status?

Wa away last week, so just wanted to check where we’re at. Simon, can
you perhaps summarize what’s still to be done, and perhaps suggest a
roadmap of how we get there? I’m not seeing any activity on the branch
schema :wink:

Bruce

We have to write the schema, which I plan to work on this weekend. I
have everything implemented in Zotero SVN.

Simon

Okay, schema committed, along with conforming sample APA style. Let
me know what you think.

Simon

I’ve recently been having problems updating from the SVN. Do you?

Beyond that, does the APA style validate against the schema? :slight_smile:

Also, you mentioned before writing a script to convert old styles to
new. A good test would be if you can do that, and if the resulting
styles validate.

Finally, you have APA as an “in-text” style; should be “author-date”.

Bruce

Okay, schema committed, along with conforming sample APA style. Let
me know what you think.

I’ve recently been having problems updating from the SVN. Do you?

Beyond that, does the APA style validate against the schema? :slight_smile:

Yes, it does. It’s slightly (but not significantly) different from
what’s in Zotero right now, because I made a few changes to attribute
names and option positions.

Also, you mentioned before writing a script to convert old styles to
new. A good test would be if you can do that, and if the resulting
styles validate.

At some point, this will get done. I have other Zotero obligations to
take care of, so I can’t make any promises as to when.

Finally, you have APA as an “in-text” style; should be “author-date”.

I changed this, because I’d rather just have “in-text” and “note”
style types. If you have better names for these, feel free to
substitute, but I fail to see why the fact that it’s an “author-date”
style makes any difference for anything besides sorting.

Simon

Also, you mentioned before writing a script to convert old styles to
new. A good test would be if you can do that, and if the resulting
styles validate.

At some point, this will get done. I have other Zotero obligations to
take care of, so I can’t make any promises as to when.

So how do we establish CSL 1.0 and say we’re done?

Perhaps someone (Jim?) ought to take a crack at writing new styles in
the updated version in the meantime? The IEEE style seems like a good
candidate. I’ll work on finishing the Bluebook styles I started, but I
can’t commit to that soon.

Finally, you have APA as an “in-text” style; should be “author-date”.

I changed this, because I’d rather just have “in-text” and “note”
style types. If you have better names for these, feel free to
substitute, but I fail to see why the fact that it’s an “author-date”
style makes any difference for anything besides sorting.

Well, if it makes no difference anymore for the schema validation or
processing, that’s fine. However, it’s crucial metadata.

I don’t generally care about any style other than author-date.
Likewise, many users won’t care about anything but note styles.

As I’ve said repeatedly, CSL needs to be designed to scale to thousands
of styles (preferably in distributed repositories), and metadata like
this helps allow that.

Beyond helping finding styles, it can help in editing them.

Bruce

Also, you mentioned before writing a script to convert old styles to
new. A good test would be if you can do that, and if the resulting
styles validate.

At some point, this will get done. I have other Zotero obligations to
take care of, so I can’t make any promises as to when.

So how do we establish CSL 1.0 and say we’re done?

Perhaps someone (Jim?) ought to take a crack at writing new styles in
the updated version in the meantime? The IEEE style seems like a good
candidate. I’ll work on finishing the Bluebook styles I started, but I
can’t commit to that soon.

There are interns at Zotero HQ working on generating new styles with
the new schema.

Finally, you have APA as an “in-text” style; should be “author-
date”.

I changed this, because I’d rather just have “in-text” and “note”
style types. If you have better names for these, feel free to
substitute, but I fail to see why the fact that it’s an “author-date”
style makes any difference for anything besides sorting.

Well, if it makes no difference anymore for the schema validation or
processing, that’s fine. However, it’s crucial metadata.

I don’t generally care about any style other than author-date.
Likewise, many users won’t care about anything but note styles.

As I’ve said repeatedly, CSL needs to be designed to scale to
thousands
of styles (preferably in distributed repositories), and metadata like
this helps allow that.

Beyond helping finding styles, it can help in editing them.

It truly belongs in the section, then. Should we use
or create a new tag?

Simon

There are interns at Zotero HQ working on generating new styles with
the new schema.

OK. Make sure they understand that the goal in the styles is to adopt
DRY wherever possible, and that in general I’d suggest they avoid
conditioning formatting on types where possible. My hunch is the more
they use macro and conditionals and the less they’re tied to type, the
more compact and robust the styles will be.

Perhaps, too, they can start documenting best practices and suggestions
on the zotero dev wiki so that we can pull that together into some more
formal documents.

I’m fine, BTW, if you guys want to host styles; just make sure they
have good stable URI IDs, and that they’re resolvable directly so that
others can use them (like say my XSLT code if/when I update it).

It truly belongs in the section, then. Should we use
or create a new tag?

I was thinking reuse category. We could just give it a dedicated scheme
URI.

So APA would get maybe:


Need to think more on the details (whether to use the URI and what it
should be), but that seems sane.

Bruce

Dan – can I get some comment on this? If possible on your end, it
seems to me we ought to move to realizing the distributed repository
idea sooner rather than later. I really don’t like to hassle you guys
to load the styles in the xbib repository, and it shouldn’t really be
necessary in any case.

My hunch is using Atom is still the best idea. Just write a little
script that generates the feeds from a directory of CSL files, and it
should be pretty easy.

I realize adding the UI for selection is more work, but perhaps not
essential at first.

In short, I want with CSL 1.0 for us to include information about
repositories, and that CSL styles ought to be publicly available at the
link URI.

Bruce