LRMI 1.0 Examples Listing

[Greg Grossmeier]

Hello,

If you come across or create any new LRMI example marked up content,
please either reply to this email on thread or send me a message. I have
started a collection at:
http://www.lrmi.net/the-specification/examples

Best,

Greg

--
Greg Grossmeier
Education Technology & Policy Coordinator
twitter: @g_gerg / identi.ca: @greg / skype: greg.grossmeier

[Monty Swiryn]

Hi,

Question about the examples on the lrmi site...
In the page source, which parts of the code are the actual tags that will affect the search results?
I see that there are many lines in the code that include "itemprop" but don't have "meta" associated with them.
Does the pertinent markup include all the example code with "itemprop", or is it only where we see "meta" and "itemprop" together?

Many thanks.
Monty

[Greg Grossmeier]

<quote name="Monty Swiryn" date="2012-06-14" time="13:34:43 -0700">

> Hi,
>
> Question about the examples on the lrmi site...
> In the page source, which parts of the code are the actual tags that will
> affect the search results?

The parts that are marked up with Schema.org and LRMI metadata.

> I see that there are many lines in the code that include "itemprop" but
> don't have "meta" associated with them.
> Does the pertinent markup include all the example code with "itemprop", or
> is it only where we see "meta" and "itemprop" together?

The use of "meta" is not required and in some cases not advisable.

If you see "<meta... > " in HTML source that means that it is not
displayed by the browsers and is instead behind the scenes. You can
apply Schema.org and LRMI metadata using <meta ...> blocks but the best
way is to annotate already displayed content.

You can do this via extending already in place HTML, eg:

"<p>This work is produced primarily for use by <span
itemprop="typicalAgeRange">13-16</span> year olds.</p>"

Does that help?

[Monty Swiryn]

Hi Greg,

That helps a lot. Thanks.  

A follow-up...  In example #2 “Signals and Systems”, the source contains the following meta data.  So according to your answer, this is not needed, correct?  

Was this put in by a tagger tool?

thanks again,

Monty

  <!-- Begin Automatic Metadata Insertion -->

    <meta content="6-003-signals-and-systems-spring-2010" name="WT.cg_n">
    <meta content="Course Home" name="WT.cg_s">
    <meta content="Signals and Systems" name="Title">
    <meta content="6.003 covers the fundamentals of signal and system analysis, focusing on representations of discrete-time and continuous-time signals (singularity functions, complex exponentials and geometrics, Fourier representations, Laplace and Z transforms, sampling) and representations of linear, time-invariant systems (difference and differential equations, block diagrams, system functions, poles and zeros, convolution, impulse and step responses, frequency responses). Applications are drawn broadly from engineering and physics, including feedback and control, communications, and signal processing." name="Description">
    <meta content="Freeman, Dennis" name="Author">
    <meta content="signal and system analysis,representations of discrete-time and continuous-time signals,representations of linear time-invariant systems,Fourier representations,Laplace and Z transforms,sampling,difference and differential equations,feedback and control,communications,signal processing" name="Keyword">
    <meta content="" name="Version">
    <meta content="6.003 Signals and Systems | Electrical Engineering and Computer Science" name="Search_Display">
    <!-- End Automatic Metadata Insertion -->

[Greg Grossmeier]

I may have been unclear in my use of "not required" in my response.

The example code you quote is indeed usable and it one way of embedding
metadata into a webpage. It isn't always the best way (marking up
already visible content with tags is the best case scenario).

What is not required is the use of the word "meta" for it to be
metadata embedded in a webpage.

Also, the bit of html you quoted below does not contain any use of LRMI
specific terms, just fyi.

Greg

<quote name="Monty Swiryn" date="2012-06-14" time="13:59:44 -0700">

> Hi Greg,
>
> That helps a lot. Thanks.

> A follow-up... In example #2 “*Signals and Systems*”, the source contains

[Monty Swiryn]

Thanks. That makes it clearer.

So... (sorry about another clarification request)...  If the code snippet does not contain any use of LRMI specific terms, do you know why it is included in the example on the LRMI site? 

Also, do you know what tool was used to do the "Automatic Metadata Insertion"?

thanks again,

Monty

[Greg Grossmeier]

<quote name="Monty Swiryn" date="2012-06-14" time="14:19:18 -0700">

> Thanks. That makes it clearer.

Glad that helped.

> So... (sorry about another clarification request)... If the code snippet
> does not contain any use of LRMI specific terms, do you know why it is
> included in the example on the LRMI site?

Because it is a part of the original page. The examples take real world
pages as they are and mark them up with LRMI metadata.

> Also, do you know what tool was used to do the "Automatic Metadata
> Insertion"?

Nope, but the person who works for MIT OCW who helped with that example
is on this list and might be able to reply.

[Monty Swiryn]

Great, thanks again. And thanks for your speedy replies.  They are much appreciated.

Monty

[Vijendra Kushwaha]

Hi Greg,
About the example you gave : in a similar example from the MIT OCW, the information has been marked a little differently - <p>This work is produced primarily for use by <span itemprop="typicalAgeRange" content="18-21"></span> Undergraduates</p>. That is, instead of delivering information as plaintext, the "content" attribute was used. One could have used <metadata> tag to deliver the same information(I understood a few other things when you said that using metadata tag alone is not advisable.. thanks for that) or using some other attribute.

I am trying to imply here that there are many ways of giving out the same information. So, if I were to try to parse the page looking for the information, I would have to know exactly how it has been presented. For the examples here, I would have to know if the information has been given in plaintext(between the start and end tag) or as a value to an attribute(like under [content]).

Could you address this in the direction of making parsing more generic?

sorry to be a bother
Vijendra Kushwaha

[Dan Brickley]

On 15 June 2012 14:15, Vijendra Kushwaha <vijend...@gmail.com> wrote:
> Hi Greg,
> About the example you gave : in a similar example from the MIT OCW, the
> information has been marked a little differently - <p>This work is produced
> primarily for use by <span itemprop="typicalAgeRange"
> content="18-21"></span> Undergraduates</p>. That is, instead of delivering
> information as plaintext, the "content" attribute was used. One could have
> used <metadata> tag to deliver the same information(I understood a few other
> things when you said that using metadata tag alone is not advisable.. thanks
> for that) or using some other attribute.
>
> I am trying to imply here that there are many ways of giving out the same
> information. So, if I were to try to parse the page looking for the
> information, I would have to know exactly how it has been presented. For the
> examples here, I would have to know if the information has been given in
> plaintext(between the start and end tag) or as a value to an attribute(like
> under [content]).
>
> Could you address this in the direction of making parsing more generic?

There are a few software libraries around, e.g. in Java Any23
http://any23.org/ which hide some of these details. Look for
'microdata parser' or if using rdfa, 'rdfa parser'...

cheers,

Dan

[Peter Pinch]

It was my understanding that one of the goals of schema.org is that it's possible to add semantic markup without having to repeat data.

In the cases where the metadata appears in the page (e.g. contributor) it's easy enough to wrap it in a tag with an itemprop attribute (line 408)

But when the metadata doesn't appear in the page, or it appears in the wrong format — e.g. I have "undergraduate" but LRMI wants an age range — I used the content attribute. (line 418).

I believe these are all supported uses of schema.org. I hope some of the libraries that Dan Brickley suggested make parsing easier.

- Peter

-----------

Peter Pinch

Production Manager, MIT OpenCourseWare

pdp...@mit.edu

http://ocw.mit.edu

[Monty Swiryn]

May I make a suggestion for the spec page on the LRMI site (http://www.lrmi.net/the-specification)?

Perhaps it would clear up some confusion if we could see examples of the proper use of each markup tag.  Currently there are columns for Property, Expected Type, and Description.  How about adding a fourth column for "Markup" that would show the actual code of the use of the tag?

Forgive me if I'm being too naive here...

thanks,

Monty

[Greg Grossmeier]

<quote name="Monty Swiryn" date="2012-06-15" time="08:21:39 -0700">

> Perhaps it would clear up some confusion if we could see examples of the
> proper use of each markup tag. Currently there are columns for Property,
> Expected Type, and Description. How about adding a fourth column for
> "Markup" that would show the actual code of the use of the tag?

Showing a full example for even just one term/property might be a bit
too long to fit in the column structure on that page. But I can pull out
some of the markup from the example pages and list them on separate
pages per property/term? Does that sound helpful? That might be a lot of
individual pages that are hard to keep track of and navigate, though.

[Monty Swiryn]

Yes, good idea... that would be helpful. I don't think we need a lot of code for each example.  Perhaps just one line for each property.

Ex:

<meta itemprop="intendedEndUserRole" content="teacher">

Thanks again,

Monty

[Monty Swiryn]

Hi,

I'm trying to get a handle on what the important differences are between the original and the LRMI versions of the three posted examples.

I did a diff of the original vs. the lrmi versions of each set.  (Files attached)  Would it be possible for someone to cull thru the diff files and edit out anything that is not illustrative of the lrmi markups?  That might help us better understand exactly what was done to the pages from the LRMI point of view.

I apologize if this is a big hassle.  If so, no need to bother...

thanks

Monty

On Saturday, June 9, 2012 8:23:52 AM UTC-7, Greg Grossmeier wrote:

[Peter Pinch]

Hi Monty. I was hoping this diff would be clearer than the one you sent, but it still shows some changes that are irrelevant:

https://github.com/pdpinch/LRMI-test-6-003/commit/782ba5fb9eb0703b690e53f579bb3ea4aa57c6c5

The relevant line numbers are:

19 - setting the itemtype to a web page and the itemscope

194 – identifying the link to a previous version of the course

252 – identifying the name of the course

253 – identifying the published date (I may consider changing this to use the Event itemtype, since this is really the date the course was given)

269 – identifying the thumbnail

287 – Using the Person itemtype to identify the author (MIT OCW considers the faculty to be the author of the course) 

297 – Using itemprop and content attributes to indicate that the "Undergraduate" label indicates the course is intended for 18-21 year olds. I'm not sure this is what LRMI intends and, frankly, it's still counterintuitive to me. I'm tempted to try using an alignmentObject as I think Greg was suggesting. 

301-305 – Various metadata that don't appear on the page: interactivityType, educationalUse, learningResourceType and intendedEndUserRole

325 – Identifying the description of the course

330-331 – adding a couple lines to indicate the time required (this was actually moved from another page on the site)

(The rest of the diffs are junk, something to do with line-endings I think.)

At the time I wrote this, I didn't intend it to be normative. So I will gladly and humbly accept any feedback. 

I am still hoping to take another run at this with a course that has video, and for which I'll try to tag at the resource level in addition to the course level. 

Good luck,

Peter

-----------

Peter Pinch

Production Manager, MIT OpenCourseWare

pdp...@mit.edu

http://ocw.mit.edu

From: Monty Swiryn <msw...@gmail.com>
Reply-To: "lr...@googlegroups.com" <lr...@googlegroups.com>
Date: Monday, June 18, 2012 4:45 PM
To: "lr...@googlegroups.com" <lr...@googlegroups.com>
Cc: LRMI TWG <lrmi...@googlegroups.com>
Subject: Re: LRMI 1.0 Examples Listing

[Monty Swiryn]

Thanks, Peter.  This is a great way to see an example of actual LRMI tagging.  Hope your example will be a model for more to come.

Monty

[Vijendra]

Hi,

I have just recently started on lrmi and there are a certain queries that keep coming up. Coincidently, Monty asks a major share of mine and I am very glad that most of the queries are answered very quickly.

I am currently working on a template page that will allow users(course managers, instructors...) to markup the data with ease. Though an easy task, I still have not got hold on certain things. I have some queries, forgive me if they are too trivial -

Firstly, it would be great if someone could give a correct usage of tag educationalAlignment. It seems none of the 3 examples contain a detailed use of this tag and its other properties.

Secondly, the tag typicalAgeRange would lose some effect when we start marking-up University courses. Like the course page from MIT OCW by Peter, undergraduates have been mentioned with 18-21. How about if we could use a tag on lines of prerequisites? University courses have that property wherein you cannot opt for a course unless you have already done another.
Similarly, even at school level, you cannot do a learning resource, like multiplication, if you have not done a prerequisite, maybe say arithmetic addition. Or do integration without having done differentiation. Leaning has to follow an order. We could exploit that order to get us a better markup.

Thirdly, about differences in marking-up a full course and a resource level(a lecture or lesson). Thankfully, Peter is already working on it.
That a course is simply a collection of resources put together - is a course considered a learning resource?

Lastly, maybe a little too trivial, but are we supposed to stick to the tags(Properties) given provided by lrmi or we can use the vocabulary used in schema.org? The examples have used properties such as 'author' for 'creator' and 'datePublished' for 'created' etc(as in Peter's example).

regards
Vijendra

[Greg Grossmeier]

Hello Vijendra,

<quote name="Vijendra" date="2012-06-20" time="04:02:53 -0700">

> Hi,
>
> I have just recently started on lrmi and there are a certain queries that

> keep coming up. Coincidently, *Monty* asks a major share of mine and I am

> very glad that most of the queries are answered very quickly.

Very glad the FAQ is already proving useful.

> I am currently working on a template page that will allow users(course
> managers, instructors...) to markup the data with ease. Though an easy
> task, I still have not got hold on certain things. I have some queries,
> forgive me if they are too trivial -
>

> Firstly, it would be great if someone could give a correct usage of tag *
> educationalAlignment*. It seems none of the 3 examples contain a detailed

> use of this tag and its other properties.

I'm all for more examples, but I would prefer to have someone propose a
page that they attempted to mark up and got stuck. Real world examples
are better than me (or anyone) making stuff up. The examples already
there have been marked-up using educationalAlignment/AlignmentObject to
the extent that the information on the pages allows.

> Secondly, the tag *typicalAgeRange* would lose some effect when we start

> marking-up University courses. Like the course page from MIT OCW by Peter,
> undergraduates have been mentioned with 18-21. How about if we could use a

> tag on lines of *prerequisites*? University courses have that property

> wherein you cannot opt for a course unless you have already done another.
> Similarly, even at school level, you cannot do a learning resource, like
> multiplication, if you have not done a prerequisite, maybe say arithmetic
> addition. Or do integration without having done differentiation. Leaning
> has to follow an order. We could exploit that order to get us a better
> markup.

Prerequisites can be modeled using the
educationalAlignment/AlignmentObject paradigm using the alignmentType
property and putting "requires" as the value. The use of the requires
value was conceived for use with educational standards (eg: Common Core
in the US) but could be expanded to meet the needs of any particular
community.

It would really only make sense within the one institution that has
CS-1003 as a prereq for CS-2004 (or what have you); which means that the
information isn't useful outside of the institution's context (which
it isn't anyway). But if that is all that is needed then that should be
fine.

> Thirdly, about differences in marking-up a full *course* and a resource
> level(a *lecture or lesson*). Thankfully, *Peter* is already working on it.

> That a course is simply a collection of resources put together - is a
> course considered a learning resource?

A course *could* be considered a resource, indeed. There is nothing
stopping one from marking up a page that describes a course with LRMI.
There are places where LRMI wasn't designed to meet some needs (eg:
describing, explicitly, the structure of a course) but it does allow you
to describe the facets of the course that are probably most important to
a learner/searcher.

> Lastly, *maybe a little too trivial*, but are we supposed to stick to the

> tags(Properties) given provided by lrmi or we can use the vocabulary used
> in schema.org? The examples have used properties such as 'author' for
> 'creator' and 'datePublished' for 'created' etc(as in Peter's example).

LRMI is an extension to Schema.org, thus, it builds on Schema.org, thus,
all of Schema.org is available to you.

In fact, a more correct way to phrase this is:
You are using Schema.org vocabulary to mark up your content, you happen
to be marking up educational content, thus, you will be using the
educational specific terms in Schema.org (those terms just happen to
have been created by the LRMI Technical Working Group and wider
community).

Of note: Schema.org has not made the inclusion of LRMI terms live on the
Schema.org vocabulary pages yet.

Hope that helps,

[Vijendra]

lovely !!!
thanks Greg for the reply, it is a great help

regards
Vijendra

[Monty Swiryn]

Hi folks,

On Wednesday, June 20, 2012 8:44:31 AM UTC-7, Greg Grossmeier wrote:

Hello Vijendra,

<quote name="Vijendra" date="2012-06-20" time="04:02:53 -0700">

...

>
> Firstly, it would be great if someone could give a correct usage of tag *
> educationalAlignment*. It seems none of the 3 examples contain a detailed
> use of this tag and its other properties.

I'm all for more examples, but I would prefer to have someone propose a
page that they attempted to mark up and got stuck. Real world examples
are better than me (or anyone) making stuff up. The examples already
there have been marked-up using educationalAlignment/AlignmentObject to
the extent that the information on the pages allows.

It seems that the confusion that Vijendra and I (and others) have about the correct use of a tag could be due to the need for an explanation of the basic syntax, such as you find in usage references for other markup "languages" or commands.  These references show the proper usage of the tag/command along with explanations of their parameters.  They are typically followed by brief examples of the usage.  As an LRMI newbie, I find it difficult to come up with examples of a tagged resource page without a starting point for "how to". 

For example, the "man" pages for unix commands typically start with something like:

NAME

     useradd - administer a new user login on the system

SYNOPSIS

     useradd [ -c comment  ]   [  -d dir  ]   [  -e expire  ]   [

     -f inactive ]  [ -g group ] ... 

A good example for css markup is on this page:

https://developer.mozilla.org/en/CSS/position

Syntax
 position:  static | relative | absolute | fixed | inherit

I think it would be super helpful to have this type of reference for the nine LRMI-specific tags.

thanks,

Monty

[Greg Grossmeier]

<quote name="Monty Swiryn" date="2012-06-20" time="11:34:01 -0700">

> Syntax
> position: static | relative | absolute | fixed | inherit

Unfortunately, that is comparing apples to oranges. Schema.org and LRMI
do not have an absolute set of values for any term. The only exception
to this is in LRMI where we have a recommended list of terms for
alignmentType (as indicated in the description of that term). See:
http://www.lrmi.net/the-specification/alignment-object

But, to be clear, there is no list of all possible values for any term.

Also, how would that be substantively different than what is available
on the specification page on lrmi.net? There we have:

Term - The expected type of that term (ie: the type of value that is acceptable) - Description

You can see a similar layout on Schema.org:
http://schema.org/CreativeWork

Schema.org does have cleaned html examples on that page, which LRMI is
lacking at the moment (in preference of having full fledge real world
examples first).

You can see one LRMI example in that form on this page:
http://www.w3.org/wiki/WebSchemas/LearningResources

The inherent issue with these examples is that they will be incomplete
in that I haven't found a real-life example that uses all LRMI terms.
Creating a fictional one is doable, but it might not map well to a real
world resource.

So, as always, if you have a specific example of content that you are
trying to mark up but you are failing some where, please share it with
us and we can work it out. The more of these real-life examples the
better.

[Phil Barker]

Hello,
May I recommend as a good explanation of how schema .org mark up works, with a fairly generic step by step example at
http://go-to-hellman.blogspot.co.uk/2011/07/spoonfeeding-library-data-to-search.html

I used this approach for a couple of examples in a presentation I gave recently:
http://www.slideshare.net/philb/what-can-schemaorg-offer-the-web-manager
(examples start at slide 13)

But the difference between something like schema .org mark-up and UNIX commands is that there is more than one way of doing schema .org, there's always a choice of which HTML container to use.

Hope this helps, Phil

-- 
<http://www.icbl.hw.ac.uk/~philb/>
Please note new email address: phil....@hw.ac.uk

---

Heriot-Watt University is the Sunday Times Scottish University of the Year 2011-2012.

Heriot-Watt University is a Scottish charity registered under charity number SC000278.

[Greg Grossmeier]

<quote name="Phil Barker" date="2012-06-21" time="16:32:31 +0100">

>
> Hello,
> May I recommend as a good explanation of how schema .org mark up
> works, with a fairly generic step by step example at
> http://go-to-hellman.blogspot.co.uk/2011/07/spoonfeeding-library-data-to-search.html

That's a great one that I completely forgot about.

Also, tangentially, Eric Hellman's Unglue.it just unglued their first
book, Oral Literature in Africa:
https://unglue.it/work/81724/

(It is sort of a Kickstarter for published books; enough funding and the
book is made available under the terms of the Creative Commons
Attribution license.)

Best,

[Monty Swiryn]

Great resource, Phil. Very helpful. Thanks for posting it.

Monty

[Monty Swiryn]

Hi Phil,

Great presentation at http://www.slideshare.net/philb/what-can-schemaorg-offer-the-web-manager
Just starting to browse your slides.  I'm wondering if you could clarify one item.  Starting with slide # 30, you're showing the example of the book "Principles of Physics".  You show "Academic Level: First year undergraduate".  In the subsequent slides showing the markup, however, I don't see the markup for "Academic Level".

For the resources on which I am working, that seems to be a very important property.

Am I missing something?

thx

Monty


On Thursday, June 21, 2012 8:32:31 AM UTC-7, Phil Barker wrote:

[Phil Barker]

Simple answer, in the context of that presentation the example comes
before the (brief) discussion of LRMI. In fact, that example was
generated before LRMI :)

How would I mark it up now we know the LRMI properties? Well, to my mind
that specific example of 'first year undergraduate' isn't a great fit
for the class of identifiable educational frameworks that the
AlignmentObject refers to, but ignoring that, this seems to work with
the rich snippet testing tool:

<p itemprop=**"educationalAlignment" itemscope itemtype="http://schema.org/alignmentObject">
<strong itemProp="alignmentType" content="educationLevel">Academic level:</strong>
<span itemprop="targetDescription">First year undergraduate</span></p>


If you know of URL that is commonly used to identify 'first year
undergraduate' in some accepted framework of education levels you would
be better using

<p itemprop=**"educationalAlignment" itemscope itemtype="http://schema.org/alignmentObject">
<strong itemProp="alignmentType" content="educationLevel">Academic level:</strong>
<link itemprop="targetURL" href="http://example.org/level/ug1">
<span itemprop="targetDescription">First year undergraduate</span></link></p>

If you actually want to link to the URL for first year undergraduate,
then use <a href=""> in place of <link href="">. (Have I said that there
is more than one way of doing this....)

Disclaimer: it's 8pm here; I'm rushing to finish this before my dinner.
While it looks OK to me there may be errors that someone can spot :-}


Phil

--
Ubuntu: not so much an operating system as a learning opportunity.

[Greg Grossmeier]

<quote name="Monty Swiryn" date="2012-06-21" time="12:27:35 -0700">
> Can anyone answer a further question about the "alignmentType"
> "targeURL"...?
> Here's an example of a specific standard from the Common Core State
> Standards:
> http://www.corestandards.org/the-standards/english-language-arts-standards/reading-literature/grade-3/#rl-3-3
>
> In other words, how would Google "know" to display "Reading Literature" in
> the rich text snippet?

Two complimentary ways:

1) Use the targetName and targetDescription terms in AlignmentObject as
appropriate.

2) Specifically for the Common Core standards, there will be metadata in
the pages at corestandards.org that describes the standard being
referenced by a URI.

See the quote from:
http://www.corestandards.org/developments-on-common-core-state-standards-identifier-and-xml-representation

"the embedded metadata within the HTML pages at www.corestandards.org
(to be available in the coming months)"

[Monty Swiryn]

Many thanks again.

Monty

[Greg Grossmeier]

<quote name="Greg Grossmeier" date="2012-06-21" time="12:52:30 -0700">

> 2) Specifically for the Common Core standards, there will be metadata in
> the pages at corestandards.org that describes the standard being
> referenced by a URI.

I should clarify that this metadata on those pages will be what Google
et al can learn what that URI represents.

[Joshua Marks]

Greg, Monty, Phil, et al

One important point here. The text wrapped by the educationalALignment tag
and associated with the reference URI should contain the text of the
statement being aligned to (No matter the source). Having the objective
statement or description in the tag itself allows semantic discovery without
the need to index the (a specific) external standard/framework structure. It
is not safe to assume Google or anyone else will utilize any specific
standard as a filtering taxonomy. Having the words of the statement being
aligned to in the tag itself assures the term associations are indexed as
part of the resource and not just the URI references.

This is perhaps the only way we will overcome the existence of multiple
promulgators for the same standard framework.

Consider lines 18-26 in this example ->
http://www.w3.org/wiki/WebSchemas/LearningResources

Now imagine there are other very similar or related resources in other
locations that are tagged with the same alignment, but the ASN URI is
replaced with the equivalent (or similar) AB or CCSSO hosted URIs. Or
imagine a different objectives or sequence based framework covering similar
skills and utilizing related terms. As the terms and natural language of the
objective statements are associated and indexed with the resource, similar
resources are more likely to be found. Furthermore, similarities and
relationships between frameworks can also be found by virtue of their
association with common resources and their semantic similarity.


Joshua Marks
CTO
Curriki: The Global Education and Learning Community
jma...@curriki.org
www.curriki.org

I welcome you to become a member of the Curriki community, to follow us
on Twitter and to say hello on our blog, Facebook and LinkedIn communities.

-----Original Message-----
From: lrmi...@googlegroups.com [mailto:lrmi...@googlegroups.com] On Behalf
Of Greg Grossmeier
Sent: Thursday, June 21, 2012 1:06 PM
To: lr...@googlegroups.com
Cc: Monty Swiryn; LRMI TWG; phil....@hw.ac.uk
Subject: Re: LRMI 1.0 Examples Listing

[Zoe]

Hi all,

I've been away on holidays - just catching up now...

I know I've said this before, but it bears repeating - Vijendra is, in my opinion, quite right. 'Age' is a very poor proxy for 'level'.

The example Vijendra gives of university study shows this clearly.

Can I also suggest that we think of the implications for special needs education, where 'age' is largely irrelevant to 'level'. I very much doubt that any SEN teacher would react positively to having resources for his high school class described as being for pupils age 7-9...

These are, however, special use cases. I can see cases in our materials where the 'age-as-proxy' problem also affects mainstream schools content - i.e. anyhwere the curriculum changes across borders, which is to say, everywhere.

[Dan Brickley]

Welcome back :)

So one thing I wanted to share with the LRMI folks ....

... in the last schema.org team meeting we agreed that (as far as we
can tell, being non-educationalists, etc.) LRMI 1.0 was ready for
inclusion in schema.org.

The schema.org partners delegated to me the task of sitting down
(virtually...) with Greg and working out the exact changelist,
including decisions for the current ongoing Audience thread.

In that light, can I encourage discussion here to framed in terms of
"what do we want to get included in schema.org in the next week or
two?" versus longer-term improvements, next phase of development, and
so on.

The suggestion under discussion in public-vocabs is that we couch
things more in terms of an explicit Audience class, which might be
refined and improved later (e.g. to avoid over-dependence on age as an
indicator of level). Is it feasible to find a quick compromise design
that leaves things open for richer elaboration later?

cheers,

Dan

[Joshua Marks]

Zoe,

Are you suggesting that audience include a reference to level and not just role? I believe we have previously covered that EducationalAlignment is suitable to encode alignments to leveling, topical or corpus based frameworks. -> http://wiki.creativecommons.org/LRMI/Properties/1.0/AlignmentObject

Is this not sufficient?

Joshua Marks
CTO
Curriki: The Global Education and Learning Community
jma...@curriki.org
www.curriki.org

US 831-685-3511

I welcome you to become a member of the Curriki community, to follow us on Twitter and to say hello on our blog, Facebook and LinkedIn communities.


-----Original Message-----

[Monty Swiryn]

Hi folks,

Great meeting you all at ISTE.  Your hard work on this project is evident.

As you may know from my posts, I agree with the complaint about using Age as a substitute for Grade Level.  In addition, the almost 2 dozen educational publishers with whom I work also agree.

It was mentioned that the logic for using Age instead of Grade Level was because Grade Level does not have the same meaning internationally.  However, Age is a poor substitute for Grade Level because age and grade level do not correlate.

The argument that the EducationalAlignment is suitable for Grade Level is not feasible for several reasons:

• If the user searches for a resource for a particular grade level, and sees Ages (but not Grade Levels) in the left hand column where checkboxes are displayed to refine the search, s/he will be frustrated, at the least.
• EducationalAlignment is recommended to contain a reference to a standards hierarchy.  But publishers who don't have their resources correlated to a standards hierarchy (and instead put the grade levels as text in the EducationalAlignment property), will be at a disadvantage.
• It is not clear what will appear in the left side of the search results as checkboxes for refining the search under EducationalAlignment. Users who are not searching by standards may not see what they are expecting under EducationalAlignment.
• One can argue that Ages can also be satisfied via the EducationalAlignment property.  So why include Ages as a separate property?  It seems to give Ages a higher priority than other EducationalAlignment sub-properties.
• Audience is not a suitable substitute for grade level, as Audience has a different meaning and purpose (e.g., Teacher, Special Ed Student, After-School Program...), and does not correlated with grade level.
• Publishers who classify their products by grade level will have a lot of extra work to do to "convert" their grade level classifications to age levels.  Why not give them a much easier option to tag their products as they currently do?

The solution is to add a "grade level" property, but call it "Academic Level".  This will satisfy the international issue. Since the "language" is wide open, publishers in the US can use language such as "Grade 6", while UK publishers can use "Sixth Formers".  Likewise, users who are searching for resources in a US curriculum would use the US curriculum language, while those searching for resources in a UK curriculum would use the British language.

In addition, Age Level should be changed to "Level" and support more than just age levels. It should support sub-properties so that publishers can enter different kinds of levels such as Reading Level, Lexile Level, Interest Level, DRA Level, etc.  Many of these levels are not contained in common standards hierarchies, but are very important to certain types of educators.  Maybe something on the order of a "multiple component" property, e.g., ItemProp=Level, ItemScope=Reading Level, ItemProp=5-6.  (Forgive my inaccurate syntax.)

Some interesting observations about EducationalAlignment seem to give evidence that the use of this property and the results it may have are as yet unknown, and difficult to implement and explain.  (Note: this is not to be critical of these examples, but to argue that the use of EducationalAlignment, especially as it pertains to Academic Levels, is far from ready for prime time.)

• At ISTE, there was a demo of a "proof of concept" that showed an idea of what Google/Bing/Yahoo search results might look like.  It was interesting that under EducationalAlignment in the left side column, actual values were not shown. I believe publishers and developers would very much like to see a proof of concept that illustrates the use of EducationalAlignment. I strongly encourage the people who are working on the proof of concept to continue to work on demonstrating this very important missing piece.
• In Phil Barker's presentation on how to use schema.org tags, the explanation of the EducationalAlignment is not included:
• http://www.slideshare.net/philb/what-can-schemaorg-offer-the-web-manager
• In a posting by Peter Pinch on June 18, 2012 8:41:40 PM UTC-7, he seems to imply that there is still some work to be done regarding how to tag the academic level "Undergraduate".

Having an "Academic Level" property would be much easier to understand and implement, and be in much more alignment with how educational publishers classify their products. It would also save them a ton of extra work.

Further, the same arguments hold true for a Subject Area property.  Subject and grade/academic level are the most important, top level educational resource classifications, and those by which educators primarily search for resources.

So what is the harm in including Academic Level and Subject Area properties?  There seems to be no downside, and plenty of benefits in adding these two properties.

If LRMI is primarily an academic exercise, then it's fine the way it is.  But if it's meant to have practical use and benefits to publishers, it should take into consideration their needs in these areas.  Perhaps the group should consult with a few more publishers, explain the schema, describe how the properties would be used, and get their feedback.

Many thanks,
Monty