.5 and core plugins
Michael Bishop
before the .5 release.
I think the metaweblog plugin might be something we want in core, and
given some love. Likewise, I personally question if the flickr silo
and Viddler silo are really core plugin material.
Perhaps also if the jWYSIWYG plugin is up to snuff, it gets moved to
core as well.
As I think the consensus is that .5 will be more of a beta release, I
think we really need to bundle plugins that most bloggers have come to
expect in functionality to be available to them out of the box.
To that end, I hope the pingback plugin gets some love before .5 is
released, I (perhaps haven't opened a ticket yet) have had some issue
posting with multiple pings and getting an error lately.
~miklb
Michael Heilemann
rick c
would also add preapproved for sure (no one wants to have to approve
every comment), and possibly pingerupdate (pinging services is
important) and blogroll (people expect one).
On Mar 7, 11:28 am, Michael Heilemann <heilem...@gmail.com> wrote:
> Might I suggest the textile (and perhaps markdown as well)?
>
Michael Bishop
On Mar 7, 2:13 pm, rick c <rickcock...@gmail.com> wrote:
>... and possibly pingerupdate (pinging services is
plugin, with an option? I'd hate to see the core plugin list get too
long.
~miklb
Michael C. Harris
>
> A bit of a proposal that we reevaluate the plugins in core and extras
> before the .5 release.
>
> I think the metaweblog plugin might be something we want in core, and
> given some love.
I was talking to Rick Cockrum about this yesterday. I hope the
metaweblog API becomes irrelevant and AtomPub replaces it. I recognise
the client landscape isn't sufficient at the moment, but I've heard
rumours about new clients, say, for the iPhone, in development. My
concern (though it's small and I will not raise my voice again if
people support the inclusion of the metaweblog plugin) is that it's
backward looking, which doesn't fit Habari's desired image.
> Likewise, I personally question if the flickr silo and Viddler silo
> are really core plugin material.
I also had concerns about this but someone (ringmaster?) convinced me
that it was reasonable to provide media silos for an image service and
a video service.
> Perhaps also if the jWYSIWYG plugin is up to snuff, it gets moved to
> core as well.
I think a wysiwyg editor of some kind would be an important core
plugin. The jWYSIWYG plugin is very lightweight, given that we already
bundle jquery. As for the plugin, I think the only thing lacking is to
be resizable. jWYSIWYG itself may need some work but joksnet (the dev)
seems to be on board. Once a jWYSIWYG tag has been made and an
external set up I would strongly support its inclusion in core.
> As I think the consensus is that .5 will be more of a beta release, I
> think we really need to bundle plugins that most bloggers have come to
> expect in functionality to be available to them out of the box.
>
> To that end, I hope the pingback plugin gets some love before .5 is
> released, I (perhaps haven't opened a ticket yet) have had some issue
> posting with multiple pings and getting an error lately.
I agree it's a very important plugin, and it's been a bit flaky.
--
Michael C. Harris, School of CS&IT, RMIT University
http://twofishcreative.com/michael/blog
Scott Merrill
> > given some love.
>
>
> I was talking to Rick Cockrum about this yesterday. I hope the
> metaweblog API becomes irrelevant and AtomPub replaces it. I recognise
> the client landscape isn't sufficient at the moment, but I've heard
> rumours about new clients, say, for the iPhone, in development. My
> concern (though it's small and I will not raise my voice again if
> people support the inclusion of the metaweblog plugin) is that it's
> backward looking, which doesn't fit Habari's desired image.
This is a good point, but the other aspect is that Habari favors
documented standards over impromptu developments from single entities.
If there is a metaWebLog design / implementation spec, that would
make it suitable for inclusion with Habari until APP (or our
implementation) matures.
> > Likewise, I personally question if the flickr silo and Viddler silo
> > are really core plugin material.
>
>
> I also had concerns about this but someone (ringmaster?) convinced me
> that it was reasonable to provide media silos for an image service and
> a video service.
I think Flickr is the de facto image hosting service these days. They
have a well defined API. I think it makes sense to bundle this. Its
presence should not preclude core bundling of other image silos. I'd
love to see a Gallery silo.
Viddler is on the cusp of user acceptance, I think (It's not as
popular as YouTube), so a reasonable argument could be made that it
doesn't belong. Echoing Michael's comments above, though, I think
it's sufficiently forward-looking as to make it a reasonable candidate
for inclusion.
> > To that end, I hope the pingback plugin gets some love before .5 is
> > released, I (perhaps haven't opened a ticket yet) have had some issue
> > posting with multiple pings and getting an error lately.
>
>
> I agree it's a very important plugin, and it's been a bit flaky.
Ditto.
Cheers,
Scott
Michael Bishop
On Mar 8, 6:56 am, "Michael C. Harris" <michael.twof...@gmail.com>
wrote:
>
> > A bit of a proposal that we reevaluate the plugins in core and extras
> > before the .5 release.
>
> > I think the metaweblog plugin might be something we want in core, and
> > given some love.
>
> I was talking to Rick Cockrum about this yesterday. I hope the
> metaweblog API becomes irrelevant and AtomPub replaces it. I recognise
> the client landscape isn't sufficient at the moment, but I've heard
> rumours about new clients, say, for the iPhone, in development. My
> concern (though it's small and I will not raise my voice again if
> people support the inclusion of the metaweblog plugin) is that it's
> backward looking, which doesn't fit Habari's desired image.
>
honestly wouldn't see this happening by a .5 release, and despite
backward looking, users rely on clients like Ecto, MarsEdit, and
Scribefire. Supporting those until Atom Pub filters down to the user
level would seem logical.
> > Likewise, I personally question if the flickr silo and Viddler silo
> > are really core plugin material.
>
> I also had concerns about this but someone (ringmaster?) convinced me
> that it was reasonable to provide media silos for an image service and
> a video service.
personally do not like the idea of Habari coming across as "pushing"
specific, commercial services in an open source app. I would be
concerned about integrity and ulterior motive issues being raised by a
larger community. If there were silos for service agnostic image/
video integration, I'd be all for inclusion.
~miklb
Randy Walker
--
Randy Walker
via iPhone
On Mar 8, 2008, at 9:21 AM, Michael Bishop <miklb....@gmail.com>
wrote:
Michael C. Harris
>
> On Mar 8, 6:56 am, "Michael C. Harris" <michael.twof...@gmail.com> wrote:
> > On Fri, Mar 07, 2008 at 05:55:36AM -0800, Michael Bishop wrote:
> >
> > > A bit of a proposal that we reevaluate the plugins in core and extras
> > > before the .5 release.
> >
> > > I think the metaweblog plugin might be something we want in core, and
> > > given some love.
> >
> > I was talking to Rick Cockrum about this yesterday. I hope the
> > metaweblog API becomes irrelevant and AtomPub replaces it. I recognise
> > the client landscape isn't sufficient at the moment, but I've heard
> > rumours about new clients, say, for the iPhone, in development. My
> > concern (though it's small and I will not raise my voice again if
> > people support the inclusion of the metaweblog plugin) is that it's
> > backward looking, which doesn't fit Habari's desired image.
>
> Completely valid points, and certainly worthy of longer term goals. I
> honestly wouldn't see this happening by a .5 release, and despite
> backward looking, users rely on clients like Ecto, MarsEdit, and
> Scribefire. Supporting those until Atom Pub filters down to the user
> level would seem logical.
I said I wasn't going to argue, but hey, I am :) I agree with both
your points - there won't be more a huge number of new AtomPub clients
before 0.5, and we should support existing clients. I argue that we
are supporting them by providing a metaweblog plugin, but that plugin
doesn't need to be in core.
> > > Likewise, I personally question if the flickr silo and Viddler silo
> > > are really core plugin material.
> >
> > I also had concerns about this but someone (ringmaster?) convinced me
> > that it was reasonable to provide media silos for an image service and
> > a video service.
>
> My only concern with this is that they are service specific. I
> personally do not like the idea of Habari coming across as "pushing"
> specific, commercial services in an open source app.
I also was concerned about the favouring of specific services,
commercial or not.
> I would be concerned about integrity and ulterior motive issues
> being raised by a larger community. If there were silos for service
> agnostic image/ video integration, I'd be all for inclusion.
I don't think such a thing is possible. So, we're left with the choice
between including only the Simple File silo or providing some
exemplars for video and image. I believe it's such a good feature that
we'd be crazy not to include the silos. As a compromise, we can also
include as many alternative silos in -extras as possible. YouTube is
already there (and I'm still waiting on the feedback people have
promised on that) and I'll commit to writing a Picasa one before 0.5.
Owen Winkler
>
> I said I wasn't going to argue, but hey, I am :) I agree with both
> your points - there won't be more a huge number of new AtomPub clients
> before 0.5, and we should support existing clients. I argue that we
> are supporting them by providing a metaweblog plugin, but that plugin
> doesn't need to be in core.
This (and other concerns of this nature) will be less of an issue if
there is an easy way to install plugins from a central location (our
repository). If this is a concern to you (2nd-person plural), then you
should consider helping get hp.o in shape for accomplishing that.
Owen
Michael C. Harris
>
> Michael C. Harris wrote:
> >
> > I said I wasn't going to argue, but hey, I am :) I agree with both
> > your points - there won't be more a huge number of new AtomPub clients
> > before 0.5, and we should support existing clients. I argue that we
> > are supporting them by providing a metaweblog plugin, but that plugin
> > doesn't need to be in core.
>
> This (and other concerns of this nature) will be less of an issue if
> there is an easy way to install plugins from a central location (our
> repository).
Absolutely. Is this just an aside, or are you expressing an opinion
one way or another about whether the metaweblog plugin should be in
core?
> If this is a concern to you (2nd-person plural), then you should
> consider helping get hp.o in shape for accomplishing that.
I'm more than happy to help, but have no idea where to start on this.
(I note the 2nd-person plural, I'm just expressing my willingness to
help.)
rick c
sense as long as the configuration panel allows you to turn either on
or off and change the list of services to update.
I've had second thoughts about the metaWeblog plugin, though. It's
important in that it lets people blog from their favorite editor.
Whatever it's birth, it's the only widely supported remote blogging
api there is at the moment. But it is limited in what it can do
without extending it so much you have another engine-specific api. As
such, it would be better off in -extras.
Michael Bishop
in core? Or once this is in place, would Habari not ship with any
plugins?
~miklb