At 13:08 08/12/2000 +0100, Ben Bucksch wrote:
>Simon P. Lucy wrote:
>
>>For those users that stray into mozilla.org user groups they need to be
>>steered correctly into the right channel for their distribution.
>
>As I said, m.users won't help us here. They will either go to
>netscape.communicator or m.apps.mailnews. Why should they go to m.users?
Why shouldn't they? I think its meant to be mozilla.users.general it would
likely get the same kind of traffic, and worse as mozilla-general. I'm not
suggesting that all support happens there, I am saying that distributors
should support their own distributions, hence
http://www.objective2k.co.uk/goodmozillauser.htm
>>If a user picks up a milestone or nightly release (and by doing so
>>they've qualified themselves as being different to a casual user), then
>>they should get supported (and currently do), within mozilla.org.
>
>There is *no* Mozilla user, not even the above-average user. There are
>only sincere testers and developers. Everything else is misplaced and we
>shouldn't even start to support it. The only alternative I see is to
>completely change the policy of mozilla.org.
No I don't think that's true, you can have sincere users as well as sincere
testers :-)
>>Nor do I think that users should be dissuaded from getting nightlies
>>and milestones because its only by having a disparate and random
>>distribution that we will get the many eyes and hands that are needed to
>>show bugs and uncover feature requirements.
>
>We have 200 people regularly downloading nightlies. There are 150 netscape
>employees and 300 relatively constant non-Netscape code-contributors, I
>guess. That's a lot of randomness. Our problem is not too few bug reports,
>but too few developers to fix them.
The kinds of things that regular users find are of a different order than
directed testers, or those that are experienced. Regular users tend to be
less organised in their bug reports, more discursive and buried in a few
paragraphs of wishing the world was a different shape to the one they think
it is. Not just 'how do I' but 'why doesn't it' or 'why does it' . Now
its true that a lot of that requires FAQs, unfortunately those same users
are unlikely to read FAQs or be able to relate them directly to their own
experience. That isn't because regular users are dumb, lazy or malicious,
but that for the most part they aren't empowered to make changes in the
software they use. That's one of the advantages of open software and to be
honest probably the only worthwhile one. But it does require work.
If you make the software open to only developers all you've done is to put
a nice canvas covered extension on the side of the Cathedral and called it
the Bazaar.
>Currently, there are not many users using Mozilla. As soon as Mozilla gets
>more stable, more famous and we start supporting users (i.e. give the
>impression that Mozilla is for users), there will be so many that they are
>not helping with bug-reporting anymore (because they report only very
>visible bugs and don't care about the rest), but consume a lot of
>resources for help. I didn't see many useful bug reports from Beonex
>users. But I spent a *ton* (too much) of time answering FAQs.
I understand that, I just think that waving FAQs at users and expecting
them to use them is missing the point. We need open support as well as
open source. That, naturally, doesn't mean free.
>To stress my point again: I think, there should be an open-source
>distribution of Mozilla for users. Either
>- mozilla.org denies users and Beonex (and maybe others) supports them or
>- mozilla.org completely changes its policy, website etc. and welcomes
>users, and I pack my things with Beonex.
>If mozilla.org supports users here and there a bit (i.e. as suggested),
>- we will continue to see clueless posts on development newsgroups
>- Beonex will have to "compete" with Mozilla, which is absolutely pointless.
>
>Any non-strict policy causes confusion, and any confusion on the users'
>side costs us a lot of time.
I don't think it matters whether there is a strict policy or not, users
will still wander into places that aren't designed for them. We need
mechanisms that direct them to the right place.
To restate this. I fundamentally agree that distributors should support
their own distribution, unless mozilla.org refrains from producing
binaries (which is possible), mozilla.org is de facto a distributor and
should support its own builds. Now I think it should encourage possible
users to take an end user distribution, either Netscape 6 or Beonex at the
moment, rather than a nightly release, however, I do think it also needs to
support distributors properly (as in the security issues you've brought up
elsewhere).
I think the only disagreement we have is how we manoever the user into the
correct channel for support.
Simon
