How ByteDance Became the World’s Most Valuable Startup

Roger Chen and Rui Max from Harvard Business Review explain how ByteDance became the world’s most valuable startup. What’s the secret? According to the editors, it’s the company’s shared-service platform, or SSP, which it uses to power innovation. From the report: Bytedance uses its SSP platform differently from most companies. The company’s product teams or units don’t control their own operating resources. Instead, many common business, technology, and operating functions (among them HR and legal) are centralized and organized into corresponding teams. The teams are highly specialized, so that the right people can be found and flexibly deployed as needed to each new venture. Cloud and shared operational tools, some of which have been developed in house, allow ByteDance to maintain this seemingly complex organizational setup. Product and related teams still focus on serving customer needs, but they rely on different SSP teams to accelerate development and growth. For example, when ByteDance tasks a new venture team with investigating user needs and market opportunities, the team can go to the user-research specialists at the SSP for data support, saving time on market analysis. In other companies, these tasks are undertaken by the product team, which is rarely best equipped for such information gathering. Subsequently, when a use case has been identified that justifies developing a new app or product feature, the product team is paired with engineers at the SSP level to develop the new product or feature.

In some cases, product teams customize existing technologies that have already been developed by the SSP. Algorithms are a case in point. Product teams at ByteDance work with SSP algorithm engineers to fine-tune their enormously powerful recommendation engines. The SSP has also brought together other important teams: user-growth teams, which help identify and acquire desired users; content teams, which establish partnerships to acquire new content; analytics teams, which help to develop deeper user insights; and sales teams, which drive monetization. As expected, because so many capabilities have been centralized into this large SSP, the actual product teams tend to be small and focused, especially in the exploration stage. Douyin, for example, began with just a handful of employees, and the education team began with just two. Importantly, the relationship between the SSP and market-facing teams is symbiotic and mutually beneficial. It’s this virtuous loop of continued discovery and improvement that has enabled ByteDance’s success.

Relying on its SSP, ByteDance has developed unique innovation and growth strategies. These strategies have five main characteristics: [broad exploration, rapid iteration, selective focus, maximum-capability cross-pollination, and productizing platform services]. […] ByteDance’s SSP strategy — accelerate new projects by providing instant access to best in class technology and operations — has been so successful that one would expect many other companies to have embraced it. Yet few companies have managed to replicate ByteDance’s success with the strategy. Why? Because they have not put in the organizational enablers that helped ByteDance overcome fiefdom mindsets, which inhibit collaboration. Three of these organizational enablers are particularly important: [OKR system, explicitly flattened hierarchy, and data-driven culture]. […]

ByteDance’s SSP-based innovation strategy has clearly played a key role in its first decade of explosive growth. It has allowed the company to incubate rapidly and broadly and to scale efficiently, by using centralized but flexibly deployed technical and operational stacks. This strategy has served the company well in part because of the similarity among its various algorithm-driven products. ByteDance is now exploring other product categories and is refining its strategy to be more suitable for its evolving organizational model and processes, but no matter how the company evolves, its SSP-based innovation strategy is sure to play an important role.

Read more of this story at Slashdot.

ACM, Ethics, and Corporate Behavior

theodp writes: In the just-published March 2022 issue Communications of the ACM, former CACM Editor-in-Chief Moshe Y. Vardi takes tech companies — and their officers and technical leaders — to task over the societal risk posed by surveillance capitalism in “ACM, Ethics, and Corporate Behavior.” Vardi writes: “Surveillance capitalism is perfectly legal, and enormously profitable, but it is unethical, many people believe, including me. After all, the ACM Code of Professional Ethics starts with ‘Computing professionals’ actions change the world. To act responsibly, they should reflect upon the wider impacts of their work, consistently supporting the public good.’ It would be extremely difficult to argue that surveillance capitalism supports the public good.” “The biggest problem that computing faces today is not that AI technology is unethical — though machine bias is a serious issue — but that AI technology is used by large and powerful corporations to support a business model that is, arguably, unethical. Yet, with the exception of FAccT, I have seen practically no serious discussion in the ACM community of its relationship with surveillance-capitalism corporations. For example, the ACM Turing Award, ACM’s highest award, is now accompanied by a prize of $1 million, supported by Google.”

“Furthermore, the issue is not just ACM’s relationship with tech companies. We must also consider how we view officers and technical leaders in these companies. Seriously holding members of our community accountable for the decisions of the institutions they lead raises important questions. How do we apply the standard of ‘have not committed any action that violates the ACM Code of Ethics and ACM’s Core Values’ to such people? It is time for us to have difficult and nuanced conversations on responsible computing, ethics, corporate behavior, and professional responsibility.”

Read more of this story at Slashdot.

TVAddons’ Adam Lackman Admits TV Show Piracy, Agrees To Pay $19.5 Million

In 2017, Bell Canada, TVA, Videotron, and Rogers teamed up in a lawsuit against the operator of TVAddons, the largest repository of Kodi add-ons. The legal action proved extremely controversial but now, after many twists and turns, the matter is now over. As part of a consent judgment (PDF), TVAddons’ founder [Adam Lackman] has admitted liability and agreed to pay a cool US$19.5 million in damages. TorrentFreak reports: In a letter dated February 18, 2022, the media companies and Lackman told the Federal Court that they had resolved their differences by agreeing to a consent judgment. That was reviewed and issued by Justice Rochester, who laid out the agreed terms in her judgment handed down February 22, 2022. Lackman admits to communicating TV shows owned by the plaintiffs to the public, including by directly or indirectly participating in the “development, hosting, distribution or promotion of Kodi add-ons that provide users with unauthorized access” to the plaintiffs’ TV shows, contrary to sections 3(1)(f) and 27(1) of the Copyright Act. The TVAddons founder further admits that he made the TV shows available to the public in a manner that provided access “from a place and at a time individually chosen by them” and induced and authorized users of the infringing add-ons to “initiate acts of infringement of the Plaintiffs’ right to communicate the Plaintiffs Programs to the public by telecommunication,” again by developing, hosting, distributing or promoting Kodi add-ons.

The Federal Court issued a permanent injunction to restrain Lackman (and anyone acting with him, under his authority, or in association) from communicating the plaintiffs’ content to the public in any way, including via the development or distribution of infringing add-ons such as the ‘FreeTelly’ and ‘Indigo’ tools. The terms of the injunction are lengthy and comprehensive, leaving no doubt that TVAddons and all related tools and services are now dead, with Lackman unable to do anything remotely similar in the future.

“THIS COURT ORDERS the Defendant Mr. Lackman to pay the Plaintiffs the amount of twenty-five million dollars ($25,000,000) in the form of a lump sum for damages, profits, punitive and exemplary damages, and costs,” Justice Rochester writes. The judgment is in Canadian dollars but for reference, that’s currently around US$19.5 million. The judgment also authorizes the bailiffs and independent supervising solicitor (with the assistance of computer forensics experts) to transfer the evidence obtained during the search of June 2017 to the media companies. Exactly what data was seized is currently unclear but it is likely to be sensitive, particularly if the trove includes user data and/or information about Kodi add-on developers. Finally, it appears the media companies will also be taking control of “login credentials, accounts, domains, subdomains and servers” in order to bring this years-long battle to a conclusion. Adam Lackman announced his relief on Twitter, noting that “It wasn’t the outcome I had hoped for, but an outcome nonetheless.”

Read more of this story at Slashdot.

Volvo Ditches PCs and Paper In Favor of iPhone and Apple Watch

Volvo has unleashed a big improvement in customer satisfaction after equipping its 1,500 service engineers with an Apple Watch to use during their day. What, on the face of it, seems a small change reflects extensive cultural change across the company, which is actively engaged in digital transformation across its business. Computerworld’s Jonny Evans reports: Volvo has equipped its engineers (Personal Service Technicians) with an Apple Watch and iPhone (running the Volvo Service app) to help them work more efficiently than before. The company’s primary focus is to improve customer service, as it recognizes that technicians are the main point of customer contact across the life of the Volvo they drive. So, how can an Apple Watch in a garage improve customer service?

– In use, the engineer will receive a Notification when a customer arrives at the garage with their car.
– The watch will show the customer’s name, relevant notes, and car details.
– During the repair, engineers can access information — and once the repair is complete, they can directly call the customer to tell them.
– They can also schedule and make a subsequent follow-up call.

The benefit is that with all this information being made available through the Watch (and accompanying iPhone app), engineers don’t need to use printed records, or access a PC to stay up to date. That’s not only time-consuming, but learning how to use these systems takes up time. The company told me it took up to 6 months to train new recruits on the 15 different IT systems Volvo used before. Now, thanks to smart analysis and smart integration of legacy systems, what technicians need to know is always with them. The result is that paperwork doesn’t disappear, technicians/engineers can stay focused, essential customer contact records aren’t lost and engineers always have clarity and purpose. It all sounds so simple. It should sound simple. But it isn’t simple. […]

The project is already generating positive results. The company told me that 80% of technicians who use the app have increased their total customer satisfaction scores. Volvo also cites a 30% increase in post-service follow up calls and emails to customers, thanks to the tech pushing complex processes out of the way. Digitalization Director Markus Lundstrom said: “With the Volvo Service app we’re connecting people through technology. At one workshop, customers report a 37% improvement in the ability to access their Personal Service Technician.” The company also reported a 40% decrease in paper printouts. Volvo is also seeing the technicians use their new kit to get other tasks done. “Some of our teams use the Walkie-Talkie feature to communicate with each other across the facility,” they said.

Read more of this story at Slashdot.

Methane Emissions From the Energy Sector Are 70% Higher Than Official Figures: IEA

New submitter Klaxton shares an excerpt from a new report released today by the International Energy Agency (IEA): Global methane emissions from the energy sector are about 70% greater than the amount national governments have officially reported, according to new IEA analysis released today, underlining the urgent need for enhanced monitoring efforts and stronger policy action to drive down emissions of the potent greenhouse gas. Methane is responsible for around 30% of the rise in global temperatures since the Industrial Revolution, and quick and sustained emission reductions are key to limiting near-term warming and improving air quality. Methane dissipates faster than carbon dioxide (CO2) but is a much more powerful greenhouse gas during its short lifespan, meaning that cutting methane emissions would have a rapid effect on limiting global warming.

The energy sector accounts for around 40% of methane emissions from human activity, and this year’s expanded edition of the IEA’s Global Methane Tracker includes country-by-country emissions from coal mines and bioenergy for the first time, in addition to continued detailed coverage of oil and natural gas operations. Methane emissions from the energy sector grew by just under 5% last year. This did not bring them back to their 2019 levels and slightly lagged the rise in overall energy use, indicating that some efforts to limit emissions may already be paying off. “At today’s elevated natural gas prices, nearly all of the methane emissions from oil and gas operations worldwide could be avoided at no net cost,” said IEA Executive Director Fatih Birol. “The International Energy Agency has been a longstanding champion of stronger action to cut methane emissions. A vital part of those efforts is transparency on the size and location of the emissions, which is why the massive underreporting revealed by our Global Methane Tracker is so alarming.”

If all methane leaks from fossil fuel operations in 2021 had been captured and sold, then natural gas markets would have been supplied with an additional 180 billion cubic meters of natural gas. That is equivalent to all the gas used in Europe’s power sector and more than enough to ease today’s market tightness. The intensity of methane emissions from fossil fuel operations range widely from country to country: the best performing countries and companies are over 100 times better than the worst. Global methane emissions from oil and gas operations would fall by more than 90% if all producing countries matched Norway’s emissions intensity, the lowest worldwide.

Read more of this story at Slashdot.

AMD Is Now Worth More Than Rival Intel

Hmmmmmm shares a report from Yahoo Finance: AMD’s market cap currently stands at $188 billion after shares rose nearly 2% in Tuesday’s session. Intel’s market cap is $182 billion. That marks the second time in a week AMD’s market value has climbed above Intel — the first time it happened was a week ago. Followers of this battle may not be surprised to see this one happen (and seeing it continue from here) for several reasons. First, AMD has been winning the battle on Wall Street for sexier investment thesis. AMD last week closed on its $35 billion acquisition for Xilinx. Secondarily, AMD has flat out posted better financials than Intel (for some time) as it has gained market share in key areas (notably in servers). AMD’s sales and profits rose 68% and 117%, respectively in 2021. The company outlined 31% revenue growth for 2022 and gross profit margins of 51%. Intel’s 2021 sales and earnings increased 2% and 7%, respectively. The company sees sales in 2022 rising about 2%. Profits are expected to drop 36% as Intel further builds out its chip-making capacity.

Read more of this story at Slashdot.

Ask Slashdot: Is It Time To Replace File Systems?

DidgetMaster writes: Hard drive costs now hover around $20 per terabyte (TB). Drives bigger than 20TB are now available. Fast SSDs are more expensive, but the average user can now afford these in TB capacities as well. Yet, we are still using antiquated file systems that were designed decades ago when the biggest drives were much less than a single gigabyte (GB). Their oversized file records and slow directory traversal search algorithms make finding files on volumes that can hold more than 100 million files a nightmare. Rather than flexible tagging systems that could make searches quick and easy, they have things like “extended attributes” that are painfully slow to search on. Indexing services can be built on top of them, but these are not an integral part of the file system so they can be bypassed and become out of sync with the file system itself.

It is time to replace file systems with something better. A local object store that can effectively manage hundreds of millions of files and find things in seconds based on file type and/or tags attached is possible. File systems are usually free and come with your operating system, so there seems to be little incentive for someone to build a new system from scratch, but just like we needed the internet to come along and change everything we need a better data storage manager.
See Didgets for an example of what is possible.
In a Substack article, Didgets developer Andy Lawrence argues his system solves many of the problems associated with the antiquated file systems still in use today. “With Didgets, each record is only 64 bytes which means a table with 200 million records is less than 13GB total, which is much more manageable,” writes Lawrence. Didgets also has “a small field in its metadata record that tells whether the file is a photo or a document or a video or some other type,” helping to dramatically speed up searches.

Do you think it’s time to replace file systems with an alternative system, such as Didgets? Why or why not?

Read more of this story at Slashdot.

Inside ‘Project Tinman’: Peloton’s Plan To Conceal Rust In Its Exercise Bikes

Dubbed internally as “Project Tinman,” executives at Peloton worked to conceal a build-up of rust on some exercise machines (Warning: source may be paywalled; alternative source) that were sent to customers instead of returned to the manufacturer. “The project was first revealed in FT Magazine last week but eight current and former Peloton employees across four US states have provided further details on the operation,” reports the Financial Times. Here’s an excerpt from the report: They described the plan as a nationwide effort to avoid yet another costly recall just months after the company’s most tragic episode — the death of a child due to the design of its treadmill. Internal documents seen by the FT showed that Tinman’s “standard operating procedures” were for corrosion to be dealt with using a chemical solution called “rust converter,” which conceals corrosion by reacting “with the rust to form a black layer.” Employees said the scheme was called Tinman to avoid terms such as “rust” that executives decided were out of step with Peloton’s quality brand.

Insiders were also angered about enacting a plan that they argued cut across Peloton’s supposed focus on its users, who are called “members” to evoke a sense that buyers are more than customers and part of a broader community. Tinman also put a spotlight on the company’s quality control process versus meeting aggressive sales targets in the search for growth. Peloton said the issue affected at least 6,000 bikes and that 120 staff had undertaken “rigorous testing” on the devices to conclude the rust — which it described as “cosmetic oxidation” — had “no impact on a bike’s performance, quality, durability, reliability, or the overall member experience.”

Read more of this story at Slashdot.

Qualcomm Will Support AV1 Video Codec In 2023, Report Says

Protocol reports that Qualcomm will finally jump on the AV1 video codec bandwagon next year. Ars Technica reports: AV1 is the web’s next open, royalty-free video codec, and widespread adoption will require hardware support from the world’s chip vendors. Qualcomm’s 2022 flagship SoC, the Snapdragon 8 Gen 1 chip, doesn’t support AV1. Samsung’s Exynos 2200 managed to ship the video codec this year in international versions of the Galaxy S22, while the MediaTek Dimensity 1000 SoC has been shipping in phones for over a year now with AV1 support. Apple is a founding member of the AV1 Alliance, but its devices also don’t support the codec yet.

The report says Qualcomm’s “upcoming flagship Snapdragon mobile processor” — model number “SM8550” — will support AV1. That would probably be called the “Snapdragon 8 Gen 2” SoC, due out in 2023. Wide adoption of AV1 seems inevitable, though it is taking a while. The codec is a successor to Google’s VP8 and VP9 codecs and is being built by the Alliance for Open Media. The alliance’s lineup is a who’s who of tech companies, with founding members like Amazon, Apple, ARM, Facebook, Google, Intel, Microsoft, Mozilla, Netflix, Nvidia, and Samsung. Netflix and Google’s YouTube are both making AV1 support “a requirement” for future products that want to support either video service. That should motivate just about every hardware and software vendor out there to get the job done. Aside from being open source and royalty-free, the report notes that the newer AV1 codec also has the benefit of being 30% more efficient than H.265.

Read more of this story at Slashdot.

Microsoft Details ‘Planet-Scale’ AI Infrastructure Packing 100,000+ GPUs

Microsoft has revealed it operates a planet-scale distributed scheduling service for AI workloads that it has modestly dubbed “Singularity.” The Register reports: Described in a pre-press paper [PDF] co-authored by 26 Microsoft employees, Singularity’s aim is described as helping the software giant control costs by driving high utilization for deep learning workloads. Singularity achieves that goal with what the paper describes as a “novel workload-aware scheduler that can transparently preempt and elastically scale deep learning workloads to drive high utilization without impacting their correctness or performance, across a global fleet of AI accelerators (e.g., GPUs, FPGAs).”

The paper spends more time on the scheduler than on Singularity itself, but does offer some figures to depict the system’s architecture. An analysis of Singularity’s performance mentions a test run on Nvidia DGX-2 servers using a Xeon Platinum 8168 with two sockets of 20 cores each, eight V100 Model GPUs per server, 692GB of RAM, and networked over InfiniBand. With hundreds of thousands of GPUs in the Singularity fleet, plus FPGAs and possibly other accelerators, Microsoft has at least tens of thousands of such servers! The paper focuses on Singularity’s scaling tech and schedulers, which it asserts are its secret sauce because they reduce cost and increase reliability.

The software automatically decouples jobs from accelerator resources, which means when jobs scale up or down “we simply change the number of devices the workers are mapped to: this is completely transparent to the user, as the world-size (i.e. total number of workers) of the job remains the same regardless of the number of physical devices running the job.” That’s possible thanks to “a novel technique called replica splicing that makes it possible to time-slice multiple workers on the same device with negligible overhead, while enabling each worker to use the entire device memory.” […] “Singularity achieves a significant breakthrough in scheduling deep learning workloads, converting niche features such as elasticity into mainstream, always-on features that the scheduler can rely on for implementing stringent SLAs,” the paper concludes.

Read more of this story at Slashdot.