Upcoming changes to Firefox

194 views
Skip to first unread message

Will S

unread,
Aug 21, 2015, 11:14:49 AM8/21/15
to zotero-dev
I know the Firefox extension signing issue has already been a pain, but how much more work are these other changes going to be for the Zotero developers? As a worst case scenario the Firefox add-on could be replaced by something like the Chrome connector though hopefully nothing that drastic is needed. The deprecation of XUL should affect Gecko and thus Zotero Standalone as well, right?

https://blog.mozilla.org/addons/2015/08/21/the-future-of-developing-firefox-add-ons/


Message has been deleted

Emiliano Heyns

unread,
Aug 22, 2015, 12:32:08 PM8/22/15
to zotero-dev
Much of the logic is UI independent, but the way it's called is not wholly decoupled from it, so this is a whopper - bigger than the promisification I'd venture to guess.

So at what point are people going to bite the bullet and just port to chrome? With the loss of xul, and the "gain" of signing, staying with mozilla could be more effort than porting away.

Will S

unread,
Sep 4, 2015, 6:39:19 PM9/4/15
to zotero-dev
I am not that familiar with Chrome extensions (I guess I am going to have to learn soon), but my understanding that they can not support all of Zotero's features.

I just wanted to post a follow up to my speculation about Zotero Standalone. It looks like XULrunner which it is built on is also on shaky ground based on the discussion here:

https://groups.google.com/forum/#!topic/mozilla.dev.planning/YpRxW0gxYyc

Frank Bennett

unread,
Sep 4, 2015, 8:07:36 PM9/4/15
to zotero-dev
Yeah, that sucks too.

I've just finished porting Juris-M and its supporting plugins to Standalone, and should have builds up sometime on Monday (JST). The alternative would have been for the entire suite to begin breaking in default installations from September 22nd, and for Juris-M to begin shedding users from that date. I submitted the plugin version for approval, but the Mozilla robot inspector times out without even offering manual review. My best shot was to file a bug report on the timeout, but I have no idea where it is in the queue, or whether anyone will even look at it.

The Standalone builds will buy some time, at least, and I'm blindly hoping that my locally fetched binaries of xulrunner will continue to run for good while.

Emiliano Heyns

unread,
Sep 5, 2015, 4:36:13 AM9/5/15
to zoter...@googlegroups.com

This is true, but then that is a limitation that we're going to have to live with *anyhow*, as Mozilla wants to replace xul with exactly that api.

And with xulrunner gone, chrome is a better platform anyhow, as several chrome-based app shells have sprung up that don't require begging for new maintainers. That main problem will be that sqlite isn't available outside xul(runner), but I have some solid ideas on how to deal with that.

--
You received this message because you are subscribed to a topic in the Google Groups "zotero-dev" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/zotero-dev/G81HPRgc8sc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to zotero-dev+...@googlegroups.com.
To post to this group, send email to zoter...@googlegroups.com.
Visit this group at http://groups.google.com/group/zotero-dev.
For more options, visit https://groups.google.com/d/optout.

Emiliano Heyns

unread,
Sep 5, 2015, 4:43:15 AM9/5/15
to zoter...@googlegroups.com

You can run the validator on your own system - I'll have a look, but this must be submitted as a bug to amo. Not at all a fan of their ham fisted signing mandate, but if they're going to be doing this, they should make sure it works. If yours is a breaking bug vorige the validator that is hard to fix and they have to push back the enforcement date, so much the better.

--

Emiliano Heyns

unread,
Sep 5, 2015, 5:16:19 AM9/5/15
to zoter...@googlegroups.com
You'll find a full report at https://gist.github.com/41a2f9a9621aa1bf40f4; it is in principle possible to have juris-m pass for automated signing by abusing known deficiencies in the validator (it's none too bright, and easily fooled), but that's a cat-and-mouse game. Mozilla is going to catch on, and either make its validator smarter (not likely) or more ham-fisted (*much* more likely).

--
Reply all
Reply to author
Forward
0 new messages