I just readed an article on cnet, about the futur of Firefox:
http://news.cnet.com/8301-30685_3-10378604-264.html. And we can learn:
"the current plan is to drop the older add-on technology with Firefox 4."
Is it true ?? Do you really want to drop it ?? Do you really want to
kill some powerful extensions like Firebug, SQLite manager etc ??
If yes, this is a terrible mistake, IMHO !!
See also this comment from an other extension developer:
http://news.cnet.com/8618-30685_3-10378604.html?communityId=2140&targetCommunityId=2140&blogId=264&messageId=8564955&tag=mncol;tback
Laurent
Cheers,
Shawn
On the other hand Laurent, if Firefox does go to multi-process,
something has to change. The current addon approach relies on a single
process model.
It seems more likely that the world of addons will split into "almost
part of the platform" like addons which will use the same technology as
the platform, and jetpack based addons which use the cheap better faster
solution.
jjb
> It seems more likely that the world of addons will split into "almost part of the platform" like addons which will use the same technology as the platform, and jetpack based addons which use the cheap better faster solution.
That's mostly right, yes.
There are presently no plans for removing support for chrome overlays packaged as XPIs in Firefox. To do so would not only foolishly throw away the efforts of hundreds as represented by our current Add-ons library, but would also fly in the face of our commitment to allowing users to take full control of their browser.
At the same time, the Jetpack project exists with the explicit goal to deprecate and provide a better solution for the large number of Add-on developers who would be better served by a solution which allows for no-restart-required installation, an API surface which can be used to maintain compatibility across versions, and a far simpler development model. As Jetpack progresses, it will attempt to provide the required tools to allow the vast majority of today's Add-ons to be implemented as Jetpacks - that's a design goal. It's our hope that this will help solve some of the technical and development challenges that our Add-on authors face.
cheers,
mike
Ok, so the article is wrong. It's a good news :-)
>
> At the same time, the Jetpack project exists with the explicit goal to deprecate and provide a better solution for the large number of Add-on developers who would be better served by a solution which allows for no-restart-required installation, an API surface which can be used to maintain compatibility across versions, and a far simpler development model. As Jetpack progresses, it will attempt to provide the required tools to allow the vast majority of today's Add-ons to be implemented as Jetpacks - that's a design goal. It's our hope that this will help solve some of the technical and development challenges that our Add-on authors face.
>
Jetpack is really good thing, and I'm happy for this kind of extensions
system, we need it. So I'm not against Jetpack, but Jetpack couldn't
satisfy all developpers, since it is like a sandbox, with an API, so it
is naturally limited, and offer a restricted environment (from the
point of view of the developer).
Jetpack and chrome extensions are complementary. Removing the
possibility to create chrome extensions, is to remove the possiblity to
access to the power of the Mozilla platform, so it will kill many
projects, and will kill one of the biggest advantage of Firefox over
other browsers. (And bye bye the "hackability" moto..).
But since you said this removing is not in the plans, I am reassured :-)
We could continue to create powerful extensions !
Thanks Mike !
Laurent