Downloads of old Firefox versions
Henri Sivonen
to the downloads of Firefox 3.6 and that even older versions are being
downloaded thousands of times a day.
There can't be *that* many regression testers and Web authors
downloading old versions for testing.
Is it known where those downloads are coming from? Are people following
old links from all over the Web and ending up installing obsolete
software?
Wouldn't it make sense to redirect all links to old downloads to a page
saying something like the following?
"You followed a link to download an obsolete version of Firefox. You
should download the latest version (link) instead in order to stay safe
online. If you really intended to download an old version for testing,
we have an archive of old releases (link)."
--
Henri Sivonen
hsiv...@iki.fi
http://hsivonen.iki.fi/
Jean-Marc Desperrier
> Wouldn't it make sense to redirect all links to old downloads to a page
> saying something like the following?
I'd add a text asking the user to report on input.mozilla.com any
problem that pushes him to stay on an older version instead of upgrading.
I think input.mozilla.com could have a dedicated section for that,
although the default help page is almost adequate already.
Jesper Kristensen
I have wondered about this same thing for a long time. As a localizer I
get download stats for downloads in my country by email once a week.
Last week the Firefox downloads for my locale in my country showed:
Firefox 4.0 4,145
Firefox 3.6 35,956
Firefox 3.5 5,063
beltzner
confirm) that many of these downloads are going through bouncer, so the bugs
should be filed against that component. Additionally we might want to put a
rule or redirect in the FTP directory if that is the source for most
downloads.
cheers,
mike
On 26/04/2011 7:26 AM, "Jesper Kristensen" <
> _______________________________________________
> dev-planning mailing list
> dev-pl...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-planning
Henri Sivonen
> These sound like great bugs to file. I suspect (but we can get stats to
> confirm) that many of these downloads are going through bouncer, so the bugs
> should be filed against that component.
Bug filed https://bugzilla.mozilla.org/show_bug.cgi?id=652789
> Additionally we might want to put a
> rule or redirect in the FTP directory if that is the source for most
> downloads.
I agree. Should be possible for http://ftp.mozilla.org/. Dunno if ftp://
can do redirects. I wonder if ftp:// should have a text file telling
people to use http:// instead.
JarrE
> I was surprised to learn that the release of Firefox 4 didn't put a stop
> to the downloads of Firefox 3.6 and that even older versions are being
> downloaded thousands of times a day.
What du you find surprising?
I administrer more than 10k computers, most of which have Firefox.
I am not going to ship Firefox 4 to these computers for still some time
* many users will have to upgrade add-ons and extensions and those
upgrades are not necessarily in place yet
* internal systems does not work as they should with a new version
* it is often better to use a stable version than a new one
Maybe you should think about 'Why do 90% of our downloaders prefer an
older version which is not the one that is easiest to download?' in other
terms than 'We need more ways of redirecting users to what we will them
to download (this is where you will go today-thinking)'?
Regards/
JarrE
Wes Garland
Presumably you are at least shipping 3.6, though? The stat that I find
surprising that 3.5 was downloaded more for this guy's locale than 4.0.
Wes
--
Wesley W. Garland
Director, Product Development
PageMail, Inc.
+1 613 542 2787 x 102
JarrE
>> I am not going to ship Firefox 4 to these computers for still some time
>
> Presumably you are at least shipping 3.6, though? The stat that I find
> surprising that 3.5 was downloaded more for this guy's locale than 4.0.
>
> Wes
Can agree to that being a surprise, though 3.5 is still maintained.
Someone knowing about any feature in 3.5 that is missing in 3.6? (I can't
remeber any personal loss at that shift)
I still have 18 Firefox 3.15 to get rid off, but yes; I want what I
believe is best for the majority of my users.
My guess today is 3.6.16 so that's what they get...
JarrE
Henri Sivonen
> On 26.04.2011 11:10, Henri Sivonen wrote:
> > I was surprised to learn that the release of Firefox 4 didn't put a stop
> > to the downloads of Firefox 3.6 and that even older versions are being
> > downloaded thousands of times a day.
>
> What du you find surprising?
I'd have expected the downloaders to be either people who have Firefox
and want to upgrade or people who are new to Firefox and, therefore,
shouldn't have a preference for a specific old version.
I'd have expected there to have been a handful of people who download
old versions for testing site compatibility in old versions.
> I administrer more than 10k computers, most of which have Firefox.
> I am not going to ship Firefox 4 to these computers for still some time
> * many users will have to upgrade add-ons and extensions and those
> upgrades are not necessarily in place yet
> * internal systems does not work as they should with a new version
> * it is often better to use a stable version than a new one
If you installed Firefox on a new computer, would go to Mozilla and
download an old version or would you use some kind of system image that
you can put easily on thousands of computers? If you'd download from
Mozilla, would you download 3.6 or an older version? The daily downloads
of 3.5, 3.0 and 2.0 are in the thousands.
> Maybe you should think about 'Why do 90% of our downloaders prefer an
> older version which is not the one that is easiest to download?' in other
> terms than 'We need more ways of redirecting users to what we will them
> to download (this is where you will go today-thinking)'?
Is there any evidence that the downloaders are showing a preference as
opposed to just following old links from somewhere?
Ron Hunter
is also prone to cause users to either delay updating (as you
described), or to eschew new features in order to preserve the
functionality of an extension that is no longer supported. Perhaps
someone needs to give some thought to determining how many people are
using what extension on 3.6 that doesn't work on 4.0. There are only
two extensions that I consider indispensable, and they, fortunately, are
very good about updating. Others may not be so lucky.
Asa Dotzler
This is most likely the case. I suspect we have links out in the wild
pointing to older versions. We should do an analysis of incoming
traffic, hunt down those links, and update or kill them.
- A