Monthly Archives: May 2022

Google has released its annual diversity report, and the company says it made some positive progress over the past year. The company saw its largest increases in representation of Black and Latinx Googlers in the US ever at 20 percent and 8 percent respectively year over year, according to chief diversity officer Melonie Parker. Google also reported improved leadership representation of Black, Latinx and Native American employees by 27 percent, Parker says.

But some data shows there is still more work to be done. The companys US workforce is 33.5 percent women and 66.5 percent men, numbers that are only slightly different than the 32.2 percent women and 67.8 percent men reported in 2021. And in its 2022 report, Google said 48.3 percent of its US workers are white, while 43.2 percent are Asian, 6.9 percent are Latinx, 5.3 percent are Black, and 0.8 percent are Native American.

The company does provide a vast amount of information about its statistics, and you can dig into all of it on the 2022 Diversity Annual Report website and in the PDF version of the report. And Google appears to be committed to moving the needle, saying that as we continue to build a more inclusive and representative Google, well hold ourselves accountable in how we work to make our goals a reality.

Google came under scrutiny for its diversity policies after it fired prominent AI ethicist Timnit Gebru, who is Black, in December 2020. The company made changes to its diversity and research policies two months later.

View original post here:

Google reports increased Black and Latinx representation in the US - The Verge

CEO of Alphabet and Google Sundar Pichai

Mateusz Wlodarczyk | Nurphoto | Getty Images

A new bipartisan proposal takes aim at Google and would force it to break up its digital advertising business if passed.

The Competition and Transparency in Digital Advertising Act was introduced Thursday by a group of key senators on the Judiciary subcommittee on antitrust: the ranking member and chair, Sens. Mike Lee, R-Utah, and Amy Klobuchar, D-Minn., as well as Sens. Ted Cruz, R-Texas, and Richard Blumenthal, D-Conn.

Advertising is a huge part of parent company Alphabet's business. In Q1, Alphabet reported $68.01 billion in revenue, $54.66 billion of which was generated by advertising up from $44.68 billion the year prior.

The bill would ban companies that process more than $20 billion annually in digital ad transactions from participating in more than one part of the digital ad process, according to The Wall Street Journal, which first reported the news.

Google infamously has a hand in multiple steps of the digital ad process, a business that has become the focus of a state-led antitrust lawsuit against the company. Google runs an auction, or exchange, where ad transactions are made and also runs tools to help companies sell and buy ads. If the new legislation passed, it would have to choose in which part of the business it would want to remain.

"When you have Google simultaneously serving as a seller and a buyer and running an exchange, that gives them an unfair, undue advantage in the marketplace, one that doesn't necessarily reflect the value they are providing," Lee told the Journal in an interview. "When a company can wear all these hats simultaneously, it can engage in conduct that harms everyone."

"Advertising tools from Google and many competitors help American websites and apps fund their content, help businesses grow, and help protect users from privacy risks and misleading ads," a Google spokesperson said in a statement. "Breaking those tools would hurt publishers and advertisers, lower ad quality, and create new privacy risks. And, at a time of heightened inflation, it would handicap small businesses looking for easy and effective ways to grow online. The real issue is low-quality data brokers who threaten Americans' privacy and flood them with spammy ads. In short, this is the wrong bill, at the wrong time, aimed at the wrong target."

The coalition behind the bill underscores the way support for reining in tech power through antitrust reform cuts across ideological lines. It's also notable that Lee, the top Republican on the subcommittee, led the bill, given he has opposed some of the other antitrust reforms on the table from Klobuchar and others.

Klobuchar, as chair of the subcommittee, has led an effort to get competition reforms passed this year. So far, two major bills have stood out as having a fighting chance of becoming law if Congress moves on them in time: the American Innovation and Choice Online Act, which would prevent dominant platforms from favoring their own products over those of competitors that rely on their services, and the Open App Markets Act, which would have a similar impact but focuses on app stores like those from Apple and Google. Lee supported the latter, but not the former, during committee votes.

Subscribe to CNBC on YouTube.

WATCH: Here's why some experts are calling for a breakup of Big Tech after the House antitrust report

Continued here:

New bipartisan bill would force Google to break up its ad business - CNBC

Google's eye-catching, 1.1 million-square-foot campus on the edge of the bay is finally complete and open to employees, marking a major expansion of the tech company's office footprint on the Peninsula.

The Bay View campus, located on NASA Ames property off of Moffett Boulevard, will be able to house roughly 4,000 employees across two large and uniquely designed office buildings. The sloped, pavilion-like roofing is clad with large, prismatic glass shingles that generate solar power for the facility, layered in a pattern that the company calls a "dragonscale" solar skin.

Inside, the offices are filled with natural lighting and vibrant colors, with lighthearted-themed areas, or "districts," such as the "Turkey Terrace" and the "Campfire Corner," the latter with meeting rooms modeled to look like tents. The so-called Plankton Palace has a light display shimmering up a stalk of faux seaweed.

The Bay View campus has been in the works for close to a decade, with plans solidifying in 2017 following a long-term lease with NASA Ames. Google's subsidiary, Planetary Ventures, inked a contract with the agency to lease 42 acres for $3.65 million in annual rent.

The company broke ground in 2017, and as of last week began inviting employees to trickle in albeit with plenty of construction workers still working on the finishing touches.

In a tour of the campus Monday, Dave Radcliffe, Google's vice president of real estate and workplace services, said this is a big moment for the company. Most of Google's offices have been inherited or leased, but Bay View marks a rare opportunity to build something new and purposefully built to support the company's workforce.

"This is the first time ... we've been able to move Googlers into a building that we were able to design from the ground up, and so it's an exciting time for us," Radcliffe said.

Even before COVID-19 abruptly forced tech companies in the Bay Area to work from home, the design for Bay View took into account telecommuting and the reality that Google's leadership, candidly, cannot precisely know what the workplace will look like in 20 or 50 years, Radcliffe said. Instead, the company sought to create a flexible workplace environment that can transform numerous times in order to meet new demands.

The first floor, with its art exhibits, food services, couches and giant camel statue, is meant to be an "activation space," Radcliffe said. It's where employees can come in a few times a week for "intentional collaboration" with colleagues, the kind of thing that can be tricky to do from home. The second level is closer to a traditional office setting where Googlers can get work done in a distraction-free environment, but even then there are no defined walls, cubicles or corner offices.

The hybrid work schedule means Bay View needs to accommodate team-focused work, said Michelle Kaufmann, Google's director of real estate and workplace services, research and development. The floor plan, desks and furniture are all movable and interchangeable, with a mix of both group desks and quiet enclosed booths.

"Rather than have the traditional office, which was fixed desks and closed meeting rooms kind of a one-size-fits-all for everyone this new workplace is really about having a much wider range of space types for the activities people are going to be coming back to do," Kaufmann said.

While Bay View's sloped roofs are designed to accommodate 90,000 solar panels and capture energy at all hours of the day, the high steel canopy is also angled to let in plenty of daylight without causing any glare. The opera hall-shaped ceiling, along with acoustic decking and soft materials throughout the building, strike a balance that keeps the office from sounding too loud or too quiet, Kaufmann said.

Google is also touting the new campus' environmentally friendly design, including the largest geothermal pile system in North America. The system involves pipes that run 80 feet underground that can be used to transfer excess heat from the building, which can then be stored for future use during the winter, said Asim Tahir, director of energy and carbon for the company's real estate and workplace services.

The geothermal system means Google will save about 5 million gallons of water each year, and can completely ditch natural gas for heating the building, which is all-electric. The plan is to fully eliminate Bay View's carbon footprint.

"This building will operate with carbon-free energy 90% of the time, and we're hoping through operational improvements we'll over the years get to the full 100%," Tahir said.

A large portion of the leased 42-acre property remains open space, including meadows and marshes, creating a buffer between the newly opened tech campus and the bay coastline. North of the office buildings is a large parking garage next door to multiple buildings with a total of 220 suites similar to motel rooms, built for people visiting Google who need to stay for extended periods.

Bay View is one of multiple Google office developments underway in and around Mountain View. The company's 595,000-square-foot Charleston East campus is still under construction and is expected to finish next year, and the Mountain View City Council approved a third office proposal, the Google Landings project, in 2020. Charleston East is expected to house up to 2,700 Google employees.

Continued here:

Google opens the doors on its massive Bay View campus next to NASA Ames - Palo Alto Online

Indian users of Google Messages are facing a deluge of ads within Androids default messaging app, 9to5Google reports. Users are reporting being sent multiple ads a day, even if they enable the apps Spam Protection feature, and blocking and reporting the offending accounts doesnt appear to prevent ads from different companies appearing. Many of the worst offenders appear to be ads for personal loans, gambling companies, and life insurance.

As 9to5Google notes, companies appear to be abusing an RCS feature designed to let businesses contact their customers with rich, interactive media. Examples listed on Googles site include sending customers QR-code tickets, or letting them make online orders. Instead, some accounts that Google identifies as Verified Business are sending users in India a barrage of ads.

Its unclear how widespread the issue currently is, but Indian-based Ishan Agarwal whose recent tweets have drawn attention to the issue tells The Verge that for him the problem has been growing worse over the course of 2022, and that he receives an average of between two and five ads a day now. Android Police reports that the problem has been around for about a year, and complaints appear to be coming overwhelmingly from users based in India.

Although blocking and reporting accounts prevents specific companies from continuing to send you spam, Android Police reports that the only way to stop receiving ads entirely is to disable Messages support for RCS (under chat features in settings). Its a pretty damning measure to have to take considering how hard Google has been pushing RCS as a successor to SMS and MMS.

Google did not respond to The Verges request for comment.

Excerpt from:

Google Messages users in India are reportedly drowning in ads - The Verge

The Department of Homeland Security has paused work on its recently announced Disinformation Governance Board, according to The Washington Post. The board's director, Nina Jankowicz, has also resigned following an unrelenting stream of harassment. The Post first reported Jankowicz's resignation, and Protocol has since confirmed it.

The board's rollout was shoddy even by DHS Sec. Alejandro Mayorkas' own admission. From the outset, DHS revealed next to nothing about the board's goals or its authorities, leading to concerns that this new entity might be surveilling social media and deciding what does and doesn't constitute disinformation. Conservatives also quickly pounced on Jankowicz, accusing her of being a partisan hack. According to the Post, DHS forbid Jankowicz from saying anything publicly in her own defense.

In truth, the goal of the board, Mayorkas explained far too late, was to do the opposite of what it was being accused of. Throughout DHS, agencies are already working on ways to combat misinformation and disinformation. The purpose of the board was to coordinate those efforts and ensure that they weren't crossing lines with regard to free speech and privacy.

But the decision to do that in a public way rather than in a private audit drew undue scrutiny to the effort. As one source familiar with DHS' plans recently told Protocol, Having a very large governance board and a really big, public rollout for it with a very well-known person in this space very publicly leading it, that probably drove their risk up a little more than it needed to."

In a statement, DHS spokesperson Angelo Fernandez said the board has been "grossly and intentionally mischaracterized," and confirmed that the Homeland Security Advisory Council is now leading a review of the board in hopes of answering two questions. "First, how can the Department most effectively and appropriately address disinformation that poses a threat to our country, while protecting free speech, civil rights, civil liberties, and privacy. Second, how can DHS achieve greater transparency across our disinformation-related work and increase trust with the public and other key stakeholders," Fernandez said.

The final recommendations are due within 75 days, during which time, the board's work will be paused.

Among the board's critics were not just the usual suspects in conservative circles, but also platform regulation scholars and legal experts who feared that the well had already been poisoned. Elon Musk and Jeff Bezos also got in on the act, with Bezos tweeting this week that the "newly created Disinformation Board should review" one of President Biden's tweets about taxing the wealthy to fix inflation.

This story has been updated to include comments from DHS.

Here is the original post:

Apple lost talent due to its return-to-office policy. Google is benefitting. - Protocol

Google announced a new initiative Tuesday aimed at securing the open-source software supply chain by curating and distributing a security-vetted collection of open-source packages to Google Cloud customers.

The new service, branded Assured Open Source Software, was introduced in a blog post from the company. In the post, Andy Chang, group product manager for security and privacy at Google Cloud, pointed to some of the challenges of securing open-source software and stressed Googles commitment to open source.

There has been an increasing awareness in the developer community, enterprises, and governments of software supply chain risks, Chang wrote, citing last years major log4j vulnerability as an example. Google continues to be one of the largest maintainers, contributors, and users of open source and is deeply involved in helping make the open source software ecosystem more secure.

Per Googles announcement, the Assured Open Source Software service will extend the benefits of Googles own extensive software auditing experience to Cloud customers. All open-source packages made available through the service are also used internally by Google, the company said, and are regularly scanned and analyzed for vulnerabilities.

Currently, a list of the 550 major open-source libraries being continuously reviewed by Google is available on GitHub. While these libraries can all be downloaded independently of Google, the Assured OSS program will see audited versions distributed through Google Cloud mitigating against incidents where developers intentionally or unintentionally corrupt widely used open-source libraries. At present, this service is in early access mode and is expected to be made available for wider customer testing in Q3 2022.

The announcement from Google comes as part of an industry-wide drive to improve the security of the open-source software supply chain and one that has also been supported by the Biden administration.

In January, a group of some of the nations largest tech companies met with representatives of federal agencies including the Department of Homeland Security and the Cybersecurity and Infrastructure Security Agency to discuss open-source software security in the wake of the log4j bug. Since then, a recent meeting of the companies involved resulted in a pledge of more than $30 million in funding to boost open-source software security.

Besides contributing funding, Google is also putting engineering hours toward keeping the supply chain secure. The company recently announced the formation of an Open Source Maintenance Crew that would work with the maintainers of popular libraries to improve security.

Read the original here:

Google will start distributing a security-vetted collection of open-source software libraries - The Verge

Why does Google need another database, and why in particular does it need to introduce a version of PostgreSQL highly tuned for Googles datacenter-scale disaggregated compute and storage?

It is a good question in the wake of the launch of the AlloyDB relational database last week at the Google I/O 2022 event.

The name Google is practically synonymous with large scale data storage and manipulation in myriad forms. The company created the MapReduce technique for querying unstructured data that inspired Hadoop, the BigTable NoSQL database, the Firestore NoSQL document database, and the Spanner geographically distributed relational SQL database. These tools were used internally at first, and then put on Google Cloud as the Dataproc, Cloud BigTable, and Cloud Spanner services.

Relational databases are back in vogue, due in part by Google showing that a true relational database is can scale with the advent of Spanner. And to try to encourage adoption of Spanner on the cloud, Google last year created a PostgreSQL interface for Spanner that makes it look and feel like that increasingly popular open source database. This is important because PostgreSQL has become the database of choice in the aftermath of Oracle buying Sun Microsystems in early 2010 and taking control of the much more widely used open source MySQL relational database that Sun itself took control of two years earlier.

The reason why Google needs a true version of PostgreSQL running in the cloud is that it needs to help enterprise customers who are stuck on IBM DB2, Oracle, and Microsoft SQL Server relational databases as their back-end datastores for their mission-critical systems of record get off those databases and not only move to a suitable PostgreSQL replacement, but to also make the move from on-premises applications and databases to the cloud.

That is the situation in a nutshell, Andi Gutmans, vice president and general manager of databases at the search engine, ad serving, and cloud computing giant.

Google has been an innovator on data, and we have had to innovate because we have had these billion user businesses, says Gutmans. But our strength has really been in cloud native, very transformative databases. But Google Cloud has accelerated its entrance into mainstream enterprises we have booming businesses in financial services, manufacturing, and healthcare, and we have focused on heritage systems and making sure that lifting and shifting applications into the cloud. Over the past two years, we have focused on supporting MySQL, PostgreSQL, SQL Server, Oracle, and Redis, but the more expensive, legacy, and proprietary relational databases like SQL Server and Oracle have unfriendly licensing models that really force them into one specific cloud. And we continue to get requests to help customers modernize off legacy and proprietary databases to open source.

The AlloyDB service is the forklift that Google created for this lift and shift, and dont expect for Google to open up all of the goodies it has added to PostgreSQL because these are highly tuned for Google own Colossus file system and its physical infrastructure. But, it could happen in the long run, just as Google took its Borg infrastructure and container controller and open sourced a variant of it as Kubernetes.

As we have pointed out before, the database, not the operating system and certainly not the server infrastructure, is arguably the stickiest thing in the datacenter, and companies make database decisions that span one or two decades and sometimes more. So having a ruggedized, scalable PostgreSQL that can span up to 64 vCPUs running on Google Cloud is important, as will be scaling it to 128 vCPUs and more in the coming years, which Gutmans says Google is working on.

But that database stickiness has to do with databases implementing different dialects of the SQL query language, and also having different ways of creating and embedding stored procedures and triggers within those databases. Stored procedures and triggers essentially embed elements of an application within the database rather than outside of it for reuse and performance, but there is no universally accepted and compatible way to implement these functions, and this has created lock in.

That is one of the reasons why Google acquired CompilerWorks last October. CompilerWorks has created a tool called Transpiler, which can be used to convert SQL, stored procedures, and triggers from one database to another. As a case in point, Gutmans says that Transpiler, which is not yet available as a commercial service, can convert about 70 percent of Oracles PL SQL statements to another format, and that Google Cloud is working with one customer that has 4.5 million lines of PL SQL code that it has to deal with. To help with database conversions, Google has tools to do data replication and scheme conversion, and has provided additional funding where they can get human help from systems integrators.

AlloyDB is not so much a distribution of PostgreSQL as it is a storage layer designed to work with Googles compute and storage infrastructure.

And while Google has vast scale for supporting multi-tenant instances of PostgreSQL, you will not that it doesnt have databases that span hundreds or even thousands of threads. IBMs DB2 on Power10 processors, which has 1,920 threads in a 240-core, 16-socket system with SMT8 simultaneous multithreading turned on, can grab any thread that is not being used by AIX or Linux and use it to scale the database, just to give you a sense of what real enterprise scale is for relational databases. But we are confident that is Google needed to create a 2,000-thread implementation of PostgreSQL, it could do it with NUMA clustering across its network and other caching techniques or by installing eight-way X86 servers that would bring 896 threads to bear with 56-core Sapphire Rapids Xeon SPs and 1,204 threads to bear with 64-core Granite Rapids Xeon SPs. (Again, the operating system would eat a bunch of these threads, but certainly not as much as the database could.) The latter approach using NUMA-scaled hardware is certainly easier when it comes to scaling AlloyDB, but it also means adding specialized infrastructure that is really only suitable for databases. And that cuts against the hyperscaler credo of using cheap servers and only a few configurations of them at that to run everything.

So what exactly did Google do to PostgreSQL to create AlloyDB? Google took the PostgreSQL storage engine and built what Gutmans called a cloud native storage fleet that is linked to the main PostgreSQL node, database logging and point in time recovery for the database runs on this distributed storage engine. Google also did a lot of work on the transaction engine at the heart of PostgreSQL and as a result, Google is able to get complete linear scaling up to 64 virtual cores on its Google Cloud infrastructure. Google has also added an ultra fast cache inside of PostgreSQL, and if there is a memory miss in the database, this cache can bring data into memory with microsecond latencies instead of the millisecond latencies that other caches have.

In initial tests running the TPC-C online transaction processing benchmark against AlloyDB, Gutmans says that AlloyDB was 4X faster than open source PostgreSQL and 2X faster than the Aurora relational database (which has a PostgreSQL compatible layer on top) from Amazon Web Services.

And to match the high reliability and availability of those legacy databases such as Oracle, SQL Server, and DB2, Google has a 99.99 percent uptime guarantee on the AlloyDB service, and this uptime importantly includes maintenance of the database. Gutmans says that other online databases only count unscheduled and unplanned downtime in their stats, not planned maintenance time. Finally, AlloyDB has an integrated columnar representation for datasets that is aimed at doing machine learning analysis on operational data stored in the database, and this columnar format can get up to 100X better performance on analytical queries than the open source PostgreSQL.

The PostgreSQL license is very permissive about allowing innovation in the database, and Google does not have to contribute these advances to the community. But that said, Gutmans adds that Google intends to contribute bug fixes and some enhancements it has made to the PostgreSQL community. He was not specific, but stuff that is tied directly to Googles underlying systems like Borg and Colossus are not going to be opened up.

So now Google has three different ways to get PostgreSQL functionality to customers on the Google Cloud. Cloud SQL for PostgreSQL is a managed version of the open source PostgreSQL. AlloyDB is s souped up version of PostgreSQL. And Spanner has a PostgreSQL layer thrown on top but it doesnt have compatibility for stored procedures and triggers because Spanner is a very different animal from a traditional SQL database.

Here is another differentiator. With the AlloyDB service, Google is pricing it based on the amount of compute and storage customers consume, but the IOPS underpinning access to the database are free. Unmetered. Unlike many cloud database services. IOPS gives people agita because it cannot be easily predicted, and it can be upwards of 60 percent of the cost of using a cloud database.

AlloyDB has been in closed preview for six months and is now in public preview. General availability on Google Cloud is expected in the second half of this year.

Which leads us to our final thought. Just how many database management systems and formats does a company need?

We think of ourselves as the pragmatists when it comes to databases, says Gutmans, who is also famous as the co-founder of the PHP programming language and the Zend company that underpins its support. If you look at the purpose built database, there is definitely a benefit, where you can actually optimize the query language and the query execution engine to deliver best in class price and performance for that specific workload. The challenge is, of course, that if you have too many of these, it starts to become cognitive overload for the developers and system managers. And so theres probably a sweet spot in the middle ground between monolithic and multimodal. You dont go multimodal completely because then you lose that benefit around price, performance, use case specific optimization. But if you go too broad with too many databases, it becomes complicated. On the relational side, customers definitely have at least one relational database and in many cases they also are dealing with legacy database. And with those legacy databases, we are definitely seeing more and more interest in standardizing on a great open source relational database. Document databases provide a lot of ease of use, especially under web facing side of applications when you want to do things like customer information and session management with a very loose schemas, to basically have a bag of information about a customer or transaction or song. I am also a big fan of graph databases. Graph is really going through a renaissance because not only is it very valuable in the traditional use cases around fraud detection and recommendation engines and drug discovery and master data management, but with machine learning, people are using graph databases to extract more relationships out of the data, which can then be used to improve inferencing. Beyond that, we have some other database models that, in my opinion, have some level of diminishing returns, like time series or geospatial databases.

PostgreSQL has very good JSON support now, so it can be morphed into a document database, and it is getting geospatial support together, too. There is a reason why Google is backing this database horse, and getting it fit for the race. It seems unlikely that any relational database could have a good graph overlay, or that a graph database could have a good relational overlay, but that latter item is something to think about another day. . . .

Continue reading here:

Google Needs Another Database To Attack Oracle, DB2, And SQL Server Directly - The Next Platform

Serial 1, the electric bike company spun out of Harley-Davidson, launched its second-generation lineup of premium e-bikes but the biggest changes will be coming to the companys app.

The updated bikes will come with a host of new software features provided by Serial 1s new partnership with Google Cloud. The company says that Google Cloud has selected Serial 1 as its new strategic eMobility partner, meaning the e-bike maker will be among the first to integrate Googles software products into its vehicles.

The software-enabled e-bikes will allow owners to track their trips, collect data, and significantly improve safety and security, Serial 1 says. It reflects a trend in the e-bike industry to install bikes with cloud-connected software as an additional selling point.

The centerpiece of the new partnership will be the Serial 1 app, in which owners can see turn-by-turn navigation, collect ride data, and control security features on their bike. Serial 1 is promising more high-tech features to come thanks to the companys access to Google Cloud analytics and business intelligence and integration with Google Cloud AI functionality.

Google Cloud will also ensure a stronger connection between the bike and the users smartphone. Most e-bikes use Bluetooth to connect to a smartphone app, but Serial 1s bikes will use cellular and GPS technology, in addition to Bluetooth, to ensure owners can connect to their bikes even when they are not in their line of sight.

Just a quick refresher: Serial 1 is a standalone electric bike company that spun out from Harley-Davidson in October 2020. Its current lineup includes four bikes, ranging in price from $3,399 to $4,999. The brand names are Mosh/Cty, a city bike, and the commuter Rush/Cty, which comes in three variants (regular, Step-Thru, and Speed). Each comes with a mid-drive motor capable of generating 250W of continuous power and hitting top speeds of 20mph except for the Rush/Cty Speed, which can go 28mph.

The powertrains will be the same in the second-generation bikes. Most of the major changes are under the surface. These include improved security features, such as flashing lights, disabled pedal-assist functionality, and real-time locations.

The Serial 1 app will integrate with Google Maps to provide better navigation, for example, by prioritizing routes with bike lanes. Serial 1s simplified digital displays are supplied by Brose, a German company that also makes the bikes powertrain, so users will likely have to mount their smartphones on the handlebars to benefit from these types of features.

The app will also feature a virtual garage in which owners can name, track, and digitally manage their e-bikes. This will include a new dashboard for owners to monitor their bikes ride data, including speed, distance, range, power output (both for the rider and the battery), efficiency, and state-of-charge, among other metrics. Serial 1 owners can record their rides to learn more about their performance and progress. And the app will provide automatic service updates when their bikes are in need of a tune-up.

The physical look and controls for the second-generation bikes will remain largely the same. I loved the bikes when I got to test them out last year. The same team that developed the batteries for Harley-Davidson electric LiveWire motorcycles also developed batteries for Serial 1. The integrated batteries are mounted very low on the frame, which helps with the mass centralization and improved handling.

With this new update, its clear Serial 1 is taking aim at major manufacturers like Giant, Trek, and Specialized, which sell premium e-bikes for high-end customers. Specialized, in particular, has been touting the connected software in its Turbo lineup. And like Harley-Davidson, the company just announced that it was spinning out its own brand called Globe that will exclusively focus on utility e-bikes.

Originally posted here:

Harley-Davidson's updated Serial 1 e-bikes will feature Google Cloud connectivity - The Verge

In the tech space, the tools at our disposal and the associated terminology fluctuate more than the late, great Prince decided to change his stage name. This time up, the Artist Formerly Known as Universal Analytics had a career makeover and would prefer to go by Google Analytics 4 moving forward. However, the name is just the beginning of the new look and sound of Googles metrics tool, and their new album drops sometime in July of 2023. After next summer, Google will phase Universal Analytics out and retire it forever. As the deadline for this pivot approaches, lets talk about what this means for your law firm, how this change may benefit your marketing efforts in the future, and what simple steps you can take to prepare. Keep reading for a breakdown of everything you need to know about Google Analytics 4 for Law Firms.

At inception, the primary deficiency of Universal Analytics was not always a weakness. When Google dropped this release, Universal Analyticss method of collecting data for your marketing team to interpret was a strength. At its launch, UA was revolutionary. Technological advances have given way to more effective means of acquiring data and led to the antiquation of Universal Analytics. Why? Because Google Analytics 4 uses event-based tracking, as opposed to session-based.

When a user visits your website, Universal Analytics records this data as a session. The session can encompass a window of several hours and include other users. On the surface, this sounds fine, but within these sessions are potential clients who may not be interested in your law firm. Before Google Analytics 4, it was hard to figure out how to weed these undesirable user impressions.

Now, Google has switched to an event-based model, which records users interaction with a website as a unique individual event. The metadata associated with these events are logged, and you can target ads to people you and your marketing team feel would be your ideal client.

Referencing some of the issues raised with tracking cookies, Google created this new platform to address security and privacy issues among their users and improve your ability to advertise effectively, leading to more frequent media buys. (There has to be something in it for them, right?)

The session-based model is somewhat incomplete as more websites push to keep the information about who visits their site private in a bid to uphold the public trust. If you envision the user experience as a destination on a map, its like drawing a line between two points and erasing random sections. You may see a path from a distance, but its incomplete and valuable data lies in those gaps. Additionally, on that map, you may see the course of other users overlap with the path youre focused on, and it muddies the waters.

The event-based model tethers one users activity to a thread that tracks them, their devices, and their habits in a resolute way. This allows you to target ads to someone without Google revealing the users identity until theyre knocking on your law firms door. Hopefully, youve attracted a qualified lead in the process without breaking user trust. This event-based model draws the path of a potential client with a fat magic marker and erases the tracks of other users who may not be an ideal fit for your law firms area of expertise.

You may be asking yourself, Should I switch to Google Analytics 4? The answer to that question should be an emphatic yes, so imagine someone screaming it in your ear. The future is here, and Googles July 2023 caveat is not like the rolling deadline to convert your drivers license to RealID. Your law firm needs to be prepared to bid bon voyage to Universal Analytics when next summer hits. Taking the time now to talk about the future means youre ready to stay ahead of the competition and learn how to reach your ideal clients before your competitor knows how.

Takeaway:

As of July 2023, Google will phase out Universal Analytics in favor of Google Analytics 4, switching from a session-based to an event-based tracking model. This shift changes how Google records data, treating each instance of a users interaction with a website as its own timestamped event rather than a session that may encompass several hours and many other users. These events are unique to a degree and revamp the session-based model that targets potential clients with users of different habits in your ads through nothing more than the limitations of the programming itself. Now, advertisers can target ads with more precision using data that reveals more about a user while maintaining their privacy and upholding trust with your firm. The result is an ad campaign that may reach more potential clients because the data holds the secrets of generating impressions that leave an impression on the right people.

As Google Analytics 4 is integrated into use, so will Good2bSocials ability to help ease the transition for your firm.

Read the rest here:

Everything Law Firms Need to Know About Google Analytics 4 - JD Supra

Over the coming weeks, AT&T is rolling out cellphone location tracking thats designed to route emergency calls to 911 more quickly. The company says the new feature will be nationwide by the end of June and should make it easier for, say, an ambulance to reach someone experiencing a medical emergency. At first glance, it seems like a no-brainer. But its also a reminder that as phone companies promise to save lives, theyre also using a lot more data about you in the process.

The AT&T upgrade is part of a broader effort to modernize the countrys approach to emergency response. T-Mobile has also started using location-based routing, and experts told Recode that the technology could eventually be universal. At the same time, the federal government is in the midst of a nationwide push to get 911 call centers to adopt a technology called Next Generation 911, which will allow people not only to call 911 but also to send texts including images and video messages to the emergency line.

Meanwhile, Apple and Google have created new software that can directly pass on information from someones device, like information stored on a health app. The hope is that more data will save crucial time during emergencies, but privacy experts are already warning that the same technology could be misused or exploited.

Get the best of Recode's essential reporting on tech and business news.

I just worry what happens the next time theres a tragedy, the next time people are scared, and the next time theres an opportunity to use this data in ways it was never intended, Albert Fox Cahn, the executive director of the Surveillance Technology Oversight Project (STOP), told Recode.

One of the main ways phone networks plan to use this data is to connect callers with the right 911 operator more quickly. Because the 911 system was designed to work with landlines, calls to 911 made via cellphones (mobile phones place the majority of 911 calls) sometimes get routed to the wrong 911 center. In places that use older technology, cellphones will generally connect to the 911 operator associated with the antenna on the cell tower that processes the call, not the 911 operator in the jurisdiction the person calling is currently in. When these calls are misdirected, it can sometimes take several minutes to be connected to the right dispatcher.

To address this problem, carriers are turning to the sensors in smartphones, like GPS, wifi antennas, accelerometers, and pressure sensors. Depending on the phone you have, either Apple or Google can then use these sensors to estimate your current location. (Googles system is called Emergency Location Service, or ELS, and Apples system is called Hybridized Emergency Location, or HELO.) With AT&Ts and T-Mobiles new systems, when someone makes a call to 911, the phone network will use this location estimate to make a best guess as to where someone is, and then connect the call to the right 911 operator. AT&T says the whole process should take about five seconds and is supposed to locate someones call within 50 meters of their actual location.

This isnt the only data 911 centers have at their disposal. Apple already allows people to load their medical information like what health conditions they have and medications theyre on into their devices, and depending on the technology used by the jurisdiction youre in, that info could be automatically sent to emergency responders when they dial 911. Some Apple Watch models also have a built-in fall detector that can dial 911 on its own.

Meanwhile, the Federal Communications Commission (FCC) has ordered carriers to start transmitting vertical location data in addition to horizontal location data, making it easier for first responders to identify what floor someone might be on in a multistory building during an emergency. And as the federal government rolls out Next Generation 911, its also laying the groundwork for 911 operators to collect data from other connected devices, like cars with certain crash notification systems, building sensors, and wearables. This is all in addition to a host of other changes that a growing number of the countrys thousands of 911 call centers have been slowly making: upgrading software, sharing and collecting more analytics, and just getting better training. The idea behind all of these updates is that, with more information, dispatchers can make better decisions about an unfolding situation.

A lot of the underlying efforts around transforming 911 is really trying to help the current nations 911 system, prioritize health and safety for call takers and dispatchers, and really just trying to ensure that the right person is being dispatched at the right time, explains Tiffany Russell, the mental health and justice partnerships project director at the Pew Charitable Trusts. This police-first model is not necessarily the best response to handle these really complex problems or issues related to mental health.

In an emergency, more information could be helpful, but there are also reasons to worry about 911 collecting additional data. Allowing 911 operators to receive image- and video-based messages could create new opportunities for racial bias, Russell points out, and texting may not be the most efficient way for an operator to communicate during an emergency. The 911 system has played a fundamental role in and contributed to some of American policings worst problems, including over-policing, racist police violence, and deeply flawed approaches to domestic violence and behavioral health.

Another growing concern is data privacy. While AT&T told Recode that location data is only used when a 911 call is in progress, there are circumstances where 911 operators can directly request that information from a carrier, even if the person who made the call has hung up, according to Brandon Abley, the director of technology at the National Emergency Number Association. There is no way for an individual user to disable the location information sent during 911 calls.

These concerns with the 911 system arent new. When the FCC rolled out enhanced 911 an early program to improve the kind of information 911 operators receive about wireless callers civil liberties organizations like the Electronic Frontier Foundation (EFF) warned about the risk that federal agencies could try to access the data created by the new technology, or it could end up in the wrong hands. A recent FBI guide to cellular data shows that law enforcement does sometimes try to collect data created by carriers enhanced 911 capabilities. Its also abundantly clear that cellphone location data generally isnt well protected. Agencies like the FBI and the Department of Homeland Security have bought app-created location data on the open market, and as long as they have the right legal paperwork, law enforcement can reach out to any company that collects data about someone and ask for information.

They are not responsible with our data, there are not proper assurances in the law to limit how they use it, Andrs Arrieta, the director of consumer privacy engineering at EFF, told Recode. Sometimes even when there are, they keep misusing it.

These risks stand to get a lot more serious and a lot murkier as 911 centers across the country start receiving far more data from peoples devices. This could take some time, since 911 call centers are generally run on the local level and vary considerably in terms of the technology they use. Still, its critical to remember that even if a new service is designed or marketed as a new way to save lives, theres no guarantee thats the only way it will be deployed.

This story was first published in the Recode newsletter. Sign up here so you dont miss the next one!

Read the rest here:

How AT&T, Google, and Apple are shaping the future of 911 - Vox.com