I like the idea of lightening your support load, however it might be worth thinking about why customers run older versions of Firefox...
In my experience, the most common reason is limited memory (resources). Newer versions of Firefox consume a lot of memory, sometimes upwards of 4GB if left running (my FF installation has used upwards of 8GB) when being used heavily for more than a day. Think about a senior (that's what, anybody over 50-60?) working on a web based genealogy program all day using their 5-10 year old computer and OS. A machine like that might only have 2-4GB of RAM and Firefox wants all of it and more! (and will want more next year) This reason alone prevents upgrading older versions of FF. There is no way to scale back newer versions of FF appetite for memory. Solution: stick with an older version of FF (or switch to Chrome), which is what people in this situation do. They can't afford, or don't want to keep throwing away and replacing their machines every couple years just to stay on the bleeding edge of IT while developers continue to tout, "this is the future". This is "today" and the rest of us would like to get some work done in a timely manner. How about supporting a Firefox Lite branch which doesn't gobble up all an older systems resources? This might work better than trying to force a "one size is going to fit all", and sorry pal but you're going to have to upgrade everything, regularly, approach. I'm not one of these people, but I end up supporting friends and family who are. They can't afford to, or justify dropping 3 grand for a new Surfacebook with 16GB RAM, like I do (btw, I'm one of those "seniors"). Nor do they want to sacrifice the -days- it can take for some of these "upgrades". They'd love to, but can't afford it, it's not in their budget, yes, really. Real people who don't work or spend a lot of time in the industry are using older more limited machines. Fact of life, and this won't change. Specifications are important. Make tools (browsers) work to specifications. LibreOffice/OpenOffice come to mind as good examples of this "work to a specification" philosophy. I can pull up a less featured 3.x version of Libre Office (or OpenOffice) and work with a document produced using the latest 5.x release. On an older machine I might -need- to run 3.x so it will run in the available memory. Both applications work because the file format written to disk is an ISO specification. It takes a -lot- of work to make a "good" light weight specification, but it's worth it. I'm one of those people who now runs Pale Moon for many things because it still supports the Sync 1.1 protocol (specification) which is what works reliably with my infrastructure. I haven't found a need for FxA yet, and current versions of FF don't support Sync 1.1. I could go on with a lot more examples, but won't for everyone's sake. Hey, let's talk about the elephant in the room. Don't get me wrong, Firefox is an amazing product and I can't wait to see what will happen with v57, I'm sure it will be really cool! Craig On 9/14/2017 7:36 AM, Shane Tomlinson wrote: > Firefox 29 is the first version of Firefox desktop that used Firefox > Accounts > to sign in to Sync [1]. Firefox 29 was released on April 29, 2014. > > In the nearly 3.5 years since Firefox 29 was released, Firefox has aged by > 26 full releases. Perhaps surprisingly, FxA still officially maintains > and supports sign in to Firefox 29. I just tried to make sure I wasn't > going > to have to eat those words. It works. Yay us. > > The thing is, as FxA adds more features that are only supported on > Fx >= version N, maintaining support for these old browsers is becoming > increasingly complex, time consuming, and expensive. Our full > functional test > suite now takes an hour to run, many of the tests are for browsers > which have > been unsupported by Mozilla for some time. > > I propose we purposely lose some of the extra baggage by officially > dropping > support for Firefox <= current ESR - 1. Why current ESR - 1? Well, Mozilla > officially supports Firefox back to ESR. There are a bunch of > companies that > lag behind even that, and since we are a nice group that doesn't like > to anger > folks, we'll support 2 full ESRs. Current ESR is based on Firefox 52. The > previous ESR was based on Firefox 45. We'd officially support down to > Firefox 45. > > For users that try to sign in to FxA on these old browsers, we could > show some > nice screen that says something along the lines of "Hey, sorry to do > this to you, > but your browser is really really behind the times. For your own > safety, here's > a link the latest and greatest." > > Thoughts? > > Shane > > > [1] - https://www.mozilla.org/en-US/firefox/29.0/releasenotes/ > > > > _______________________________________________ > Sync-dev mailing list > [email protected] > https://mail.mozilla.org/listinfo/sync-dev
_______________________________________________ Sync-dev mailing list [email protected] https://mail.mozilla.org/listinfo/sync-dev
