[sugar] Squeak / Etoys RPMs

[Christopher Blizzard]

Ian Bicking wrote:
> Perhaps this is a function of our different backgrounds.  To me issues 
> of content management and basic persistence seem pretty concrete, 
> because that's a lot of what I work with.  It's also my belief that the 
> content will ultimately be a greater investment than the software, and 
> that the content has to outlive any particular software associated with 
> it.  I'm not an obsessive archivalist, and I don't think that's 
> necessary for this project (especially for content just on the laptop 
> that isn't hosted somewhere else).  But I do think we need to consider 
> content quite seriously as its own entity, because we're managing it on 
> behalf of the child, it's not ours.
> 
> And I feel really quite strongly that content belongs to the user, not 
> the activity.  That means that when possible the content should have a 
> purpose and meaning outside of the activity that created it.  Even for 
> the most obscure or niche activity this applies, because someone might 
> fork that activity and still understand its obscure internal format. 
> This is not based on goals of "it should work in 30 seconds" but rather 
> "it should work in two years".
> 
> So I feel protective of content, and you feel protective of user 
> experience, and that means we'll probably be coming at this from 
> different perspectives, and I don't know if either one will entirely 
> drive the other.
> 
> 

I really don't think that these two viewpoints are at odds.  I happen to 
strongly agree with your point that contant belongs to the user.  But I 
also feel that it should also have context.  Content without context 
just turns you into a file server. :)

It's becomes a question of how do we make it easy to do something useful 
with a piece of content.  We want to make sure it's really obvious what 
to do with anything on the laptop, while at the same time not preventing 
it from being used in other contexts.  The current metaphors we have 
today do a really shitty job of connecting content with programs that 
can do something useful.

We have the journal, we (will) have a decent data store - let's take 
advantage of that to build something that really makes things easier.

And just to be clear, we're not arguing against choice in how something 
is used.  Just that it's really easy to figure how the which-to-use 
question up front.

--Chris