RE: P&P suggestions

Subject: RE: P&P suggestions
From: "Sweet, Gregory (HEALTH)" <gregory -dot- sweet -at- health -dot- ny -dot- gov>
To: "techwr-l -at- lists -dot- techwr-l -dot- com" <techwr-l -at- lists -dot- techwr-l -dot- com>
Date: Tue, 23 Sep 2014 13:44:59 +0000

There are those of us in government service who take the content in our care, and the ability of the average person to find and use it very seriously. Over my tenure with the NYSDOH Health Commerce System, we've implemented every single one of Tony's recommendations, and more.

The case study is below. Jay, feel free to email me off-list to dive into any of these details.

Our site:
* Secure extranet requiring positive proof of identity to access.
* ~100k document library.
* ~260 applications.
* ~275k users.

When I started here 10 years ago, site content was organized by agency program, and yes real users described our site as "The library of Congress, in random order." But since then we have implemented a portal and content management system to organize content from the consumer's point of view.

*Metadata and abstracts are required.

The CMS, among requiring other metadata (dates, abstract, etc.), forces content providers to tag content with information about who that content is useful to. The content providers gets to see all of their content in one place, while the content consumer gets a pre-filtered view of the data based on the tags.

* Strict implementation of tags.
* Tags presented as "folders" to users.
* There is flexibility for incrementing values.

The "folder" structure created by this approach is strictly restricted to five levels deep so no single document is ever more than 6 clicks from the homepage.

Content creators cannot create new tags on the fly but must select from the tag taxonomy, or provide a really strong use case as to why what we've provided will not work.

We are flexible on tags that should increment. For example so documents should be tagged by year. No one need request a 2015 when 2015 arrives, we simply increment the tag as necessary.

* Expiration dates are required.
* Renewals are automated by the content submission tool.
* Review is required to renew an expiration.

Every document must also have an expiration date with the farthest date out being two years from publication, and the default set to six months. The CMS automatically sends email reminders to the content poster/owner 30 days prior to expiration and offers an opportunity to renew the expiration date should the document still be relevant. Content providers must open the original posting request and review the document before an expiration date renewal button becomes available.

* User base is authenticated and known so content is pre-filtered.
* Pre-filtering does not preclude browsing or search almost all content.
* Some content reserved for exclusive groups.

As our site is secure, so we go a step further and pre-filter content by assigning users to specified groups. For example if we know you are a physician in a hospital organization you are automatically assigned to the hospital and physician groups. All users can self-select which groups to subscribe to, include un-subscribing from your pre-assigned groups. There is one group reserved for government employees.

* On initial implementation, every document was reviewed.

When we set the system up, one of the largest tasks we had was forcing the content owners to review the content they had on the site. We distributed spreadsheets with all documents (and their URLs) listed to the content owners and required they be classified into the new taxonomy. Sufficient time was allotted for review. Any document that was not classified was not brought forward into the CMS. Almost as much content was marked expired and removed as was classified and brought forward.

* Search is provided by a dedicated Google appliance.
* Indexes are constantly tuned to assure search accuracy.

So with organization sorted out, let's talk about search. We use a Google appliance to provide search. Search cuts across nearly all pre-filtering groups (remember the one group for gov. employees -- you cannot search that unless you are a gov. employee), so it will find the most relevant content even if you have not subscribed to the particular group a document is in. Search executes a full text and metadata search of documents. So you can search for actual document text or by any metadata applied to the documents. We pay close attention to the search logs to tune the system matching most relevant content to most frequent search terms.

* Usability testing suggests we've got it right.

In practice, our usability research shows, that despite the pre-filtering a large majority of users prefer to search for content and a majority of content is found accurately within three clicks. In actual user letter-grade scoring of the site we went from a D- to a B+ inside eight months, and have only improved from there. Out latest enhancements introduce RWD and a new service architecture that has the site running blistering fast, even over 3g/4g cellular networks.

* Server-side tools for bookmarking content built-in.

We provide inbuilt bookmarking of applications and documents, so users can literally sit down at any internet connected computer in the world and have their most important content at their fingertips immediately upon log in.

Our efforts right now are focused on replacing the outdated back-end pieces that make this all work. There's no real change from the user's perspective, except possibly noticing a faster, more stable site.

Sure there's lots of crappy websites out there, government or no, but there are those of us who work hard every day to make sure we get it right.


> -----Original Message-----
> From: techwr-l-bounces+gregory -dot- sweet=health -dot- ny -dot- gov -at- lists -dot- techwr-l -dot- com
> [mailto:techwr-l-bounces+gregory -dot- sweet=health -dot- ny -dot- gov -at- lists -dot- techwr-l -dot- com]
> On Behalf Of Peter Neilson
> Sent: Friday, September 19, 2014 3:38 PM
> To: techwr-l -at- lists -dot- techwr-l -dot- com
> Subject: Re: P&P suggestions
> The usual way to arrange "government" documents is according to the
> internal structure of the organization.
> Thus the dogcatcher (NOT under Animal Control as you would think) is
> strangely subservient to the secretary to the City Council. The only way the
> interested reader can find the dogcatcher is by clicking on "City Council" first,
> then on "Staff" and then on "Lucy Millner" that being the name of the former
> secretary. THEN you can get to dogcatcher. And yes there is a search, but it's
> not properly implemented, and only displays a photo of City Hall, labeled
> Mayor's Office.
> Ya gotta knows tha territory before ya can reads tha map!
> Oh, and Animal Control is part of Parks and Recreation, of course.
> On Fri, 19 Sep 2014 12:15:53 -0400, Jay Maechtlen
> <techwriter -at- laserpubs -dot- com> wrote:

Read about how Georgia System Operation Corporation improved teamwork, communication, and efficiency using Doc-To-Help |


You are currently subscribed to TECHWR-L as archive -at- web -dot- techwr-l -dot- com -dot-

To unsubscribe send a blank email to
techwr-l-leave -at- lists -dot- techwr-l -dot- com

Send administrative questions to admin -at- techwr-l -dot- com -dot- Visit for more resources and info.

Looking for articles on Technical Communications? Head over to our online magazine at

Looking for the archived Techwr-l email discussions? Search our public email archives @


P&P suggestions: From: Jay Maechtlen
Re: P&P suggestions: From: Peter Neilson

Previous by Author: RE: Questions about the work day for the Whirlers
Next by Author: RE: Document tagging and metadata (was: P&P suggestions)
Previous by Thread: Re: P&P suggestions
Next by Thread: Re: P&P suggestions

What this post helpful? Share it with friends and colleagues:

Sponsored Ads