[openbeos] Re: Tracker icons
- From: Michael Phipps <mphipps1@xxxxxxxxxxxxxxxx>
- To: openbeos@xxxxxxxxxxxxx
- Date: Sun, 24 Jul 2005 01:23:08 -0400
I don't understand, after all of the times that this has been explained, how
this could *still* not be clear.
GE was never intended to be, and never will be anything more than a discussion
group. The *GOAL* of GE is to talk about, consider critically, and recommend
ideas for R2.
Why is that? Well, a lot of reasons. GE is a place for non-devs and devs to get
together. It is seperate from the main list because there are a lot of devs who
find that list to be too much. And there were many, many very long emails,
email threads, etc, justifying the seperation of the two lists. GE is a place
to talk about crazy, way out ideas that are *so* not in scope for R1.
Another reason for GE to be talk only is that, unlike most OSS projects, we
don't have an attitude of "if you want it, code it". Honestly, I think that a
lot of the problem that most OSS has comes from that attitude. I have met many
people in the community who can't code but do have good ideas. GE is a way for
them to be involved.
Thirdly, strictly speaking, any conversation about what R2 is and is not has to
come from and involve the people who will be writing it. As stated above, I
think that it is a flaw to have the developers only dictating the ideas. But I
think that it would be an equal flaw (one that I think that a lot of commercial
companies fall into) to have the users dictate the direction. This is sort of a
bicameral approach.
Fourthly, even *if* we were into integrating new features, I think that
planning for them and integrating into the overall package is better than
piecemeal; that is what bloats and overly complicates other systems. XML is a
prime example. SVG uses it. SOAP uses it. If you are going to include all three
of those into an OS release, it would be wise to have the SVG and SOAP
implementations use the XML implementation, rather than the way some other OS's
do things. The GE approach of putting everything sort of on a virtual
whiteboard, letting people take potshots at the ideas and coming up with
something that is the best of everyone's ideas is generally considered a
software best practice.
So. A focused R1 is a good thing. A place for devs and non-devs to come
together and talk about features for the future is a good thing. I don't quite
understand why there should be some sort of an artificial timetable for
deciding what ideas from GE should "move up the ladder". That decision is *far*
better left deferred until it needs to be made, when all of the ideas are on
the table in a clear and coherent fashion. FWIW, this is how all of the
software development companies that I have ever worked with do their product
planning; they gather requirements and desires from customers. When they have
everything put together and it is time to plan a release, they get all of the
stakeholders in the room, decide what will go in and what won't. I have yet to
see any other form of development methodology that makes ANY sense at all. Yes,
four years or so is a long time to be in "requirements gathering". No question
about that. But in the same way, in a corporation, you wouldn't want the
marketing people to start coding the application, it would be crazy to start
pulling people off of coding R1 to play on R2. There is no R2 branch for just
that reason - R1 is the sole focus of the development team.
I am more than open to constructive criticism. If someone has a *better* idea,
they can always feel free to bring it up to me either here or privately. But
the ideas of splitting our forces to start with R2 while R1 is in progress, of
people other than the dev team making unilateral decisions on what will be in
R2 or even forcing the decisions of what is good and what is not before we need
to don't make any sense at all, to me. The first would delay both and waste a
lot of time. The second would never be implemented. The third would result in
poor decision being made.
I hope that this makes more sense to everyone now.
> The constant referring to GE of new ideas or concepts that more often than
> not don't seem to be overly complex/complicated isn't only nauseating but
> counter-productive. But given the previous responses to a request in change
> in attitude by those in command, I politely suggest the closing of the GE
> list. It just doesn't serve a purpose in its current incarnation,
> irrespective of the noble idea behind it when it was launched - unless people
> like expending their time on just talking about stuff without leading/going
> anywhere.
>
> Helmar
Other related posts:
