| Chromium | |
|---|---|
 | |
.webp) | |
| Original author | Google[1] |
| Developers | The Chromium Projects,[2] controlled by Google[3] |
| Initial release | 2 September 2008; 17 years ago (2008-09-02)[1] |
| Repository | |
| Written in | C++ primarily,[4]HTML,CSS,JavaScript forUI and test suite[5][6] |
| Engines | V8,Blink (WebKit oniOS/iPadOS)[7] |
| Operating system | |
| Platform | IA-32,x86-64,ARM,ARM64 |
| License | LGPL,BSD-3 and others[8][9] |
| Website | www |
Chromium is afree and open-sourceweb browser project, primarily developed and maintained byGoogle.[3] It is a widely usedcodebase, providing the vast majority ofcode forGoogle Chrome and many other browsers, includingMicrosoft Edge,Samsung Internet, andOpera. The code is also used by severalapp frameworks.
Chromium is afree and open-source software project. Owing to its origins in theWebKit project, Chromium incorporates code subject to a variety of licenses, includingLGPL,MIT,Ms-PL, andMPL (with the MPL/GPL/LGPLtri-license clause).[9] Original Google-authored portions and some portions written by Apple are shared under the3-clause BSD license.[8]
This licensing permits any party tobuild thecodebase and share the resultingbrowserexecutable with the Chromium name and logo. Thus manyLinux distributions do this, as well asFreeBSD andOpenBSD.[10] Chromium also forms the basis ofElectron, which in turn is used for many commercial and proprietarycross-platform applications, such asSlack.
Chromium provides the vast majority of source code forGoogle Chrome, but there are important differences.
Chromium lacks the followingChrome features:[10]
Google first chose the name "Chrome" for its browser.[12][13] Then the open-source release was named "Chromium" becausechromium metal is used to makechrome plating.[1]
Chrome has the sameuser interface functionality as Chromium, but with a Google-branded color scheme. Unlike Chromium, Chrome is not open-source; its binaries are licensed asfreeware under the Google Chrome Terms of Service.[14]
The Chromium browsercodebase contains over 36 millionsource lines of code, excludingcomments andblank lines.[4]
Chromium has been a Google project since its inception,[1][3] and Google employees have done the bulk of thedevelopment work.[15]
Google refers to this project and the offshootChromiumOS as "the Chromium projects",[2] and its employees use @chromium.orgemail addresses for this development work. However, in terms of governance, the Chromium projects are not independent entities; Google retains firm control of them.[3]
The Chromiumbrowser codebase is widely used, so others have made important contributions, most notablyMicrosoft,Igalia,Yandex,Intel,Samsung,LG,Opera,Vivaldi, andBrave.[15][3] Some employees of these companies also have @chromium.org email addresses.
Google designed the first multi-process browser.[16][17] Compared to single-process designs, this architecture has better responsiveness with many browsertabs open and security benefits ofprocess isolation, but with the trade-off of higher memory usage.[18][19] This was later refined as per-process website isolation, providing additional security.[18]
Another important design decision was for the browser'suser interface to beminimalistic.[17][20][21]
Thebrowser engine was originally based onApple'sWebKit, which Google deemed the "obvious choice" of available options.[22] However, Google's novel multi-process design required engine changes. This divergence from Apple's version increased over time, so in 2013 Google officiallyforked its version as theBlink engine.[22][23]
C++ is the primarylanguage, comprising over half of the codebase.[4] This includes theBlink andV8engines, the implementation ofHTTP and otherprotocols, the internalcaching system, theextension API, and most of theuser interface.[24]
The rest of the user interface, called the WebUI, is implemented inHTML,CSS, andJavaScript (usually viaTypeScripttranspilation).[24][5] An extensive collection ofweb platform tests is also written in these languages plusXML.[6][4]
Third-partylibraries that provide essential functionality,[25] such asSQLite and numerouscodecs, are written inC, C++, or beginning in 2023,[26][27] the newerRust language.[4]
Support formobileoperating systems requires special languages: forAndroid bothJava andKotlin, and foriOS bothObjective-C andSwift.[5][4]
Python is the main language of the build system,[4] which also has special configuration files for Google's GN tool.[28]
Thebug tracking system is a publicly accessible website.[29] Participants are identified by theiremail addresses.
The Chromiumcontinuous integration system automaticallybuilds and tests the codebase several times a day.[30]
Builds are identified by a four-partversion number that is major.minor.build.patch.[31] This versioning scheme and thebranch points that occur every six to seven weeks are fromGoogle Chrome and itsdevelopment cycle.[32]
Google Chrome debuted in September 2008, and along with its release, the Chromium source code was also made available, allowing builds to be constructed from it.[1][33][34]
Upon release, Chrome was criticized for storing auser'spasswords without the protection of a master password. Google has insisted that a master password provides no realsecurity against knowledgeablehackers, but users argued that it would protect against co-workers or family members borrowing a computer and being able to view stored passwords asplaintext.[35] In December 2009, Chromium developer P. Kasting stated: "A master password was issue 1397. That issue is closed. We will not implement a master password. Not now, not ever. Arguing for it won't make it happen. 'A bunch of people would like it' won't make it happen. Our design decisions are not democratic. You cannot always have what you want."[36]
Version 3 was the firstalpha available forLinux. Chromium soon incorporated native theming for Linux, using theGTK+ toolkit to allow it fit into theGNOME desktop environment.[37][38][39][40] Version 3 also introducedJavaScript engine optimizations and user-selectable themes.[41]
Version 6 introduced features for user interfaceminimalism, as one of Google's goals was to make the browser "feel lightweight (cognitively and physically) and fast".[21] The changes were a unified tools menu, no home button by default (although user configurable), a combined reload/stop button, and the bookmark bar deactivated by default. It also introduced an integratedPDF reader,WebM andVP8 support for use withHTML video, and a smarterURL bar.[42][43]
Version 7 boosted performance to twice that of prior versions viahardware acceleration.[44][45][46]
Version 8 focused on improved integration intoChromeOS and improvedcloud features. These include backgroundweb applications, host remoting (allowing users centrally to control features and settings on other computers) and cloud printing.[46]
Version 9 introduced a number of features, including a URL bar feature for exposingphishing attacks andsandboxing for theAdobe Flash plug-in. Other additions were theWebGL library and access for the newChrome Web Store.[47][48][49]
In February, Google announced that it was considering large-scaleuser interface (UI) changes, including at least partial elimination of theURL bar, which had been a mainstay of browsers since the early years ofthe Web. The proposed UI was to be a consolidation of the row of tabs and the row of navigation buttons, the menu, and URL bar into a single row. The justification was freeing up more screen space forweb page content. Google acknowledged that this would result in URLs not always being visible to the user, that navigation controls and menus may lose their context, and that the resulting single line could be quite crowded.[51] However, by August, Google decided that these changes were too risky and shelved the idea.[52]
In March, Google announced other directions for the project. Development priorities focused on reducing the size of theexecutable, integratingweb applications andplug-ins,cloud computing, andtouch interface support.[53][54] Thus a multi-profile button was introduced to the UI, allowing users to log into multiple Google and other accounts in the same browser instance. Other additions weremalware detection and support for hardware-acceleratedCSS transforms.[55][56]
By May, the results of Google's attempts to reduce the file size of Chromium were already being noted. Much of the early work in this area concentrated on shrinking the size ofWebKit, theimage resizer, and theAndroid build system.[57] Subsequent work introduced a more compact mobile version that reduced the vertical space of the UI.[58][59]
Other changes in 2011 wereGPU acceleration on all pages, adding support for the newWeb Audio API, and theGoogle Native Client (NaCl) which permits native code supplied by third parties as platform-neutral binaries to be securely executed within the browser itself.[60][61] Google'sSkia graphics library was also made available for all Chromium versions.[62][63]
The sync service added for Google Chrome in 2012 could also be used by Chromium builds.[64][65] The same year, a newAPI for high-quality video and audio communication was added, enablingweb applications to access the user'swebcam andmicrophone after asking permission to do so.[66][67] ThenGPU accelerated video decoding forWindows and support for theQUIC protocol were added.[68][69]
In 2013, Chromium's modifiedWebKitrendering engine was officiallyforked as theBlink engine.[22][23]
Other changes in 2013 were the ability to reset user profiles and newbrowser extension APIs.[70] Tab indicators for audio and webcam usage were also added, as was automatic blocking of files detected asmalware.[71]
Version 67 added the security benefit of per-process website isolation.[18] Then version 69 introduced a new browser theme, as part of the tenth anniversary of Google Chrome.[72] The same year, new measures were added to curtail abusive advertising.[73]
Since 2021, the Google Chrome sync service can no longer be used by Chromium builds.[74][11] Also, starting with version 89, onlyIntel andAMD processors with theSSE3 instruction set or later are supported.[75]
Starting with version 110, onlyWindows 10 and later are supported for Windows users.[76]
In addition toGoogle Chrome, many other activelydeveloped browsers are based on the Chromium code. Most of these areproprietary, like Chrome, but some remainfree and open-source software (FOSS), like Chromium.[77][78][79][80]
Primarily non-English
These notableapp frameworks embed a Chromiumbrowser as the functional core of customapps:
The two prominent Chromium-basedWebView components also provide a similar way to make apps:
With either approach, the custom app is implemented withHTML,CSS,JavaScript, and other web technologies. Moreover, the app can be readily deployed on theoperating systems supported by Chromium. Since the 2010s, many apps have been created this way. (Two examples areSpotify andSlack.)[86][87]
Due to constraints of the iOS platform, all browsers must be built on top of the WebKit rendering engine.
There is a reason when we built Chrome we minimized everything to do with Chrome so that all you spent time on was the website you cared about at the given time. We wanted the users to focus on the content they were using. The reason the product was named "Chrome" was we wanted to minimize the chrome of the browser. That's how we thought about it.
Chrome was the first browser with a multi-process architecture.
Here are the sources to the great Chromium Embedded Framework that is used by the Spotify Desktop client.
