Hi everyone,

I'm adding our draft FAQ content type for comments. We already use a FAQ content type in Percussion which creates an xml feed ingested by our Google Search Appliance. So when someone searches on a trigger word the FAQ Question and a 'short answer' appears directly on the SERP with a link to full FAQ page.

As we adapt this to a new Drupal content type, we are working through the following:

• We have recommended lengths for Questions and 'Short' Answers, but not a strict character limit. Some of the questions folks wish to include can become prohibitively long and complex. So we hope to show them visually why the shorter question is more useful but allow them to let that sentence run on as long as necessary if there is truly no option. We could have gone the short question /long question route as we have done with titles and descriptions on Article, but we decided against it.
• However, a 'short answer' is required. We want to encourage content authors to give quick answers upfront so that users know if the full FAQ is actually going to cover what they expect and / or if it can statisfy their goal without clicking through to the full FAQ.
• We have a site-wide hhs FAQ section as well as numerous Staffdiv / topic specific FAQs. We'd like to change the field name 'type' but haven't settled on a replacement - perhaps FAQ Set. Each set will have their own 'category' taxonomy as the topic specific faqs can get very SME specific.
• The Site-Wide FAQ, however, will conform to the same category taxonomy we use for Blogs and News Releases.
• We use an 'Audience' taxonomy as well (which is site-wide) as our FAQs can be targeted to general consumers, SMEs, Researches, Policymakers, etc. and should be written accordingly.
• We are considering adding an image field (which would not be required), but could be used for iconography or logos as well as images.
• Like our Article content type we will add Twitter and OpenGraph markup. I've only listed the Schema.org markup here.
• We identify the schema.org/question property in the markup header. However we aren't actually using any of the itemproperties from it (instead sticking with the basic name/text properties of CreativeWorks). Schema.org/question is tailored for Quora type q/a sites (that crowdsource answers and use upvoting for example). This is an area we've decided to keep simple for now and extend once we have more experience with Schema.org. If you have suggestions on this, let us know!
  

http://gsa.github.io/Open-And-Structured-Content-Models/

Should we mention that these models are mostly derived from schema.org? It might help folks sell it in their agencies if they can say that they're in cahoots with Google, Facebook, Etsy, Github, etc...

source: http://getschema.org/index.php/List_of_websites_using_Schema.org

Hi team - We have a challenge that I hope you all can help with. Since the development of DigitalGov.gov we realize that a lot of the resources</a href> (laws & regs/guidelines, tools and reference materials) we’ve curated and provide to agencies:

• aren't always owned by us (we link to laws / memos on OMB, WH etc)
• aren't (in a lot of cases) able to be searched (pdfs / docs)
• are popular and not easily findable
• are in need of better organization
• and / or a combination of these

We have been calling these bits of information “assets” and they include: documents, pdfs, powerpoints presentations, video, audio, photo, and outside links to these same items.

We took a stab at creating a template and realized that we were creating "fields" and essentially it could be a content model. We looked at the current article and we don’t think that it applies in this situation.

I have talked this through with @jpgsa and he suggested I take a look at Schema.org to see if there is an equivelent. It occurs to me that the Creative Works</a href> model or perhaps more specifically the Media Works</a href> (specific types below) might work well.
More specific Types:
AudioObject</a href>
DataDownload</a href>
ImageObject</a href>
MusicVideoObject</a href>
VideoObject</a href>

Ultimately, we envision a page for each asset with the minimal but most important information for the customer, tag the page, and use feeds and queries to pull the information via "search" specific and unique to the assets.

Does anyone have a similar approach or need like this? Also, just throwing it out there, do you all think this would be a good government-wide model?

Please let me know if you have questions about what we are trying to do.

Perhaps the conversation isn't one or the other but a combination of the two.

Some of the more simple needs could be met with a basic feature. (smallest possible piece)
Standardize field names, help editor text, etc

More complex requirements may require module development.

Features

• ideally no dependancies
• smallest possible piece
• content types and views are not in the same feature

Hi!
The working group appreciates your feedback on the current content models. Feel free to post questions and comments related to the content models in this space.

USA.gov wants to have API open for submissions from agency partners.
-USA.gov wants to help agencies get more mileage out of content agencies have prepared through their networks.
Is there value in this type of service for agencies? What are the questions? Opportunities? Questions?

Hi All,

Building off of last week's chat around managing change, check out this week's Responsive Design podcast from Karen McGrane and Ethan Marcotte featuring NPR mobile strategy. They ask about adaptive content around 18 minutes in. The guests talk about how their approach has changed. It involves taking away the "preview" button.

http://responsivewebdesign.com/podcast/npr.html

Here's a very short description of the project.

Purpose:

1. We want to improve our search engine results.
2. We want to improve our site/page maintenance.
3. We want to improve our inventory reporting capabilities.

Questions/Issues:

1. We want to know what pages we have; what content is used within those pages; what content is duplicated; and what content is related to multiple pages.
2. This issue is specifically for the GSA work group. Content modeling is one issue to contemplate and implement, but semantically tying that content to the page is another issue that is not described sufficiently. Guidance is sparse using general search engines (e.g. Google) which makes me believe the solution is typically relegated to an automated tool such as a WCMS. Problem is, some agencies don’t have that utility. Even so, not all WCMSs are consistent with that capability.

Approach:
Starting with a combination of the Google Custom Search for help identifying meta usage, the Screaming frog for a similar but different inventory, and internal policy reviews on implementing more useful meta tags, we will be able to automatically identify our pages. This allows a faster, more targeted, and more consistent solution. However, the content of those pages is loosely coupled at best.

If you can relate to this, I'm interested to know what your agency doing? How are you handling this type of scenario?

How can we easily leverage the Article and other models in Drupal?

After the recent workshop, I spoke to Erik Arnold at DigitalGov Search (DGS), who applauded the efforts of the Content Model group but pointed out (again) that because the DGS service doesn’t parse web content, they can’t directly use the markup to improve their search results. He suggested that the only player able to do that would be Google.

However, if the metadata is aggregated into a structured file, they can use it – and presumably the metadata becomes more useful to other third party processes. I’m interested if others are packaging their metadata and if that makes it more useful to other users?

A theme I heard at the workshop was how to sell structured content to management. Our markup may or may not improve Google search results for our content, but it may be hard to point to as a clear benefit.

I think we stand a better chance of demonstrating the value of structured content if we can package the metadata and point to third party apps that can consume it. In other words, markup is the first step, but an important second step is aggregation.

Thoughts?

-- Bob Rand, SEC

We're creating a Drupal site, and I want to work with our developer to incorporate the content model fields. I'm still learning Drupal, but what they tell me is the definition of "article" in Drupal is not quite how the content model treats article. Here is how the article content type is defined in Drupal.org:

"The Article content type (formerly, "story") is enabled in Drupal in the default installation profile. Articles are generally used for information that is updated more frequently and often cross-referenced and categorized (such as news items or resources). By default, Articles are sorted with the most recent post at the top, but this can be customized with contributed modules like Views."

My developer is encouraging me to use the Drupal "article" content type for press releases and blog posts, but not for simple informational pages.

I'm wondering if there has been any discussion about this. Do any of the Content Model workgroup members use Drupal? Should we be using the Drupal content type differently, or simply add content model data fields to a wider variety of content types, not just the article content type?

thanks, Lois Nilsen, State of NC

HHS.gov is creating new content types for a migration to Drupal in 2015. We've altered our fields to map to the GSA Article model. We intend to map the fields to Schema.org and OpenGraph markup in the page source as well (although this will be invisible to the editor.) I am adding images below because I see no other way to share xls or pdf files. The spreadsheet is what we edit with our developers. We use the other images when discussing with clients. This is very much a work in progress.

It's worth noting we've made the following decisions:

• We are not including ArticleSection (Title and Body) as required fields. We will only have one Body field, although the Detailed Description may also serve as a Summary in Drupal.
• We are using two dates only. Created Date is the date the item is initially drafted and operates as a 'published or posted' date as well.. If an item is published days or weeks later it will keep the Created Date unless an editor makes the change. Last modified date also requires manual editing by the content manager when an item is reviewed or updated.
• I just noticed there is an inconsistency... right now we expect to limit the characters in the short description to 160 (not 165).
• We are considering how we might identify FOIA related content in the markup so common electronic reading room items can be identified across agencies. DOJ is working on this (I think).

I'm sure there will be other questions. I'd be happy to discuss and we're looking forward to your feedback!

Bonnie Strong
Content Stratagist (Consultant) at HHS.gov

There have been a number of questions about making the case for structured content in government organizations and I've curated some articles the working group has written about "making the case" for structured content.

These articles include case studies on developing strategy and selling the approach in your organization.

Strategy
Open and Structured Content Models Workshop Recap
Top 5 Reasons to Use Government-wide Open and Structured Content Models
What Structured Content Can Do For You Webinar Recap
Government Open and Structured Content Models Are Here
Always Future Ready: The Benefits of Open Content Models and Structured Data Webinar
Why the National Cancer Institute Moved to a Structured Content Model

Changing the Conversation in Your Organization
Structured Content in Government: How Department of Health and Human Services and the National Cancer Institute Are Getting Started
The Content Corner: Humanizing Structured Content

You can always find the most recent Open and Structured Content posts at: https://www.digitalgov.gov/tag/open-and-structured-content-models/

Nice structures! I wonder if there are plans to build sample pages using these structures in WordPress and Drupal...and maybe a third content management system? Then test them with current federal content creators.