Re: [learningreg-dev] Fwd: Augmenting others' metadata

15 views
Skip to first unread message

Jim Klo

unread,
Jun 14, 2013, 4:54:02 PM6/14/13
to <learning-registry-collaborate@googlegroups.com>, Jerome Grimmer, Jeanne Kitchens, learnin...@googlegroups.com, Michael Parsons
Sorry for the multiple repost..  the collaborate list was bouncing… 


On Jun 14, 2013, at 1:49 PM, Jim Klo <jim...@sri.com>
 wrote:

All,

Jerome brings up an interesting discussion… adding the collaborate list as I think there are several that might have some guidance.

+LR Collaborate List

Begin forwarded message:

From: Jerome Grimmer <jgri...@siuccwd.com>
Subject: Augmenting others' metadata
Date: June 14, 2013 1:41:45 PM PDT
Cc: Jeanne Kitchens <jkit...@siuccwd.com>, Michael Parsons <mpar...@siuccwd.com>, Jim Klo <jim...@sri.com>

Hey guys,
I have a question about a very real situation we have here in Illinois.  We have lots of resources that we’ve harvested from the Learning Registry, which of course are published by people other than us.  We’re going to grant some users of our system the ability to augment others’ metadata with their own. 
 
Here’s the use case: Suppose there’s a resource which had metadata published to the LR (by someone other than an Illinois Shared Learning Environment user) indicating it was suitable for 5th grade.  A teacher comes along and uses this resource for a gifted student in her 3rd grade class.  She comes into our system and tags the resource as suitable for 3rdgrade (LRMI educationalAlignment field), and also adds a tag in the LRMI educationUse field, indicating that it’s useful as an “Enhancement.”
 
Here’s the problem: The recipe we have for the paradata verb “tagged” (see the “tagged” recipe in http://goo.gl/Cq6im, about ¾ of the way through the document) the way we see it is basically treating all tags in a fashion similar to keywords.  There doesn’t seem to be anything to indicate that “3rd Grade” is a grade level and “Enhancement” is an Educational Use.  Which way would it be better (for the greater good of the community) to publish this information?
·         Publish a new metadata record by ISLE that shows all the metadata we previously had, plus the additional “3rd grade” educationalAlignment and the “Enhancement” educational use (and flip the submitter field in our database so it looks like it came from us), or
·         Publish a paradata statement using the “tagged” paradata recipe, realizing that it is up to the consumer of the data to figure out that “3rd grade” is an educationalAlignment and Enhancement is an Ed. Use.
 
Thanks,
Jerome Grimmer
Southern Illinois University Carbondale
2450 Foundation Drive Suite 100
Springfield, IL
NOTE: My E-mail address has changed
"If you think you're too small to make a difference, you've never spent a night with a mosquito." - An African Proverb
 


--
You received this message because you are subscribed to the Google Groups "Learning Registry Developers List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to learningreg-d...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

Jerome Grimmer

unread,
Jun 14, 2013, 5:42:32 PM6/14/13
to learnin...@googlegroups.com, learning-regis...@googlegroups.com, Jeanne Kitchens, Michael Parsons, Jim Klo

Hi Joe,

Thanks for the link.  On the page you gave me, I saw a link to http://activitystrea.ms/specs/json/schema/activity-schema.html#verbs and http://activitystrea.ms/specs/json/schema/activity-schema.html#tags-property, which gave me an idea.  Would something like this work (I’ve not included the entire envelope, just the stuff I thought was essential to convey the needed information)?  I still wonder if it would be better to publish it as metadata though.

 

"resource_data": {

  "activity": {

    "actor": "Teacher",

    "verb": {

      "action": "tagged"

      .

      .

      .

    },

    "object": {

      "id": "http://www.example.com/resource/12345"

    },

    "tags": [

      {

        "objectType": "gradeLevel",

        "description": ["Grade 3"]

      },

      {

        "objectType": "educationalUse",

        "description": ["Enhancement"]

      }

    ]

  }

},

"resource_data_type": "paradata",

"payload_schema": ["LR Paradata 1.0"]

 

Jerome Grimmer

Southern Illinois University Carbondale

2450 Foundation Drive Suite 100

Springfield, IL

Phone: 217-786-3010 ext. 5857

Toll-free: 1-800-252-4822 ext. 5857

NOTE: My E-mail address has changed

jgri...@siuccwd.com

"If you think you're too small to make a difference, you've never spent a night with a mosquito." - An African Proverb

 

From: learnin...@googlegroups.com [mailto:learnin...@googlegroups.com] On Behalf Of joe hobson
Sent: Friday, June 14, 2013 4:13 PM
To: learnin...@googlegroups.com
Cc: Jeanne Kitchens; Michael Parsons; Jim Klo
Subject: Re: [learningreg-dev] Augmenting others' metadata

 

I'm sure there will be some metadata purists out there that will speak up about the potential issues of people publishing their own metadata about other people's resources. I don't see the harm in that approach, but I can see how the paradata route would be easier for other systems to consume and will likely satisfy the purists. As one of the people that piloted the paradata schema, i can't see why you have to stick to just the examples shown in the doc. Its origins are in activity stream so I'd say look at the verbs in that spec for examples of how you can report how other users are describing a resource: http://activitystrea.ms/registry/verbs/

 

 

On Jun 14, 2013, at 1:41 PM, Jerome Grimmer wrote:



Hey guys,

I have a question about a very real situation we have here in Illinois.  We have lots of resources that we’ve harvested from the Learning Registry, which of course are published by people other than us.  We’re going to grant some users of our system the ability to augment others’ metadata with their own. 

 

Here’s the use case: Suppose there’s a resource which had metadata published to the LR (by someone other than an Illinois Shared Learning Environment user) indicating it was suitable for 5th grade.  A teacher comes along and uses this resource for a gifted student in her 3rd grade class.  She comes into our system and tags the resource as suitable for 3rd grade (LRMI educationalAlignment field), and also adds a tag in the LRMI educationUse field, indicating that it’s useful as an “Enhancement.”

 

Here’s the problem: The recipe we have for the paradata verb “tagged” (see the “tagged” recipe in http://goo.gl/Cq6im, about ¾ of the way through the document) the way we see it is basically treating all tags in a fashion similar to keywords.  There doesn’t seem to be anything to indicate that “3rd Grade” is a grade level and “Enhancement” is an Educational Use.  Which way would it be better (for the greater good of the community) to publish this information?

·         Publish a new metadata record by ISLE that shows all the metadata we previously had, plus the additional “3rd grade” educationalAlignment and the “Enhancement” educational use (and flip the submitter field in our database so it looks like it came from us), or

·         Publish a paradata statement using the “tagged” paradata recipe, realizing that it is up to the consumer of the data to figure out that “3rd grade” is an educationalAlignment and Enhancement is an Ed. Use.

 

Thanks,

Jerome Grimmer

Southern Illinois University Carbondale

2450 Foundation Drive Suite 100

Springfield, IL

NOTE: My E-mail address has changed

"If you think you're too small to make a difference, you've never spent a night with a mosquito." - An African Proverb

 

 

-- 
You received this message because you are subscribed to the Google Groups "Learning Registry Developers List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to learningreg-d...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

Jim Klo

unread,
Jun 14, 2013, 5:48:58 PM6/14/13
to <learning-registry-collaborate@googlegroups.com>, Jerome Grimmer, Jeanne Kitchens, learnin...@googlegroups.com, Michael Parsons
Now that I've got the right message thread working…

Jerome,

I've been thinking about this for some time, as this was one of the problems with the current LR Paradata 1.0 spec in that it lacked the rigor in specificity for objects.  One thought I've been tossing around with several folks is to leverage the Schema.org vocabulary (which now includes most of LRMI) to make objects more specific.

There is some new work with Schema.org around ActivityActions: http://www.w3.org/wiki/WebSchemas/ActivityActions. The new Google GMail apparently makes use of this new vocabulary. With this, the WebSchemas community seems to be circling around JSON-LD (JSON Linked Data) expresssions. 

At this point in time, since none of the previous has been discussed really here - and more to your question - I think paradata is the correct solution and by the way I read the "spec" you could theoretically twist it to do something like this:

{
"verb": {
"action": "tagged",
"measure": {
"@context": "schema.org",
"targetName": "3rd grade",
"educationalFramework": "grade level"
},
"context": "educationalAlignment"
},

}


Which is legal in that measure does not specify a schema.  I know this seems odd - but  in this example, I'm nested JSON-LD instance of an AlignmentObject… theoretically you could put anything in as a measure.

What do you all think?  Am I way off my rocker and gone off the deep end?


Jim Klo
Senior Software Engineer
Center for Software Engineering
SRI International
t. @nsomnac

On Jun 14, 2013, at 1:54 PM, Jim Klo <jim...@sri.com>
 wrote:

--
--
You received this message because you are subscribed to the Google
Groups "Learning Registry: Collaborate" group.
 
To post: learning-regis...@googlegroups.com
To unsubscribe:learning-registry-collabo...@googlegroups.com
 
For more options, visit this group at
http://groups.google.com/group/learning-registry-collaborate?hl=en?hl=en
 
---
You received this message because you are subscribed to the Google Groups "Learning Registry: Collaborate" group.
To unsubscribe from this group and stop receiving emails from it, send an email to learning-registry-co...@googlegroups.com.

Jim Klo

unread,
Jun 14, 2013, 5:55:02 PM6/14/13
to Jerome Grimmer, learnin...@googlegroups.com, learning-regis...@googlegroups.com, Jeanne Kitchens, Michael Parsons
That seems reasonable…  To be inline with the spec… make one minor change, tags -> related:

"resource_data": {
  "activity": {
    "actor": "Teacher",
    "verb": {
      "action": "tagged"
      .
      .
      .
    },
    "object": {
    },
    "related": [
      {
        "objectType": "gradeLevel",
        "description": ["Grade 3"]
      },
      {
        "objectType": "educationalUse",
        "description": ["Enhancement"]
      }
    ]
  }
},
"resource_data_type": "paradata",
"payload_schema": ["LR Paradata 1.0"]
Jim Klo
Senior Software Engineer
Center for Software Engineering
SRI International

On Jun 14, 2013, at 2:42 PM, Jerome Grimmer <jgri...@siuccwd.com>
 wrote:

Hi Joe,
Thanks for the link.  On the page you gave me, I saw a link to http://activitystrea.ms/specs/json/schema/activity-schema.html#verbs andhttp://activitystrea.ms/specs/json/schema/activity-schema.html#tags-property, which gave me an idea.  Would something like this work (I’ve not included the entire envelope, just the stuff I thought was essential to convey the needed information)?  I still wonder if it would be better to publish it as metadata though.

Jerome Grimmer

unread,
Jun 14, 2013, 5:56:00 PM6/14/13
to learning-regis...@googlegroups.com, Jeanne Kitchens, learnin...@googlegroups.com, Michael Parsons

Wow I can see several ways to do this.  We should hash this out and come up with a “standard” way of doing it for interoperability purposes so that InBloom, OER Commons, ISLE, and others can all understand each other’s additional tags once someone else’s metadata has been published.

 

Jerome Grimmer

Southern Illinois University Carbondale

2450 Foundation Drive Suite 100

Springfield, IL

Phone: 217-786-3010 ext. 5857

Toll-free: 1-800-252-4822 ext. 5857

NOTE: My E-mail address has changed

jgri...@siuccwd.com

"If you think you're too small to make a difference, you've never spent a night with a mosquito." - An African Proverb

 

Jerome Grimmer

unread,
Jun 14, 2013, 5:57:38 PM6/14/13
to learning-regis...@googlegroups.com, learnin...@googlegroups.com, Jeanne Kitchens, Michael Parsons

Jim, to be honest I pulled “tags” right out of the activitystrea.ms document at http://activitystrea.ms/specs/json/schema/activity-schema.html#tags-property.  Like I said in a previous email, we should come up with a standard way of doing this so everybody understands everyone else’s “tags”.

 

Jerome Grimmer

Southern Illinois University Carbondale

2450 Foundation Drive Suite 100

Springfield, IL

Phone: 217-786-3010 ext. 5857

Toll-free: 1-800-252-4822 ext. 5857

NOTE: My E-mail address has changed

jgri...@siuccwd.com

"If you think you're too small to make a difference, you've never spent a night with a mosquito." - An African Proverb

 

--

--
You received this message because you are subscribed to the Google

Groups "Learning Registry: Collaborate" group.
 
To post: learning-regis...@googlegroups.com
To unsubscribe:learning-registry-co...@googlegroups.com
 
For more options, visit this group at
http://groups.google.com/group/learning-registry-collaborate?hl=en?hl=en
 
---

You received this message because you are subscribed to the Google Groups "Learning Registry: Collaborate" group.
To unsubscribe from this group and stop receiving emails from it, send an email to learning-registry-co...@googlegroups.com.

Steve Midgley

unread,
Jun 14, 2013, 6:31:30 PM6/14/13
to <learning-registry-collaborate@googlegroups.com>, learnin...@googlegroups.com, Jeanne Kitchens, Michael Parsons
I'm not a purist (I don't think) but I'm wondering why you don't just emit LRMI JSON with the resource URL and whatever tags you've generated that are new? (If you can't tell the new tags from the old ones, it seems fine, per Joe's point, to just emit everything). Because you're using LRMI expressions, why not just emit LRMI JSON? Is it b/c you're also trying to emit something related to a paradata verb? (Or associate the LRMI with a paradata subject?)

Steve

Jim G

unread,
Jun 17, 2013, 7:50:32 AM6/17/13
to learning-regis...@googlegroups.com, learnin...@googlegroups.com
CEDS has established specific vocabularies for U.S. education that may help solve this kind of problem. 

Schema.org and LRMI are designed to be international standards for tagging and therefore avoid country-specific vocabularies such as grade-level.  LRMI  provides the alignment object and assumes the implementer will reference the appropriate vocabulary.  LRMI group discussions have recognized this need for the use of standard vocabularies external to the LRMI/Schema.org spec.  In the U.S. the Common Education Data Standards provide the vocabulary fixed vocabularies (with fixed URIs) for things like "grade level,"  giving an option set and definition of the term.  Referencing the CEDS URI/term would make it clear that “3rd Grade” is a grade level.

Let me suggest that the LR/LRMI/CEDS communities to work together on an implementation approach and guidance for use of the standards vocabularies in Learning Registry assertions and LRMI tagging.


On Friday, June 14, 2013 4:54:02 PM UTC-4, Jim Klo wrote:
Sorry for the multiple repost..  the collaborate list was bouncing… 

On Jun 14, 2013, at 1:49 PM, Jim Klo <jim...@sri.com>
 wrote:

All,

Jerome brings up an interesting discussion… adding the collaborate list as I think there are several that might have some guidance.

+LR Collaborate List

Begin forwarded message:

From: Jerome Grimmer <jgri...@siuccwd.com>
Subject: Augmenting others' metadata
Date: June 14, 2013 1:41:45 PM PDT

Hey guys,
I have a question about a very real situation we have here in Illinois.  We have lots of resources that we’ve harvested from the Learning Registry, which of course are published by people other than us.  We’re going to grant some users of our system the ability to augment others’ metadata with their own. 
 
Here’s the use case: Suppose there’s a resource which had metadata published to the LR (by someone other than an Illinois Shared Learning Environment user) indicating it was suitable for 5th grade.  A teacher comes along and uses this resource for a gifted student in her 3rd grade class.  She comes into our system and tags the resource as suitable for 3rdgrade (LRMI educationalAlignment field), and also adds a tag in the LRMI educationUse field, indicating that it’s useful as an “Enhancement.”
 
Here’s the problem: The recipe we have for the paradata verb “tagged” (see the “tagged” recipe in http://goo.gl/Cq6im, about ¾ of the way through the document) the way we see it is basically treating all tags in a fashion similar to keywords.  There doesn’t seem to be anything to indicate that “3rd Grade” is a grade level and “Enhancement” is an Educational Use.  Which way would it be better (for the greater good of the community) to publish this information?
·         Publish a new metadata record by ISLE that shows all the metadata we previously had, plus the additional “3rd grade” educationalAlignment and the “Enhancement” educational use (and flip the submitter field in our database so it looks like it came from us), or
·         Publish a paradata statement using the “tagged” paradata recipe, realizing that it is up to the consumer of the data to figure out that “3rd grade” is an educationalAlignment and Enhancement is an Ed. Use.
 
Thanks,
Jerome Grimmer
Southern Illinois University Carbondale
 

Steve Midgley

unread,
Jun 17, 2013, 8:11:26 PM6/17/13
to <learning-registry-collaborate@googlegroups.com>, learnin...@googlegroups.com
That would be great Jim. So far in my experience, no one near to the LR project cares at all what vocabulary is used - they all just want to have one they can implement. This means having a set of fields and values, and *also* a practical binding to a format so they can actually send data on the wire to each other.

Jim, Walt, Jason and others have done a lot of work binding LRMI/schema to JSON so that it can fit inside LR (the "HTML" binding isn't so great for transport). 

To be concrete, is your suggestion that we work with CEDS to identify specific vocabularies so that we can use the alignment property to tag them?

This would be basically agreeing on what to fill out for this object: http://schema.org/AlignmentObject

Does that sound right?

Steve





--
--
You received this message because you are subscribed to the Google
Groups "Learning Registry: Collaborate" group.
 
To post: learning-regis...@googlegroups.com
To unsubscribe:learning-registry-co...@googlegroups.com
 
For more options, visit this group at
http://groups.google.com/group/learning-registry-collaborate?hl=en?hl=en
 
---
You received this message because you are subscribed to the Google Groups "Learning Registry: Collaborate" group.
To unsubscribe from this group and stop receiving emails from it, send an email to learning-registry-co...@googlegroups.com.

Steve Midgley

unread,
Jun 19, 2013, 3:22:49 PM6/19/13
to learnin...@googlegroups.com, <learning-registry-collaborate@googlegroups.com>
Hi Dave,

My thinking is that you could probably express alignment with "pure" schema.org/lrmi, and that you don't need paradata format at all. Jim and Walt have some examples of what pure schema.org looks like in JSON, but without the binding, you're basically just emitting these fields:

Name: [name]
Description: [descr]
keywords: (if you have any, they're great if you do)
datePublished: 2013-6-18
educationalAlignment: {
  alignmentType: teaches (or whatever appropriate)
  educationalFramework: Common Core State Standards
  targetDescription:Determine the meaning of symbols, key terms, and other domain-specific words and phrases as they are used in a specific scientific or technical context relevant to grades 6—8 texts and topics. (optional)
  targetName:CCSS.ELA-Literacy.RST.6-8.4
}

What do you think? We can work up a JSON binding if this seems reasonable. But since you're just emitting a "fact" not an activity or an opinion of a user, it seems like using schema.org is better than paradata? I got all my field names and structure from these pages:

Steve



On Tue, Jun 18, 2013 at 12:50 PM, <finke...@gmail.com> wrote:
Steve,
     NSDL also needs to do this for our comm_anno records,. You mentioned filling out http://schema.org/AlignmentObject as the context, which is how I was planning on doing it. So I'm glad you mentioned it.

so my plan would be our activity in LR Paradata format would be something like

"activity": {
       "verb": {
            "action":"aligned",
            "date":"2013-6-18",
            "context":{
                {'type':'http://schema.org/AlignmentObject',
                 'properties': {'targetUrl':'http://asn.jesandco.org/resources
                                /S1143AC5'}
                }
           
}
         }

I was hoping to we could use a different tag called aligned instead of tagged. Even though tagged would work to. Is it considered bad practice to go against the current list of verbs?

Is this what you were thinking?
thanks
Dave - NSDL



On Monday, June 17, 2013 6:11:26 PM UTC-6, Steve Midgley wrote:
That would be great Jim. So far in my experience, no one near to the LR project cares at all what vocabulary is used - they all just want to have one they can implement. This means having a set of fields and values, and *also* a practical binding to a format so they can actually send data on the wire to each other.

Jim, Walt, Jason and others have done a lot of work binding LRMI/schema to JSON so that it can fit inside LR (the "HTML" binding isn't so great for transport). 

To be concrete, is your suggestion that we work with CEDS to identify specific vocabularies so that we can use the alignment property to tag them?

This would be basically agreeing on what to fill out for this object: http://schema.org/AlignmentObject

Does that sound right?

Steve



Reply all
Reply to author
Forward
0 new messages