Mediaboss Marketing

The Plex service is widely appreciated by multimedia enthusiasts all over the world. It is a free software package that allows users to stream their media video, photos, and music to any device running the same software. Plex is supported on many platforms including consoles, smart TVs, tablets, and so forth. However, the company recentlymade an announcementthatupseta lot of people. Ithad planned to make data collection mandatory, without giving users an option to opt out. That plan has been canceled due to thenegative backlash.

In this day and age ofdata collection and invading ofprivacy, any decision made by companies in this regard will be scrutinized. Plex is no exception, as the software package is used by millions of people around the world. It is one of the most convenient ways to access and stream different types of media to any device. However, a newly announce plan by the company over the weekend had users up in arms, as they felt their privacy would have been invaded due to this decision.

The announcement of the change by the Plex team was bound tohave major repercussions. The company aimed to collect data onhow customers were usingthe software and services for which it is known. Up until now, that has always been a matter of users opting in to share this data or keeping it private at all times. The company decided the time had come to make abig change in this regardand make data collection mandatory at all times. Indeed, there would be no opt-out feature whatsoever.

The updated privacy policy reflecting these changes was not received all that well by the Plex community. In fact, they voiced their outrage on social media platforms and on the companys forum as well. Most Plex users do not wishto share their information with the company or detail how they use the software. After all, Plex is often used to stream less-than-legal content to other devices, which would give the company information onwhich users are known to havepossessedpirated content. It is unclear how that information would have been used, and with no opt-out feature, a lot of problems could have ensued.

According to the explanation provided by the company, this change was indeed necessary. A lot of information wasalready being transmitted due to servers connecting to the cloud, new services designed using metadata, and communication through the Plex cloud infrastructure to relay playback requests. Providing an opt-out feature in the setup, Plex claimed, would giveusers a false sense of privacy as a whole. The Plex software already collects a ton of information from which people cannot opt-out, and thus this updated privacy policy only reflected things which had been present for quite some already.

It did not take long until the Plex user base caught wind of this change andassumed the natureof an angry mob. Most of the users voicing concern promised to take their business elsewhere to other services, either paid or free of charge. Although thechange was supposed to go into effect onSeptember 20th, it appears the company has already reversed its decision altogether. This demonstrates thatfeedback from the public can certainly influence a companys decision-making. Even though Plex meant no harm whatsoever, the potential implications of thedata collection could have been catastrophic.

Additionally, Plex would have also been in violation of European Union law. These laws clearly state any company collecting users information needs the permission of the user specifically. Any updates made in regards to which information is collected has to be approved by users as well, which Plex had no intentionof doing. The fact that theplan will not go into effect after all is a small victory for privacy advocates. The opt-out feature will remain where it is right now, although the amount of data collected when users opt in will still reflect the proposed changes.

See the original post:
Plex Users Outraged Over Company's Plans to Collect More Information - The Merkle

International automakers and the Austrian Transport Ministry agreed Tuesday that the owners of 600,000 diesel cars in the country would be entitled to free software upgrades with a view to reducing nitrogen oxide emissions.

They said the deal would include 350,000 recalled Volkswagen vehicles that had used illegal emissions-cheating devices.

Austrian car owners would get incentives such as vouchers to get as many upgrades as feasible, Austrian Transport Minister Jrg Leichtfried told reporters.

Improved results

In addition, car importers agreed to offer a premium to drivers, who would be willing to trade in their diesel cars for models that emitted less exhaust gases.

While a recent diesel summit in Germany led to a similar agreement that covered German carmakers Daimler, Volkswagen and BMW as well as French-owned Opel, the Austrian deal also involves other French, Korean and Japanese manufacturers.

"We have improved the meager results that were achieved in Germany and negotiated a better deal for Austria," Leichtfried said.

He acknowledged, though, that neither he nor German politicians had managed to convince automakers to install emissions-reducing hardware in older models.

hg/jd (dpa, Reuters)

Read the original here:
Austria follows German drive to make diesel cars cleaner - Deutsche Welle

Tech Advisor

Best free text-to-speech software Tech Advisor ... free text-to-speech programs available right now is Balabolka. This diminutive download taps into the existing Microsoft Speech Program that is included in Windows. Using the Microsoft Speech API (SAPI) it's able to harness the conversion software ...

The rest is here:
Best free text-to-speech software - Tech Advisor

Sacramento Bee

Google to serve next version of Android as 'Oreo" Sacramento Bee Oreo boasts several new features, including the ability to respond to notifications directly on a phone's home screen and the ability to access apps without installing them on a device. The free software is scheduled to be released this fall, most ... Android Oreo: 10 things you need to knowTechRadar Android 8.0 Oreo officially unveiled, here's when you can expect to receive itThaiVisa News all 785 news articles »

See the original post here:
Google to serve next version of Android as 'Oreo" - Sacramento Bee

Enlarge / Look, it's really hard to find stock imagery for "evolutionary algorithm."

Adobe Stock

With 1.3 billion daily users, the Facebook site and its apps are the most-used pieces of software in the world.Only a handful of software companies have ascended to a similar echelon of ubiquity, including Microsoft, Google, and Apple. For better or worse, that is the world we now live in, where a large percentage our waking hours is spent interacting with softwareand Facebook leads the pack, with the average userspending 50 minutes per day mostly watching videos and liking photos of babies.Television is the only leisure activity in the worldthat receives more attention than Facebook. And don't forget that Facebook now owns Instagram and WhatsApp, too.

It is understandable, then, that Facebook cares a lot about the quality of its software. If Facebook pushes out a new version of its Android app with a crashing bug, millions of users could be affected. Those users might be inclined toswitch to another social network, or even worse: put down their phone and interact with the real world. The net effect is the same, either way: Facebook's share of your attention, and thus potential revenue, decreases.

That's why Facebook has some advanced bug-finding toolsincluding a devilishly clever dynamic analysis tool, initially devised by students at University College London and then acquired and further developed by Facebook's London office. This is the first time they've shown the inner workings of this new tool, dubbed Sapienz, to the press.

Eachtechnique serves a different purpose, and a big software company would usually use both. Static analysis is perfect for formally verifying that an algorithm works as intended, or for highlighting bad code that might allow for a buffer overflow orother security vulnerability. Dynamic analysis is better at finding the gnarly edge cases that cause crashes. Humans can manually perform both analyses, of course, but computers are obviously a lot quicker when it comes to testing millions of possible inputs.

Facebook's static analyser is called Infer. The company open-sourced the toolin 2013, and a lot of big names (Uber, Spotify, Mozilla) use it. There isn't a whole lot to say about it, other than it seems to be very popular and effective; download it today!

Sapienz has three main tricks up its sleeve.First, it uses a search-based evolutionary algorithm, rather than a random or model-based approach. Second, the fitness function that guides how the algorithm evolves is incredibly complex: there are multiple objectives, entwined by Pareto optimality, that must be fulfilled for each evolutionary change to be a success. And third, Facebook can run Sapienz on its One World test platform, which lets engineers find crashing bugs on hundreds of different Android devices simultaneously. (Sapienz only supportsAndroid apps currently, though there are plans to expand to otherplatforms and app types.)

The key to a successful evolutionary algorithm is its fitness function. I'm not your college computer science lecturer, so I won't go into too much detail, but a fitness function essentially looks at the result of a test case, and decides how close that result is to a desired outcome/objective. The results that don't fulfil the fitness function are tied up in a burlap sack and thrown in the river; the good ones are bred together, to form the basis of the next round of testing.

According to Facebook's engineers, most of their secret sauce is in Sapienz's fitness function, which has three objectives: to test as many of the app's methods and statements as possible, find as many crashes as possible, and minimise the length of the test sequences required to cause crashes. The first two are all about producing better, crash-free software; the third is to improve the efficiency of the system, so that a decent number of crashes can be found in a reasonable amount of time.

These three objectives are assessed by the fitness function for Pareto efficiency. That is, one objective isn't more important than the others: if the evolutionary algorithm is only producing long test sequences, but they're providing good coverage and finding lots of crashes, then those long tests will be kept alive. Over time the systemtries to hit Pareto optimality: where it's impossible to improve one objective without negatively impacting another. So, in this case, the algorithm would attempt to reduce the test sequence length without reducing coverage or the fault revelation.

Sapienz also strays slightly across the border into static analysis: it attempts to reverse-engineer the app (an Android APK in this case) to pull out some strings, which it then uses as natural-language inputs when testing begins. "We found this seeding to be particularly useful when testing apps that require a lot of user-generated content, e.g., Facebook, because it enables Sapienz to post and comment in an apparently more human-meaningful way," say the researchers.

Originally posted here:
Facebook's evolutionary search for crashing software bugs - Ars Technica UK