Firefox Lite

xunxun xunxun1982 at gmail.com
Wed Jul 8 03:25:50 UTC 2015


If you mean that we should remove some modules from firefox, in the past, I
tried the way:
https://groups.google.com/forum/#!topic/mozilla.dev.builds/6Qw7JhBLBvQ

I plan to generate the patches again based on the newer source code, if I
think one edition is stable enough.

2015-06-29 10:29 GMT+08:00 Eric Rescorla <ekr at rtfm.com>:

>
>
> On Sun, Jun 28, 2015 at 5:11 PM, M V <mvocom at gmail.com> wrote:
>
>> Hello JS4Lf, Eric, Daniel and David,
>>
>> Thank you for replying and for the useful information.
>>
>> 1) Each component in itself may indeed not be a primary source of memory
>> consumption. All of them combined might be significant.
>>
>
> They might be, but that's not the expectation of the people who work on
> this
> code. If you can provide evidence that some set of these features is
> consuming a significant amount of memory, we can take a look at that.
>
>
>
>>     Please don't get me wrong: all the features I have mentioned are
>> great, but why impose them on those who never use them?
>>
>
> Because while any particular feature may not be important to one person,
> that doesn't mean that one of the other features aren't, so it's not clear
> that there's a subset that would have a significant size reduction and also
> be useful to a big enough fraction of people. For instance, I don't use
> Pocket but I use WebRTC.
>
> Moreover, making any kind of subset adds complexity for development,
> build, and test.
>
> 2)  IMO, the binary size matters as well.
>>
>> 3) I mentioned ABP because of its popularity and high memory consumption.
>> But even if (or rather when; thank you all who are working
>>     on it) this issue is resolved, the basic question remains: why not
>> adapt Firefox to the vast majority of users?
>>
>
> I think there's a natural presumption that having all the features
> available does
> that.
>
> -Ekr
>
>
> Best regards.
>>
>>
>> 2015-06-29 1:00 GMT+03:00 Daniel Holbert <dholbert at mozilla.com>:
>>
>>> Someone posted basically the same request last year, here:
>>>   https://groups.google.com/d/msg/firefox-dev/otM2HQy_QXI/z5pKvsMjCPcJ
>>> Please see the responses in that thread. Basically, the features you
>>> mention are (or should be) designed to be "lazily loaded", so they don't
>>> consume any memory at all until they're used. So, removing them wouldn't
>>> actually help.
>>>
>>> We do have people working on finding & fixing memory leaks and reducing
>>> memory usage, for what it's worth. (And in many cases, add-ons like ABP
>>> are partly to blame; see for example this post from last year:
>>>
>>> https://blog.mozilla.org/nnethercote/2014/05/14/adblock-pluss-effect-on-firefoxs-memory-usage/
>>> )
>>>
>>> On 06/27/2015 05:38 PM, M V wrote:
>>> > Hello,
>>> >
>>> > FF 39 consumes approximately 240 MB on start-up (with ABP installed).
>>> >
>>> > The average user doesn't use any of the Dev Tools.
>>> > Also, many users never use WebRTC, Pocket, Reader etc..
>>> >
>>> > How about offering users Firefox Lite?
>>> > - Either a separate FF version, or providing the option to choose the
>>> > components to install.
>>> > I'm quite sure it would enlarge Firefox market share.
>>>
>>
>>
>> _______________________________________________
>> firefox-dev mailing list
>> firefox-dev at mozilla.org
>> https://mail.mozilla.org/listinfo/firefox-dev
>>
>>
>
> _______________________________________________
> firefox-dev mailing list
> firefox-dev at mozilla.org
> https://mail.mozilla.org/listinfo/firefox-dev
>
>


-- 
Best Regards,
xunxun
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.mozilla.org/pipermail/firefox-dev/attachments/20150708/a16d1dcf/attachment.html>


More information about the firefox-dev mailing list