army reservists from being paid and deployed on time. The army spent $2.6 million to investigate and repair their systems. They did manage to restore all the data, but the man was ordered to pay $1.5 million in restitution and received jail time.
To add insult to injury, Tinley’s lawyers argued he never made any money by being hired to go in and fix the spreadsheets. Instead, he was motivated by his desire to protect his proprietary work, they claim.
It would appear then, that crime really doesn’t pay!