At the Mozilla All-Hands in SF we had some discussions about Mozilla's future plans on Android. Currently, Mozilla are actively supporting four Android app web browsers - Fennec[0], Focus[1], Rocket[2], and Firefox on Fire TV[3].
Focus, Rocket, and Firefox on Fire TV are all minimal browsers, and they do not have the same features as Fennec. Fennec is the codebase Orfox uses, and it is the browser we are currently planning on using for Tor Browser. However, Fennec is not being actively developed, but Mozilla are keeping Fennec alive and they do not have an EOL date.
The two important parts from this are: 1) The Fennec code will not change much over the next 12+ months 2) We can safely continuing using Fennec for the 12+ months
This effectively means Fennec 67 will be very close to Fennec 60ESR. Therefore, I propose we backout our current plan of following Fennec releases and we make this simple and follow ESR (exactly the same as desktop). We can track on-going changes in mozilla-central that affect Fennec, but these changes should not happen often - backporting any high priority patches shouldn't be difficult.
The two additional components I should mention are GeckoView[4] and Fenix[5]. GeckoView is an Android library acting as a glue-layer between the Gecko native code and the UI/Browser app layer. Firefox Focus is currenty using Android WebView (Android's built-in renderer), but they are moving Focus onto using GeckoView in the very near future. Fennec currently uses parts of GeckoView, so if we decide we should use the ESR then we should also watch the work done on GeckoView in case there are any improvements we want.
Fenix is the last piece in this puzzle. It is a new browser Mozilla will be working on, and it has a lot of promise - but currently it does not exist in a form we can use. We may want to consider using Fenix in 2-3 years.
Mozilla's Android Components project is also interesting. When it is ready, we could create our own browser by picking-and-choosing the parts we want. Unfortunately, this isn't as easy as hardening an existing private browsing mode. This is the reason I think we should continue working on Fennec right now.
In conclusion, I propose we develop Tor Browser for Android based on Firefox Fennec ESR. This is not an official ESR, but the changes between Fennec releases will be very small (if any changes at all). This will allow us maintain the same branches for desktop and mobile.
[0] https://play.google.com/store/apps/details?id=org.mozilla.firefox [1] https://play.google.com/store/apps/details?id=org.mozilla.focus [2] https://play.google.com/store/apps/details?id=org.mozilla.rocket [3] https://www.amazon.com/Mozilla-Firefox-for-Fire-TV/dp/B078B5YMPD [4] https://wiki.mozilla.org/Mobile/GeckoView [5] https://github.com/mozilla-mobile/fenix [6] https://github.com/mozilla-mobile/android-components
Hi Matthew,
Thanks for the heads up. I have one question, Are they going to freeze even the Android SDK and NDK versions?
On 06/14/2018 08:51 PM, Matthew Finkel wrote:
At the Mozilla All-Hands in SF we had some discussions about Mozilla's future plans on Android. Currently, Mozilla are actively supporting four Android app web browsers - Fennec[0], Focus[1], Rocket[2], and Firefox on Fire TV[3].
Focus, Rocket, and Firefox on Fire TV are all minimal browsers, and they do not have the same features as Fennec. Fennec is the codebase Orfox uses, and it is the browser we are currently planning on using for Tor Browser. However, Fennec is not being actively developed, but Mozilla are keeping Fennec alive and they do not have an EOL date.
The two important parts from this are:
- The Fennec code will not change much over the next 12+ months
- We can safely continuing using Fennec for the 12+ months
This effectively means Fennec 67 will be very close to Fennec 60ESR. Therefore, I propose we backout our current plan of following Fennec releases and we make this simple and follow ESR (exactly the same as desktop). We can track on-going changes in mozilla-central that affect Fennec, but these changes should not happen often - backporting any high priority patches shouldn't be difficult.
The two additional components I should mention are GeckoView[4] and Fenix[5]. GeckoView is an Android library acting as a glue-layer between the Gecko native code and the UI/Browser app layer. Firefox Focus is currenty using Android WebView (Android's built-in renderer), but they are moving Focus onto using GeckoView in the very near future. Fennec currently uses parts of GeckoView, so if we decide we should use the ESR then we should also watch the work done on GeckoView in case there are any improvements we want.
Fenix is the last piece in this puzzle. It is a new browser Mozilla will be working on, and it has a lot of promise - but currently it does not exist in a form we can use. We may want to consider using Fenix in 2-3 years.
Mozilla's Android Components project is also interesting. When it is ready, we could create our own browser by picking-and-choosing the parts we want. Unfortunately, this isn't as easy as hardening an existing private browsing mode. This is the reason I think we should continue working on Fennec right now.
In conclusion, I propose we develop Tor Browser for Android based on Firefox Fennec ESR. This is not an official ESR, but the changes between Fennec releases will be very small (if any changes at all). This will allow us maintain the same branches for desktop and mobile.
[0] https://play.google.com/store/apps/details?id=org.mozilla.firefox [1] https://play.google.com/store/apps/details?id=org.mozilla.focus [2] https://play.google.com/store/apps/details?id=org.mozilla.rocket [3] https://www.amazon.com/Mozilla-Firefox-for-Fire-TV/dp/B078B5YMPD [4] https://wiki.mozilla.org/Mobile/GeckoView [5] https://github.com/mozilla-mobile/fenix [6] https://github.com/mozilla-mobile/android-components _______________________________________________ tbb-dev mailing list tbb-dev@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tbb-dev
On Thu, Jun 14, 2018 at 10:00:46PM -0300, Igor Oliveira wrote:
Hi Matthew,
Thanks for the heads up. I have one question, Are they going to freeze even the Android SDK and NDK versions?
Mozilla are actively developing GeckoView, so they won't freeze the SDK/NDK versions. However, historically they've been careful about when they increase the toolchain and SDK versions, and they only increased it when they decided it was necessary.
On 06/14/2018 08:51 PM, Matthew Finkel wrote:
At the Mozilla All-Hands in SF we had some discussions about Mozilla's future plans on Android. Currently, Mozilla are actively supporting four Android app web browsers - Fennec[0], Focus[1], Rocket[2], and Firefox on Fire TV[3].
Focus, Rocket, and Firefox on Fire TV are all minimal browsers, and they do not have the same features as Fennec. Fennec is the codebase Orfox uses, and it is the browser we are currently planning on using for Tor Browser. However, Fennec is not being actively developed, but Mozilla are keeping Fennec alive and they do not have an EOL date.
The two important parts from this are:
- The Fennec code will not change much over the next 12+ months
- We can safely continuing using Fennec for the 12+ months
This effectively means Fennec 67 will be very close to Fennec 60ESR. Therefore, I propose we backout our current plan of following Fennec releases and we make this simple and follow ESR (exactly the same as desktop). We can track on-going changes in mozilla-central that affect Fennec, but these changes should not happen often - backporting any high priority patches shouldn't be difficult.
The two additional components I should mention are GeckoView[4] and Fenix[5]. GeckoView is an Android library acting as a glue-layer between the Gecko native code and the UI/Browser app layer. Firefox Focus is currenty using Android WebView (Android's built-in renderer), but they are moving Focus onto using GeckoView in the very near future. Fennec currently uses parts of GeckoView, so if we decide we should use the ESR then we should also watch the work done on GeckoView in case there are any improvements we want.
Fenix is the last piece in this puzzle. It is a new browser Mozilla will be working on, and it has a lot of promise - but currently it does not exist in a form we can use. We may want to consider using Fenix in 2-3 years.
Mozilla's Android Components project is also interesting. When it is ready, we could create our own browser by picking-and-choosing the parts we want. Unfortunately, this isn't as easy as hardening an existing private browsing mode. This is the reason I think we should continue working on Fennec right now.
In conclusion, I propose we develop Tor Browser for Android based on Firefox Fennec ESR. This is not an official ESR, but the changes between Fennec releases will be very small (if any changes at all). This will allow us maintain the same branches for desktop and mobile.
[0] https://play.google.com/store/apps/details?id=org.mozilla.firefox [1] https://play.google.com/store/apps/details?id=org.mozilla.focus [2] https://play.google.com/store/apps/details?id=org.mozilla.rocket [3] https://www.amazon.com/Mozilla-Firefox-for-Fire-TV/dp/B078B5YMPD [4] https://wiki.mozilla.org/Mobile/GeckoView [5] https://github.com/mozilla-mobile/fenix [6] https://github.com/mozilla-mobile/android-components _______________________________________________ tbb-dev mailing list tbb-dev@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tbb-dev
tbb-dev mailing list tbb-dev@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tbb-dev
Matthew Finkel:
At the Mozilla All-Hands in SF we had some discussions about Mozilla's future plans on Android. Currently, Mozilla are actively supporting four Android app web browsers - Fennec[0], Focus[1], Rocket[2], and Firefox on Fire TV[3].
Focus, Rocket, and Firefox on Fire TV are all minimal browsers, and they do not have the same features as Fennec. Fennec is the codebase Orfox uses, and it is the browser we are currently planning on using for Tor Browser. However, Fennec is not being actively developed, but Mozilla are keeping Fennec alive and they do not have an EOL date.
The two important parts from this are:
- The Fennec code will not change much over the next 12+ months
- We can safely continuing using Fennec for the 12+ months
This effectively means Fennec 67 will be very close to Fennec 60ESR. Therefore, I propose we backout our current plan of following Fennec releases and we make this simple and follow ESR (exactly the same as desktop). We can track on-going changes in mozilla-central that affect Fennec, but these changes should not happen often - backporting any high priority patches shouldn't be difficult.
The two additional components I should mention are GeckoView[4] and Fenix[5]. GeckoView is an Android library acting as a glue-layer between the Gecko native code and the UI/Browser app layer. Firefox Focus is currenty using Android WebView (Android's built-in renderer), but they are moving Focus onto using GeckoView in the very near future. Fennec currently uses parts of GeckoView, so if we decide we should use the ESR then we should also watch the work done on GeckoView in case there are any improvements we want.
Fenix is the last piece in this puzzle. It is a new browser Mozilla will be working on, and it has a lot of promise - but currently it does not exist in a form we can use. We may want to consider using Fenix in 2-3 years.
Mozilla's Android Components project is also interesting. When it is ready, we could create our own browser by picking-and-choosing the parts we want. Unfortunately, this isn't as easy as hardening an existing private browsing mode. This is the reason I think we should continue working on Fennec right now.
In conclusion, I propose we develop Tor Browser for Android based on Firefox Fennec ESR. This is not an official ESR, but the changes between Fennec releases will be very small (if any changes at all). This will allow us maintain the same branches for desktop and mobile.
Sounds almost like a best-case scenario for us, great. We should think about how we get timely information about possible security updates we need to backport. I guess we can create a weekly "monitor mozilla-central for possible security/privacy related backports"-role and/or I can try to get a more timely access to Fennec-only security bugs.
Georg
[0] https://play.google.com/store/apps/details?id=org.mozilla.firefox [1] https://play.google.com/store/apps/details?id=org.mozilla.focus [2] https://play.google.com/store/apps/details?id=org.mozilla.rocket [3] https://www.amazon.com/Mozilla-Firefox-for-Fire-TV/dp/B078B5YMPD [4] https://wiki.mozilla.org/Mobile/GeckoView [5] https://github.com/mozilla-mobile/fenix [6] https://github.com/mozilla-mobile/android-components _______________________________________________ tbb-dev mailing list tbb-dev@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tbb-dev
On Mon, Jun 18, 2018 at 12:44:00PM +0000, Georg Koppen wrote:
Matthew Finkel:
In conclusion, I propose we develop Tor Browser for Android based on Firefox Fennec ESR. This is not an official ESR, but the changes between Fennec releases will be very small (if any changes at all). This will allow us maintain the same branches for desktop and mobile.
Sounds almost like a best-case scenario for us, great. We should think about how we get timely information about possible security updates we need to backport. I guess we can create a weekly "monitor mozilla-central for possible security/privacy related backports"-role and/or I can try to get a more timely access to Fennec-only security bugs.
Yes, I agree. Igor and I can watch non-security-related tickets[0], but we'll need your help or Tom's help with getting the security-related tickets.
[0] https://bugzilla.mozilla.org/buglist.cgi?list_id=14195161&order=Importan...
On 06/18/2018 11:09 AM, Matthew Finkel wrote:
On Mon, Jun 18, 2018 at 12:44:00PM +0000, Georg Koppen wrote:
Matthew Finkel:
In conclusion, I propose we develop Tor Browser for Android based on Firefox Fennec ESR. This is not an official ESR, but the changes between Fennec releases will be very small (if any changes at all). This will allow us maintain the same branches for desktop and mobile.
Sounds almost like a best-case scenario for us, great. We should think about how we get timely information about possible security updates we need to backport. I guess we can create a weekly "monitor mozilla-central for possible security/privacy related backports"-role and/or I can try to get a more timely access to Fennec-only security bugs.
Agreed! Seems like staying the course, with a bit more resources and attention to high priority issues, makes the most sense here!
+n