Thunderbird mailing-list / NNTP decision rationale
dmose at mozillamessaging.com
Tue Aug 3 21:57:26 UTC 2010
It has become very clear from recent discussions that how much (if at
all) a person is bothered by the lack of a newsgroup presentation of
Thunderbird-related newsgroups depends heavily on personal working
style. For many folks, the mailing list presentation works, if not
well, at least Well Enough. However, it's also clear that there are
non-trivial usability issues in the tb-planning community with the
current email-only presentation (mostly in Thunderbird, but also in a
few other clients).
Here's a short summary of the most important interests of multiple
people in the tb-planning community that are not met by the current
email-only presentation of group conversations:
* they are not automatically separated from the Inbox, which is
experienced as clutter
* the unread count for them is propagated to the folder pane, dock, and
biff, which feels like clutter and is distracting
* it's difficult to interact with historical or in-progress discussions
* there is a barrier to entry and exit for participation
* some tools for dealing with volume (watch, ignore, ...) are missing or
inferior to that of newsgroups
we've said that we're targeting individual and SOHO (Small Office /
Home Office) users. We've also said that we'll be doing this by
"focusing on conversations that occur over mainstream and emerging
communication channels. These include email, web forums, social
networks, and microblogging services."
All of the interests in the list above are not only our own interests as
a development community, they are also shared by our users when they're
having conversations over email. And those rough edges of handling
conversations over email are part of our mainstream users' day-to-day
experience of our product.
While it makes a lot of sense for us as the development community to
enjoy the benefits of the NNTP experience today, it is also acting as a
crutch: because we're not eating our own dogfood, it's preventing us
from having a deeper understanding of how those problems feel to our users.
This in turn makes it much harder for us to make the gut prioritization
choices for solutions to these issues both individually and as a group.
And when I say individually, what I'm trying to communicate is that
while some amount of feature work and fixes happens as a result of
driver-led prioritization, lots also depends on decisions made by
individual community members. In particular, it depends on the
decisions of individual developers, triagers, bug filers, and designers
about "what feels like the next most important thing for me to spend my
own time on?".
So I'd like to be clear that at this point I still feel that the
decision to leave the NNTP gateways off is the right one, though not an
easy one. I understand that this requires some work flow changes and,
even with those, it won't be entirely comfortable, and I'm sympathetic
to that. But it has the advantage of serving the interests of _both_ us
in the development community _and_ the interests of our users, rather
than serving the interests of the development community while working
against the interests of our users.
It's not impossible that I could be convinced to change my mind, but at
this point, I feel like the bar is pretty high. I have a decent
understanding of the pain points, and for me to change my mind, I'd have
to get new information that made me believe that creating those gateways
really would serve the overall goals of Thunderbird better.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the tb-planning