Ted Husted wrote:
On 5/6/06, Don Brown <[EMAIL PROTECTED]> wrote:
Please don't judge until we are done.
Silence can be taken to imply consent, and if people with questions
don't speak up now, then other people might be disappointed later,
when a binding veto lands on the table.
We need to first agree on our
design goals, then we can pick through implementations. The proposal is
about agreeing to the laid out design goals, but it is too early to say
that we are or aren't meeting them. First cuts are never put into
production as is, and this API is no different. Besides, even if we did
finally go with something more sweeping, to make it acceptable under the
original agreed upon proposal, we'd have to ensure existing apps would
be able to migrate in hours so you can be sure the end user will have a
smooth migration. Otherwise, we'd have to move the API to a new version.
+1
Under the original proposal, the community agreed that the goal is for
Action 2.0 to be a smooth *transition* for WebWork developers and as
smooth a migration as we can make it for Action 1 developers.
Of course, we all want to continue to improve the framework, and under
the original proposal we agreed that more ambitious improvements are
to take place in a second phase, which may be a 2.1 or 3.0 series.
It's healthy to have these discussions, but we do not want to create
unrealistic expectations of what can be accomplished in the SAF 2.0
timeframe.
What we need here is cooperation and a willingness to compromise and
follow a common path that may not be exactly what everyone wants, but it
is what the community agreed is best for the project and its users.
True, and the community has already agreed that Action 2.0 will be a
transitional release. If we agree that some or all of the proposed API
changes are transitional, and if someone actually comes up with a
patch, and if we accept the patch, then fine. But there is no
shortages of *ifs* in that statement. We do *not* want to give people
the impression that by having these discussions, we have decided to
change course and revoke the original agreement.
My
guess is you will always think a new API goes too far, Bob will think it
doesn't go far enough, and the rest of the developers will lie along
that spectrum. We could either fight every step of the way to force our
point of view, or we could work together to find a common vision. I
think we owe it to our users to do the latter.
Hmmm, I can't think of a *user* to whom we owe more than Jason -- or
Patrick -- or Toby -- or Rainer -- or anyone else who not only *uses*
the framework but also makes concrete contributions to the code and
documentation. We all are users.
The people mentioned above surely do use Webwork. However, they are not
regular users in the sense of this discussion. When you know something
and its internal logic and so on as well as these people do, things that
are unintuitive and confusing to regular end &users may come naturally
to you. There is a real hazard in this, I think.
In some recent private correspondence, somebody commented to me that a
lot of ASF projects seemed to be "developer-centric", rather than
"user-centric" and that this had some very negative consequences in
practice.
Well, is this dichotomy between a "developer-centric" and "user-centric"
really a legitimate difference of viewpoint? It seems to me that any
software development project should be user-centric. Similarly, a book
writing project should be reader-centric and the development of a
consumer device should be consumer-centric.
It is my considered opinion that, without that kind of relentless focus
on what is useable by and helpful to other people, a project is at
extreme risk of degrading into some kind of exercise in intellectual
masturbation.
In any case, I (unlike some people ;-)) do not presume to know
everything there is to know about open source software development.
However, I have certain basic beliefs about this and some experience to
know whereof I speak. On the basis of this, my sincere advice is that,
if you want to do something that's useful, focus on convenience for the
broad user base, not your own convenience.
Regards,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
-Ted.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
