Experiments status + experiments proposal

Mark Banner mbanner at mozillamessaging.com
Mon Apr 11 08:47:03 UTC 2011

On 10/04/2011 22:03, Jonathan Protzenko wrote:
> I also suggest two more experiments that are, imho, simple enough that 
> we could try to land them in the 3.3 timeframe.
So I think generally we need to stop thinking of landing things in "xxx" 
timeframe and just work out what our priorities are and work towards 
getting them landed when they are ready. If we keep tagging things as 
landing in "xxx" then we'll keep potentially delaying "xxx" until 
they've landed, and then we add more into "xxx" and "xxx" never gets 

> _Compose in a tab_ is making progress, although at a slow pace. It is 
> such a massive undertaking that it will probably require more than 
> just one person working on it to make it happen, that is, if we want 
> to make it happen, and I'm still not sure of that :-). So no plans for 
> landing there either.
It sounds like we need to get together and discuss the issues around 
what you're doing in more detail if that's the case.
>         *Unknown status*
> - _Thunderbird Sync (a.k.a. Weave for Thunderbird)._
Whilst I believe the basics are there, it needs more work to let it work 
with trunk, and I believe it needs some UI of some kind or some work on 
the UX for initial syncing of preferences. At the moment we're working 
on other things, but it'd certainly be something to pick up again.

>         *Experiments slated for landing*
> Jim (:squib) has been doing fantastic work on the _account summary and 
> folder summary_ patches. These experiments have been revived: they've 
> been lying there for quite some time, but there was no one to pick 
> them up. IIRC, Jim said it required a whole lot of work to make this 
> work properly, and I think he expects to land this in time for 3.4 
> rather than 3.3.
Getting just one of those landed as soon as its ready would be a good 
thing IMO. I'm not sure about the amount of work - I thought the account 
summary was almost ready (Jim?)?
> _OpenSearch_ and _GetAnAccount_ are two addons that are expected to 
> land within the 3.3 timeframe. That's awesome.
We're hoping those add-ons can land in before 3.3, but some parts are 
wrapped up in external dependencies. We'll be working through those as 
soon as we can.

> I think mconley's excellent work on Ubuntu integration is expected to 
> land in 3.3.
I believe that the significant functionality here is going to stay as 
extensions that Ubuntu integrates into Thunderbird with their build 
mechanism. I'm sure Mike will correct me if I'm wrong.
> Can we do more? I believe the answer is yes!
I think that's the wrong question... see above.

>         *Suggestions for new experiments*
> 1) A *very* cool addon created by davida is _webtabs_ (build at 
> http://jonathan.xulforum.org/files/webtabs@momo.xpi, source at 
> http://hg.mozilla.org/users/dascher_mozilla.com/webtabs/). It allows 
> you to register some websites, and have a small favicon appear for 
> each one of them in the tab bar, next to the quickfilter button. 
> Clicking the favicon opens a content tab for the given website. I've 
> been using it for months and it works just perfectly.
> How does that sound to you? I feel this could be a very valuable 
> addition to Thunderbird. Do other people feel the same way? Is it 
> realistic to expect this for Thunderbird 3.3?
iirc we're not quite convinced that's the right UX for Thunderbird users 
for what we want to achieve there. I certainly feel that'd it need some 
bigger UX rework before we could land it if that's what we want to do. 
Of course, there's nothing to stop this being experimented with as an 
add-on to gain wider feedback.

> 2) Another idea of mine that I've been playing with is _RSS Tab_ 
> (screenshot at http://twitpic.com/4hxz6j/full, source at 
> https://github.com/protz/Rss-tab, build at 
> http://jonathan.xulforum.org/files/rsstab-nightlies/201104091950-master.xpi).
So I've not looked at this in detail yet, but it feels like something we 
should have as an experiment for a little while to gain wider testing. I 
can also think of some different ways of where to present that pane and 

I think this is a useful discussion/update, but I really think that 
especially wrt Miramar we need to be thinking about get our blocker list 
down to zero and for it ready to be shipped before we start thinking 
about adding too many more new UX features. There's a good platform 
update in it, and until we ship it, things like Conversations and the 
other experiments that depend on Miramar won't get wider usage/feedback 
because of the dependencies.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.mozilla.org/pipermail/tb-planning/attachments/20110411/1e973032/attachment.html>

More information about the tb-planning mailing list