Showing posts with label Browser Trends. Show all posts
Showing posts with label Browser Trends. Show all posts
The Firefox-maker is poised to introduce a novel sandboxing technology called RLBox in Firefox 95, which technology was developed in collaboration with researchers at the University of California San Diego and the University of Texas.

RLBox Sandboxing technology uses WebAssembly to isolate potentially-buggy code, making the isolation of subcomponents easier for more secure browsing and will be extended to all supported Firefox platforms (desktop and mobile), with five different modules: Graphite, Hunspell, Ogg, Expat and Woff2.

Even with these modules as untrusted code, or a zero-day vulnerability in any of them should pose no threat to Firefox with the RLBox technology in place.

How Isolating with RLBox makes the Difference?



The major browsers including Chrome, Safari and Microsoft Edge, all run Web content in its own sandboxed process, which in theory helps to prevent it from exploitations by a browser vulnerability to compromise your computer.



And having isolated things along trust boundaries, the next logical step is to also isolate across functional boundaries. This has meant hoisting a subcomponent into its own process, as Firefox runs audio and video codecs in a dedicated, locked-down process with a limited interface to the rest of the system. But, there are some serious limitations to this approach, it requires decoupling the code and making it asynchronous, which is time-consuming and could impose a performance cost.

As a result, nobody would want to seriously consider hoisting something like the XML parser into its own process, because to isolate at that level of granularity, we need a different approach.

Now, with RLBox, rather than hoisting the code into a separate process, it can instead be compiled into WebAssembly and then the WebAssembly compiled into native code and no need of shipping any .wasm files in Firefox, since the WebAssembly step is only an intermediate representation in the build process.

RLBox makes a big win on several fronts: it protects users from accidental defects as well as supply-chain attacks, and reduces the need to scramble when such issues are disclosed upstream.

How to Upgrade to the latest Firefox 95?



Firefox 95 is available for Windows, macOS and Linux, and can be downloaded from Mozilla's official site or from the respective app stores. But as Firefox updates happen in the background, users will only need to relaunch the browser to receive the latest version.

If perhaps you failed to get automatic update, you can manually update on Windows, by pulling up the menu under the horizontal bars at the upper right, then click on the help icon, which is the question mark within the circle. And select "About Firefox" and the upgrade will begin.

For macOS users, they should select "About Firefox" which can be found under the "Firefox" menu. And the page will show whether the browser is up to date or recommend the refresh process.

Mozilla tightens Sandboxing in Firefox 95 with RLBox technology

Google has rolled out Chrome 91 on the stable channel, with the latest version of the browser bringing tons of features including the super-fast non-optimizing Sparkplug compiler, that compiles bytecode to machine code.

While Google had earlier introduced a two-tier compiler system in the V8 engine made up of Ignition and Turbofan for JavaScript execution in Chrome browser. The Ignition is responsible for executing the JavaScript while Turbofan optimizes the code for maximum performance, these compilers make tradeoffs during the different phases of JavaScript execution.

As part of Chrome’s V8 JavaScript/WebAssembly engine, Sparkplug will be nestled between the Ignition interpreter and the TurboFan optimizing compiler; thus filling the gap between the JavaScript execution tradeoffs.

How Sparkplug compiler will ensure faster performance in Chrome?



Sparkplug does not depend on gathered information while executing JavaScript for generating machine native codes, which allows for quicker execution and generating high-performance codes.



It compiles directly from bytecode instead of JavaScript source; thus Sparkplug compiles functions already compiled to bytecode, and the bytecode compiler having done work such as determining if parentheses are arrow functions, variable resolution, and desugaring destructuring statements.

And given that it compiles directly to machine code through a single linear pass over the bytecode, it emits code that matches the execution of that bytecode.

How to Download or Upgrade to Chrome 91?



Google updates for Chrome happens in the background, and also, users can install the latest version manually, by clicking on "About Google Chrome" from the Help menu; with the result tab display showing if the browser has been updated or the download-and-upgrade button.

Apart from the Sparkplug compiler feature, the new update also brings many performance improvement under the hood, which according to Google, Chrome 91 is 23 percent faster than the previous version.

Chrome Sparkplug compiler to boost JavaScript performance

Mozilla's Python-in-the-browser project, Pyodide is an experimental project to create a full Python data science stack that runs entirely in the browser.

While Pyodide is the offshoot of another Mozilla project, known as Iodide, which is a tool for data science experimentation and communication based on state-of-the-art web technologies. Now, Mozilla has spin out Pyodide into an independent, community-driven project.

Going forward, the project will be maintained by community of volunteers and Mozilla has published a governance document with a project roadmap, which outlines what the goals are, such as reducing download sizes, better performance of Python code, and simplification of package loading.

Pyodide aim to Bring the scientific Python stack to the browser



JavaScript, the common browser language doesn’t have a mature suite of data science libraries, and missing a number of features that are necessary for numerical computing, like operator overloading.



Pyodide is designed to perform data science computation within the browser rather than a remote kernel, it gives you a full, standard Python interpreter that runs entirely in the browser, with access to the browser’s Web APIs. Also, Pyodide can install Python packages with a pure Python wheel directly from PyPi, the Python Package Index.

Pyodide includes a foreign function interface which exposes Python packages to JavaScript and the browser UI, including the DOM, to Python, making several Python scientific packages, including Matplotlib, SciPy, NumPy, Pandas, and Scikit-learn, available in the browser.

What does the Community-driven development for Pyodide means?



The Community-driven development for Pyodide means that control of the development process, resources and decision making authority will now come directly from groups in the community.

Therefore, developers are invited to try out Pyodide in a REPL in their browser, even as Mozilla has recently announced the release of Pyodide 0.17, bringing major improvements, and a redesign of central APIs, with elimination of error and memory leaks.

Mozilla's Python-in-the-browser project now Community-driven

Compat2021 is a lofty project led by Google and Microsoft, with the “broader web community” to pool resources that will improve browser compatibility along five critical areas identified by the group.

According to the group, the last couple of years have highlighted browser compatibility issues as the top challenges faced by developers, and research in the MDN Browser Compatibility Report has helped hone that signal into five areas where browser compatibility is particularly an issue, namely: CSS Flexbox, CSS Grid, CSS position: sticky, the CSS aspect-ratio property, and CSS transforms.

Albeit, the effort seems to be centered around improving the open source Chromium engine which powers Google Chrome and Microsoft’s Edge browsers.

What the Compat2021 project hopes to achieve?



The Compat2021 project working group identified the focus areas above based on feature, number of upvotes on given bugs in their different tracking system, with various feedback via survey, CanIUse data, and results from web-platform-tests.



Microsoft Edge team, on its part, intends to contribute fixes to Chromium to surpass 100% of CSS Grid tests and support the improvement of interop across browsers, as well as assist with triage in web-platform-tests. While the Chromium project had already started work on improving the compatibility of the browser in 2020 with the fine-tuning of the scope of the changes offered by Microsoft.

Still, there are compatibility issues in basically all of the web platform, even though the focus of this project remains on a rather small number of the most important areas, those voted as top issues for developers.

How to get Involved with the Compat2021 Project?



If you are a developer and encountering compatibility issues in the above listed areas, it is advised that you continue to file bugs in the appropriate tool via the “Send Feedback” or directly in the appropriate project like Chromium, Webkit, or Gecko.

And you can follow up on the project’s progress on the Compat2021 Dashboard on web-platform-tests, and by subscribing to the mailing list for updates.

Compat2021 Project towards improving Browser Compatibility

Brave, the company behind the popular privacy-focused browser by the same moniker recently announced its acquisition of Tailcat, an open search engine developed by the team responsible for the private search and browser products at Cliqz.

With the acquisition, Brave is developing a privacy-focused alternative search engine to Google that promises above all things to never profile users activities. Brave refers to its upcoming search engine as an antidote to “Big Tech,” and the need to harvest users search history to serve targeted ads and recommendations.

Interestingly, Brave will also be bringing a combination of privacy-focused browser and search engine, which is perhaps what made Google successful, given the heavy adoption of its chrome browser.

What Brave Search is bringing to the table?



The open search engine, Tailcat will become the foundation of Brave Search, with Brave Search and the Brave browser constituting the industry’s first true independent, privacy-focused alternative to Google and Chrome, which rely heavily on tracking its users across all websites.



As Brave browser offers the stringent privacy protections to users, even so is the Brave Search being developed according to the same principles, as follows:

  • Brave Search offers Choice: Providing options for ad-free paid search and ad-supported search.
  • Brave Search is Independent: Relies on anonymized contributions from the community to improve and refine its Search.
  • Brave Search is User-first: Meaning that Brave serves the user first, not the ad and data industries.
  • Brave Search is Private: it doesn't track or profile users.


Brave Search comes as part of the family of privacy-preserving Brave products, even as consumers are increasingly shifting to user-first alternatives. Brave browser, as a result, recorded an unprecedented growth in 2021, reaching more than 25 million monthly active users, lending to the fact that privacy is now becoming mainstream.

Does Brave search stand any chance with Google's strangle-hold on search?



Brave has grown its browser market share significantly over the past year, from about 11 million monthly active users to over 25 million. Therefore, it is expected that even greater demand for Brave in 2021 will spur the adoption of their search service, as more and more users demand real privacy solutions to escape the Big Tech’s invasive practices.

As a case in point, DuckDuckGo, another privacy-focused search engine has been growing steadily since it was launched in 2008, and has even scaled its efforts to capitalize on growing international reach for its pro-privacy products, which Google has recently recognized the importance of offering a private search option, by adding DuckDuckGo in the available search engines on Chrome for over 59 countries.

Brave Search: Privacy-focused Search Engine to counter Google

Google announced a new change coming to the open-source Chromium project, whereby it will limit access to several of Chrome application programming interfaces (API) starting from March 15, 2021.

The change was necessitated, according to Google, as a result of third-party Chromium based browsers integrating Google Chrome features, such as Chrome sync and Click to Call, which are intended only for Google’s services.

And it has such security implications that users could easily sign into their Google Account and store their personal data via Chrome sync, with just any third-party Chromium based browsers.

How the Change would impact other Chromium-based browsers



Google's plan to limit some of the Chrome APIs features that it includes inside Chromium will impact not only Chrome Sync, but also some other features such as the Chrome Translate Element, Chrome Spelling API, the Contacts API, and many more.



These APIs implemented within Chromium source code, with the open-source shell which is the base of Chrome browser, are usually removed or modified by other companies that build browsers based on the Chromium code to their own systems, over which they have control.

With the coming changes, Google has given third-party companies two months to remove every Chrome-specific APIs from their code before access is cut off. And Chromium browsers ranging from Microsoft Edge, Brave and Opera, will have to develop and implement their own to retain the specific-features.

What's the impact on Programmers and End-users?



The change will definitely affect Linux developers at multiple facet, especially, for Linux Chromium browser users to discover that the latest versions of their browsers can't perform some basic tasks.

Albeit, there is a growing mistrust between Google and Linux vendors, with the later edging closer to dumping Chromium. This latest development will leave several others with a bad taste over how Google has failed the open-source community, as it will make all Linux Chromium builds significantly less functional.

Google set to change the rules for Chromium APIs access

Apple brought a number of countermeasure to prevent attacks with iOS 14.4 update, such as the introduction of a sandboxed 'BlastDoor' feature for the parsing of untrusted data in iMessage.

Now, the company has gone a bit harder on the bad actors, with iOS 14.5 update, which will introduce fraudulent website check for Safari browser, which also passes traffic through its own proxy servers to prevent leaking of IP addresses and protect users' privacy.

The "Fraudulent Website Warning" feature will alert users about any dangerous website that have been previously blacklisted or reported as malicious, deceptive, or harmful.

How the Safari browser Fraudulent Website Alert will work?



Apple will rely on Google Safe Browsing, which is a blacklist service that offers a list of web resources that contain malicious or phishing content, against a hash prefix calculated from the addresses and checked for fraudulent activities.



The database will then prompt Safari browser to request for the full list of URLs that correspond to the hashed prefix, and subsequently, block access to the site with a warning to the user. While this approach will ensure that the actual website the user is trying to visit is never shared with the safe browsing provider, but it will definitely leak the IP address of the device.

Albeit, Safari will proxy the Fraudulent Website Alert service through Apple servers to limit the risk of information leak.

Additional Privacy measures that Apple will be rolling out



Apple will also require that apps request for users' permission before tracking them in iOS 14.5 and websites using the device's ad identifier as part of App Tracking Transparency.

The new anti-tracking features has been seen as a big hit in the ad revenue generated by internet marketers and those that rely on cross-device user tracking and the resulting ad revenue generation to continue their offering of free services.

Fraudulent Website Alert coming to Safari browser with iOS 14.5 update

IPFS, which stands for InterPlanetary File System, is a peer-to-peer networking protocol similar to BitTorrent designed to serve as a decentralized storage system.

While Brave browser is the first web browser to integrate IPFS support, and offers an alternative way of using the internet instead of the HTTPS (Hypertext Transfer Protocol Over Secure Socket Layer) protocol that is currently the norm.

Albeit, IPFS is a relatively obscure transport protocol with promises of improving on the dominant HTTP standard by ensuring faster access to content and more resiliency to failure and control.

Why does IPFS support matter in Brave's Web Browser?



Normally, web users uses URLs (universal resource locators) to access the different sites from centralized servers and you are left with either using HTTP or HTTPS on your browser. Now, the physical proximity to the servers does affect how long it takes to load a page and the amount of bandwidth.



But, IPFS instead of relying on URLs and servers, distributes website related data across a network. And the data can also be accessed with URIs (universal resource identifier), which in simple terms, is akin to how BitTorrent and blockchain function. With every PC or mobile device known as “nodes” and temporarily stores a piece of a website’s data, meaning that whenever a user accesses a website via IPFS, it loads data from any nearby nodes on the network.

The obvious advantages of the new approach include faster speed as data is distributed and stored closer to the users who are accessing the information, as well as lower costs of server for the original publisher of the content. And most importantly, IPFS has the potential to make content more resilient to failures and censorship.

As a user, you can also choose to not be a node and access IPFS content through public gateway, though the public gateways can view and log your IP address.

Some Privacy concerns in Brave’s IPFS network



There are privacy concerns in Brave’s IPFS network, whether you’re a local node or simply loading IPFS content. If you’re a node, Brave gives you a unique ID, which ID though hashed, is still viewable by other users to see what you are hosting and accessing. Your PC and network’s resources will also be used when another user accesses the IPFS data you’re hosting.

And even if you choose to not be a node and simply access IPFS content via a public gateway, the public gateways can still view and log your IP address. But, Brave for now uses IPFS alongside HTTP/HTTPS, so it will affect only the content that’s configured for IPFS hosting.

What function does the IPFS protocol in Brave's Web Browser serve?