[sakai2-tcc] We need a blog tool for Sakai 2.9

[Noah Botimer]

I personally don't want to anticipate any significant effort redistribution or dispersion. I just don't think that packaging has any bearing on it. Any freed effort will be spent as it will be spent by those who command it.

(I also don't read David's comment of focus as anticipating redistribution, but simply as to a smaller release package, maybe with subtly better QA, though I wouldn't personally assume that.)

I would much rather focus on the friction exerted by ugly packaging and the core/contrib boundary. As you say, Seth, there is an empirical question of whether our current "indie" packaging provides us with substantial leverage on the problems we face. It is multi-faceted, in terms of the ability to revise components independently, strict version dependency issues, and inclusion within the core release or local installation.

Though the process of moving some projects to "indie" has given us some wisdom, I don't personally think it addresses our range of difficulties yet. This is not a disparaging comment on the hard work done. I think that the work needs some additional help and recognize that this implies incremental resource. I hope we can find that for 2.9.

I'll leave aside distinctions of indispensability. I am only interested in a mechanism that makes such a decision practicable for the community and individual institutions. The gold star of "in the release" has, unfortunately, been the sole determinator of whether something is readily adopted. To me, that's what the release has ultimately meant: these things will be able to be run by the majority -- it has implied rather little about the final quality, completeness, or maintenance of the components, inside or outside of the release.

Thanks,
-Noah

On Mar 3, 2011, at 10:53 AM, Seth Theriault wrote:

> David Haines wrote:
> 
>> We shouldn't underestimate the potential positive impact of an
>> "app store" on the release process.  App store installation would
>> make indy tool releases / distribution / development much
>> quicker.  With more installable tools the Sakai releases could
>> focus much more on the core indispensable tools and kernel and so
>> enable community development.  Current releases are held captive
>> to the tool code that is slowest to be compatible with the latest
>> release.
> 
> Playing devil's advocate for a moment, what happens when everyone
> is working on "cool app store tools" and not on the core
> foundational things that they rely on? Why should we assume that
> any "app store" is going to encourage community involvement since
> it allows people to wall themselves off? We have any number of
> "core indispensable tools" that a few in the community are
> maintaining and few if any are actively testing.
> 
> Part of this discussion probably needs to include a review of the
> efficiency of "indie" releases based on actual production
> experience with large numbers of them, something that is fairly
> new in this community.
> 
> Seth
> 
>