Scalable and future-proof knowledge portal. TheGMP-BERATER', a standard work in the pharmaceutical industry
Who is GMP-Verlag and what was this project about?
GMP-Verlag Peither AG is a German specialist publisher. He specializes in publishing products on Good Manufacturing Practice (GMP) for pharmaceutical companies and their suppliers. The knowledge portals GMP-BERATER (German) and GMP Compliance Adviser (English) are among the standard works in the pharmaceutical industry.
Take a look at the structure of the GMP-BERATER here.
The project we are presenting here was about future-proofing the GMP knowledge portals with a new solution based on modern technologies.
New, perhaps even today unknown, use cases should be made possible with a flexible content architecture.
users the portals are to be supported even better than before in accessing content. Because that is the core functionality of modern knowledge portals, that relevant information can be found quickly.
The users the portals and their expectations
Let's put ourselves in the shoes of the typical users. You need information on a specific topic quickly.
For example, as shown here in the video from GMP-Verlag
How does the new portal technology based on c-rex.net support finding the right answer quickly?
We'll tell you in advance: through clever use of metadata and targeted preparation of related information. Key words here are 'using object dependencies' and 'context search'.
From search and find
Finding the right information quickly is one of the central aspects in a knowledge portal.
In addition to simple content indexing, we have implemented advanced search functions for GMP. You almost want to say 'advanced find functions'!
Thus, content is weighted depending on metadata, i.e. a concrete context. This ensures relevance in the search results.
This can be both a context given by user selection and an implicit context.
Implicit searches run in the background depending on what users are doing or what role or licenses they are assigned. Their results are displayed if they match an action of the users.
Want an example? Of course!
When users open a specific chapter, additional information such as appendices, supplementary information or product recommendations related to that chapter via meta-information are dynamically displayed.
You may have seen it briefly in the video just now. Here again a screenshot for illustration.
Metadata and its generation in the publication process
Metadata is responsible for ensuring that search weighting works and that contextual information can be taken into account.
Now you may wonder how the metadata is obtained? Do writers have to enter metadata completely manually?
No, not for the most part. There is some metadata that is captured during content creation, for example, who is the author an article. However, much metadata is the result of the publishing process, which is performed with c-rex.net software components within minutes of document release and includes the following steps.
Contents are restructured
Contents are modularized
Content is classified with metadata
Content is full text indexed
License information is assigned
Graphics are prepared
HTML modules are generated
All these measures lead to a more precise access to content and provide users with appropriate results. And by the way: we naturally work according to the single-source principle. The classic print editions are also published from the same sources that fill the online portals.
Speaking of licensing
Parts of the portal require a valid license to gain access. Currently, licenses are assigned at the portal level, but in the future they can also be restricted down to chapters and subchapters in a dedicated manner.
If users have a license for GMP-BERATER and/or GMP-Compliance Adviser, they get access to all contents of the respective portal.
Two different licensing options are offered to GMP's customers, a personal license or a corporate license as a monthly or annual subscription.
The licenses are maintained in GMP's CRM system and regularly synchronized with the user administration (c-rex.net Identity) in an automated manner. The leading system is the CRM.
An extended licensing model is being discussed as a future expansion stage. It should also be possible to license only individual chapters, even for a limited period of time.
For this purpose, one or more license validities are assigned to individual chapters and then compared with the assigned licenses of the user administration when they are accessed.
Furthermore, additional portals based on Mediawiki are to be connected to the user administration in the form of a single sign-on.
This means that all customer and license information for all GMP applications is maintained centrally in the CRM and made accessible to all applications via the central user administration.
The CRM is thus the single point of truth for all applications from GMP-Verlag, and the user administration or integrated license management assigns the necessary access rights centrally.
More details on the standard and special properties of the GMP knowledge portal
Some features of the portal came more or less by themselves with the switch to c-rex software components, because they are standard components of c-rex.net and state-of-the-art nowadays.
For example, the Responsive Design. The display of the content adapts to the end device on which the portal is used.
Or the output of search suggestions when typing a term into the search slot.
Adding and managing chapters as favorites, printing selected chapters or pages are also two functions that are part of c-rex.net standards.
Other functions that focus on users and their needs have been programmed specifically for GMP-Verlag. Sometimes just little things, but important for the target group and tailored to them.
This includes the display of the found hits in the table of contents.
The benefit for users here is to be able to quickly navigate to the relevant places and to be able to limit the result set to practice or regulation sections.
For the areas General, GMP Practical Knowledge and GMP Regulations, the total number of hits is displayed in the table of contents.
Behind each chapter name is the number of hits in the respective chapter.
Clicking on a chapter displays the hit breakdown in the respective subchapters.
This allows users users to jump directly to the content in the desired context via the familiar chapter structure and further narrow the search results at chapter or subchapter level.
The hit list itself is displayed in the main content window. There you can display a preview via the eye or open the chapter directly.
In the chapters, the hits are highlighted in color. Arrows take you from one location to the next.
In the following screenshot, you can see another helpful feature that comes into play when searching. We call this function'quick response'.
If a glossary entry or an abbreviation is found via the search term, the content appears directly as a highlighted answer at the top of the hit list.
Customize the display of the content is another small feature we implemented specifically for GMP-Verlag improve the user experience.
users can hide the header and navigation elements to have a larger reading area.
A very practical function are the tooltips. Terms from the glossary and the list of abbreviations are automatically tagged by c-rex.net.
This can be used to highlight occurrences of these terms in the text. Clicking on the term dynamically displays the contents of the glossary term or abbreviation.
Content creation by the editorial team and authors authors
Let's take a look at how content gets into the portal. This is done in two different ways for the GMP-BERATER.
Let's first look at the content creation for the practical part.
In this area, the editorial staff of the cooperates with authors authors.
These authors create their articles in MS Word. The communication between the authors and the editorial office runs via Word.
After approval, the article is converted from Word to structured FrameMaker and then to XML.
The situation is only slightly different when it comes to creating content for the regulations section.
Here, the editorial team is constantly researching updates and new guidelines that affect the GMP environment. The final files for upload are available in PDF.
This PDF data is transferred to the structured FrameMaker and then to XML.
Regulations that are available in English are translated (and marked as translation) by the technical editors.
As soon as new content is ready, it is transformed by our data conversion service c-rex.net DTS, enriched with metadata and published via the c-rex.net IDS Content API.
This is the publication process already mentioned above.
Online and offline
The portals are currently delivered on a DVD or can alternatively be downloaded from the website. This means that customers also have offline access to the information if the Internet is not available.
For this purpose, the online application is made available locally with full functionality offline.
Modern technology in the background stands for scalability and future security
Attention! Now comes some info for nerds and techies.
If you don't like that, just skip to the next section.
Scalability and future-proofing result from the use of standards (content and technology) as well as our API approach based on the REST architecture.
OAuth2/OpenID Connect for the single sign-on (SSO mechanism)
Linked Data approach based on iiRDS
Content API based on REST architecture
Headless CDP, API-Driven Content Delivery
Faceted search, full text search, based on Lucene
Use of object dependencies in and outside the portal
HTML5 and Responsive Design
Conclusion and outlook
The essential part about the solution is the headless CDP or API-driven approach.
On the one hand, this approach enables the implementation of the GMP knowledge portals. On the other hand, it also guarantees future viability.
In this way, the content that is currently available in the portal can be integrated into any other portals.
It will also be possible for customersintegrate the content directly into their intranet in to order to provide their users with direct access to the necessary information.
More. In the future, loosely coupled user-generated content can be dynamically integrated into documentation.
Alternatively, customerstheir own customized documentation can receive and integrate content from the knowledge portals into their own documentation.
We are talking about integration by reference, which ensures that this inserted content automatically updates itself when there are new features.