Content Management

1
Content Management
SGN-5226 Content Sharing Technologies and Services
15.11.2010
2
Background
Organizations have a lot of content. The value comes when:
• Content turns into information
• And information turns into knowledge
This is not possible, unless somebody knows:
•
•
•
That the content is there
How to access it with minimal effort
H
How
tto repurpose it iin order
d tto create
t new iinformation,
f
ti
ffrom existing
i ti content
t t
Content that is not accessible is not an asset for an organization
15.11.2010
3
Sample problem scenarios
A company has content from many of its suppliers
• Catalogs
• White papers
• Product documentation
• Technical manuals
• Bulletins
Theyy are delivered in p
paper,
p , or electronically
y ((CDs,, Internet).
)
Most data ends up being duplicated in different sources, and when an error
is spotted, it is never really fixed in all the instances…
15.11.2010
4
Sample problem scenarios
In another company, an employee in finance collects information coming
form multiple sources:
• From colleagues and from external partners
• Through official and unofficial documents
documents, via memos and emails
Information is located at multiple servers (intranet, extranet, Internet), but
she does not always have access rights to the resources she needs.
She would like to have a system that simply makes her life easier, if it would
take care of those content seeking and access issues.
15.11.2010
5
Content Management
Managing content is a great challenge for organizations, when sharing
content with employees, customers, partners, suppliers.
Core Requirements: Content must be
•
•
•
•
•
•
•
Easy to find
Accurate and always up-to-date
Complete, in order to satisfy the needs of its users/consumers
Quick to search and retrieve
Translated in the different needed languages
Linked/associated to other relevant content
Customized for the needs of each user/consumer and his/her experience and
kno ledge le
knowledge
level
el
15.11.2010
6
Content Management
Content Management is not about tools or technology (even though they
play an important role)
Content Management is about organizing,
organizing categorizing and structuring
information resources, so that they can be stored, retrieved, published
and reused in multiple ways.
It alls
ll starts
t t with
ith an Information
I f
ti M
Model.
d l It is
i the
th framework
f
k for
f organizing
i i
content, as an answer to the users’ needs, and setting the structure
under which the content is stored.
• Build your Information Model -> Manage it with the appropriate content
management technology -> Deliver information in multiple dynamic ways
15.11.2010
7
Conceptual Model
Delivery
Authoring /
Repository /
Assembly
Information Model
User Community
15.11.2010
8
Most organizations originally designed their websites
statically
Templates
es
Home
Page
Page
1
Page
2
Page
3
Page
n
15.11.2010
9
After a few expansions…
Templates
Home
Page
Page
n
Page
n
Page
n
Page
n
Page
2
Page
1
Page
n
Page
n
Page
n
Page
3
Page
1
Page
n
Page
n
Page
n
Page
n
15.11.2010
10
… ended up unmanageable
Templates
Home
Page
Page
n
Page
Page
n n
Page
n
Page
n
Page
n
Page
n
Page
1
Page Page
n
1
Page
n
Page
n
Page
n
Page
2
Page
n
Page
Page
2 n
Page
n
Page
3
Page
3
Page
1
Page
Page 1
n
Page
n
Page
n
Page
n
Page
n
Page
n
15.11.2010
11
Why it does not scale?
• Mixed presentation (HTML) and data is problematic
• Slow process
• Hand coding is an individual process (humans needed)
• Posting to the web – again individual process
• Errors discovered needed time/effort to be fixed, updated everywhere and
redeployed
• Links needs to be checked everywhere
• Affect
• Wrong data can be costly -> service calls, employee mistakes, lawsuits
N d ffor automated
Need
t
t d processes in
i ttwo levels:
l
l
• Posting and author workflow
15.11.2010
12
Implementing a Content Management Solution
We need tools for supporting four core components
• Content acquisition and authoring
• Content management (repository for storing & retrieving content)
• Content assembly and linking
• Content delivery and publishing (towards customers/users)
15.11.2010
13
Content Management System core components
4. Publishing
1. Authoring
2. Repository
3. Linking and
Assembly
15.11.2010
14
1. Authoring and acquiring content
15.11.2010
15
Authoring and acquiring content
The process of authoring creates the components of content for an
organization
Authoring starts with every business function that documents its activities
and reporting its work: employees writing memos and emails, create
reports and documents. Moreover, a lot of material is coming from
outside the organization.
Lets use the term “information asset” for all kinds of information, including
text, graphics, slide presentations and other media.
15.11.2010
16
Authoring and acquiring content
Diversity in authoring, throughout an organization
• Different tools used of authoring, from a variety of applications
• Word processing, spreadsheets, email, etc.
When creating plans for a content management system, we need to study
the different models of authoring that occur in the organization
Examples:
• Unstructured authoring (no templates of style tags), e.g. e-mail
• Forms-based authoring (forms completed with field based content), e.g.
content stored in databases
• Format-based authoring (templates and style tags), e.g. Word
• Structured authoring (content based templates and content tags), e.g.
XML
15.11.2010
17
Unstructured authoring
• Most authoring us typically unstructured in an organization. Typically
because
• Authors do not believe they have a need for structure in the document they
create. Thy just consider the document as a whole, and add a new header,
figure or table when they simply feel that it is appropriate
• Authors do not want to take the to learn how to use style templates to create
documents
People typically work with freely drafted emails, memos and many times
find a similar document with the one they want to create, and work on it.
• They take the original, delete some sections, apply new ones and make
changes before delivering it to the customer
15.11.2010
18
Forms-based authoring
• Forms provide both consistent structure and content
• Highly structured manner, and uniform. E.g.
• Separate fields for first name and last name
• Select a country name from a list (instead of freely typing it)
• Typically the output requirements govern the input requirements of the
forms
• E.g.
E g a customer database requires two fields for first and last name
• Forms-based authoring can be easily transformed into an XML structure
15.11.2010
19
Format-based authoring
Word processors provide functionality for structuring the format of a
document, by using tags (e.g. heading levels, body text, bulleted lists).
However, most people use the “normal”
However
normal style mode throughout the
document
Even those limited tags are typically focused on the rendering aspects (i.e.
h
how
th
the ttextt will
ill appear and
db
be printed),
i t d) and
d th
the fformatt ttags are nott used
d
consistently, so that they could be easily reused in a content
management system.
15.11.2010
20
Structured authoring
More sophisticated and powerful than format based authoring
Languages that can be used:
• SGML (Standardized General Markup Language)
• XML (eXtensible Markup Language)
The author is using
g a tagging
gg g language
g g to identify
y elements of the
document, based on their content rather than their appearance
Based on a well-defined set of descriptive tags
E.g.
<product_name>Cola drink</product_name>
<ingredient>sugar</ingredient >
<ingredient>water</ingredient
g
/ g
>
…
15.11.2010
21
Summary of content acquisition and authoring
• First step in a content-management system
• Content and documents are not the same
• Without labeling the content, you cannot make it reusable for
sophisticated manipulation
manipulation, post processing or rendering
rendering, without
significant extra effort
15.11.2010
22
2. Content-management repository
15.11.2010
23
Content-management repository
The central point of a content-management solution, where data is stored
Typically based on a standard database (e.g. MySQL)
• Database stores either the actual content,
content or pointes to the content on a file
system
Text
XML
Text
Image
PDF
Content Management System
Video
Repository
15.11.2010
24
Advantage of content management system
• Content is more easily accessible to everyone
• Accessible from one virtual location (no need for users to search on
multiple servers, or hard drives)
• Latest version always accessible via a single point of access
• Editorial metadata and tracked workflow process (who updated the
content, when)
• Centralized security process and policies (who can access what)
• Content can be opened to externals (customers and others) via
assembling, linking and rendering
15.11.2010
25
From database to a content management software
Typically, a traditional database is overlaid with management software
which provides a set of features/functions. Most important ones:
• Single point of access, via standard login process
• Access policies (who can access,
access read
read, modify
modify, what)
• Lock process, so that a document cannot be modified at the same time
by multiple users
• Version control (what changed in every version, and by whom)
• Different search and retrieval functions
• Optionally, workflow tracking of documents through a number o users
who need to perform actions on it (e.g. review, approve, etc.)
15.11.2010
26
Remember
A content management system is different than a document management
system
In a content management system the individual elements (information
assets) inside a document are handled as objects in the repository
In a document management system the whole document is stored in the
repository as an object
• When something changes, the whole document needs to re-uploaded
re uploaded
15.11.2010
27
Busting Configuration
In a content-management system (where documents are developed using
structured content units, with tags) content elements are burst apart for
storing them.
15.11.2010
28
Reuse strategy
With individual modules content can be reused in a variety of outputs
• E.g. think the company logo. It is probably reused in every single
document.
• Updating the logo on the content-management
content management repository would
automatically update all the output documents
Translating much easier, as modules need to be translated, and the whole
new document
d
t can b
be automatically
t
ti ll constructed
t t d
There is of course the possibility of specific policies (so that a change does
not p
propagate
p g
automatically
y to all output)
p ) if so desired
• Think a country specific update
15.11.2010
29
Link management
Links can be maintained in a content management repository, so that
relations between objects are maintained
A “virtual
virtual document”
document could simply be a collection of links towards content
objects in the repository.
Links maintained in the repository should be also usable when content is
published
bli h d outside.
t id F
For example,
l when
h creating
ti a CD
CD-ROM
ROM with
ith content,
t t
the content associations should be correctly transferred to this off-line
medium
Dynamic checking of links could be possible, with notification to the authors
• E.g. if a referenced page was deleted
• Helps avoiding the problem link problem on the web
15.11.2010
30
Multiple Repositories
It is a good idea to think that an organization would probably have multiple
repositories
• Working repository
repository, where authors execute their daily job/tasks during the
authoring process
• Publishing repository, a repository that is handling the calls from the
public
bli web
b site
it (focused
(f
d on performance
f
and
d scale).
l ) C
Content
t t wise
i it could
ld
simply have a subset mirror dataset of the working repository
• Staging
g g area,, repository
p
y for testing
g changes,
g , tweaking,
g, etc.,, before
deploying the changes to the working/publishing repositories
• Caching, a memory repository (seating “before” the main repository) that
caches frequently used data
data, so calls do no always reach the slower
database component
15.11.2010
31
Summary of content-management repository
• Component that stores and retrieves content
• Content can be stored as whole documents, or content entities can be
burst apart to individual components for better data re
re-usage
usage
• The centralized authority of the repository can help towards solving the
broken link problem
• Repositories are typically build by applying special content management
software on top of standard databases
15.11.2010
32
3. Assembling and linking content
15.11.2010
33
Assembling and linking content
Having individual content elements stored as separate entries, in the
repository, there is the need for combining them to create compound
documents and linked structures (e.g. a PDF file, or a web site).
Control on how the data is linked can be either at the hands of authors, or
the content consumers
Typically
T
i ll build
b ild lilists
t are used,
d as simple
i l lilists
t off pointers
i t
tto th
the appropriate
i t
content elements that need to be linked (in a similar way as a multi-level
table of contents).
• Those build lists could also include dynamic elements (e.g. scripts)
• For example, different results if a US customer is asking for a document, vs. a
customer from Finland
15.11.2010
34
Linking
• Authors can specify the links between information modules/documents.
• The links are maintained within the content management system, at the
repository
• The actual linking functionality happens at the place where the information is
delivered. For example, at the user’s web browser, when clinking from one
page to another (via a link), or at the publishing system if printing a manual
• The relationships between (e.g. web site) navigation and the presentation
can be established via linking.
• For example, there can be a “tab-based” web page, or a step-by-step click
through guide, or a forum-style bulletin board
• No matter the navigation paradigm, the interconnections are based on links.
• Links are however, not within the information objects, rather they are defined in
a link structure, which is part of the site’s information system
15.11.2010
35
4. Delivering content
15.11.2010
36
Delivering content
Customization is an important aspect when creating/delivering the final
document, in multiple styles, for multiple media.
Content Management System
Repository
Content
Style sheets
Product A
product
web page
User manual
Software
help file
15.11.2010
37
Delivering content
HTML
Content
CD-ROM
Design
Print
Input
Forms
Template
e
Content
Database
WML
Design
Database
PDA
RSSXML
Design
Form
15.11.2010
38
Information Model
15.11.2010
39
Information Model
• The Information Model provides the framework for organizing your
content, in such as way that it can be delivered and reused efficiently
• With a defined information model you can label information so that search
and retrieval is easy and efficient, both for end users and authors
15.11.2010
40
How do you know that a web site does not have an
Information Model?
• You can’t tell how to get to the web page that has the information you
need
• You click a link, where you hope that the information is, but the result
surprises
p
yyou negatively
g
y
• Instead of drilling down the page hierarchy, for getting closer to the
information you need, you end up going further away
• Every time you start over, from the main page, you end up in the wrong
place
• Searching/browsing articles is guided only by their titles
15.11.2010
41
Static Information Models
• University libraries use a quite stable and consistent information model
for storing and organizing their books. Once you get familiar with it, you
know exactly what to find where (with or without the help on an electronic
catalogue system).
• Each book has a code
• You find books where you expect them to be
• Now imagine going to another library (e
(e.g.
g a municipality one) that uses
different way of categorizing books. A different coding scheme… You
have to figure out all over again how the system works.
• For example, the books by your favorite author are placed in a different location
that the books about your favorite author
• You can learn the new system, but it will take time until you feel comfortable
again
• Sites are sometimes designed like this
this, having some specific people in
mind, thus the system of navigating is not natural always for the end
users
15.11.2010
42
Dynamic Information Models
• Now imagine walking into a library, and the books are automatically
reorganized according to your preferred system/logic
• You don’t need to think or worry, as the location of the books is exactly a
yyou thought
g it would be
• And imagine the library reorganizing, on the fly, for every visitors that
enters
• Yep, that could be messy in real life, but possible in the digital world
• With the usage of dynamic information models
15.11.2010
43
Example
Think of a cooking site. Users might want to search for recipes based on
different criteria. Those needs to be identified
For example,
example based on the “primary
primary recipe ingredient”
ingredient , or based on the
“ethnicity”, “role in meal” and finally “special diets”
Those attributes and their values
would
ld b
be translated
l d iinto tags
that should be embedded in
the respective documents
15.11.2010
44
Example
Attribute (dimension)
Value (subcategory)
Primary recipe ingredient
Beef, Lamb, Chicken, Fish,
Shellfish, Vegetables
Ethnicity
Italian, Mexican, Chinese, Irish,
Thai, Vietnamese
Role in a meal
Starters, Soups, Sandwiches,
Salads, Main courses, Side
dishes, Desserts
Special diets
Low fat, Low salt, Low calorie,
Low cholesterol
15.11.2010
45
Example of labels in a document
• Spring rolls recipe
<recipe= “Spring rolls”>
<primary=“Vegetables”>
<ethnicity= Chinese >
<ethnicity=“Chinese”>
<role=“Starter”>
<diet=“Low fat”>
• Canneloni recipe
<recipe= “Canneloni”>
<primary=“Beef”>
<ethnicity=“Italian”>
<role=“Side dish”>
15.11.2010
46
Benefit
• With this tags in place, at each information module, it is possible for each
person searching for a menu to find out what he/she is after
• No matter their conceptual model (starting from the ethnicity, the main
ingredient, etc…)
• Not only users can search based on the core categories, but also make
combinations. Somebody might be looking for low fat chicken recipes…
• Data can be easily reusable
reusable. For example
example, if we would like to publish a
book on Asian cuisine, we could easily extract all the relevant recipes
from the system
• Typically, the information model gives us very concrete model for
designing the repository database of the content management system
15.11.2010
47
A product example
• A users guide for every product model
• But common parts manual parts can be reused
User guide
5800
User guide
N97
User guide
N900
15.11.2010
48
A product example
• If information topics have been labeled by the target languages, then the
company can easily print the information in all languages
Käyttöohje
5800
Handbuch
5800
User guide
5800
15.11.2010
49
A product example
• If information topics have been labeled by different delivery targets
(service manual, user guide, web etc.) information is delivered to the
appropriate channels. Any change to the core topics is reflected to all
targets
Service
guide 5800
User guide
5800
Web site
15.11.2010
50
Dynamic Web sites
Most of the content you find on the web is static, especially sites related to
products.
The content is pre
pre-planned
planned, developed and organized by the authors
authors.
• Same applies to Frequently Asked Questions (FAQ)
• Technical updates
• News releases
However, in content-rich sites, the focus is on current events and data
constantly updated either automatically by feeding it into the system (e.g.
weather information) or the end users themselves (community contributed
content)
t t)
15.11.2010
51
The New York Times
15.11.2010
52
The WSJ
15.11.2010
53
Developing customized content
• Previous examples are based on getting data from a traditional repository
(e.g. financial and enterprise databases) and moving this data to the web,
in a appropriate context
• Dynamic customization allows delivering variations to specific users who
navigate of search the site, which are assembled on the fly.
• D
Dynamic
i customization
t i ti and
d assembly
bl iis mostt appropriate
i t when
h th
there are
more variations than are practical to store on a static site.
• Variations could be based on:
•
•
•
•
Product models
User profiles (remember the context advertising we discussed in Lecture 3)
Local environments, such as country and region
…
15.11.2010
54
Developing customized content
15.11.2010
55
Developing customized content
15.11.2010
56
Tracking customers
Customers returning to Amazon get pages related to the books they
previously purchased
In a clothes shop,
shop you might get the style and size based on your previous
purchases
15.11.2010
57
Tracking customers
f
15.11.2010
58
Tracking customers
Another way of tracking is by following the path the users had on the site
• For example, if somebody went straight to the technical documentation of a
product, that product information might be displayed to him next time
Of course, really knowing if the user found what he was really after is
difficult. Some times it is OK to ask…
15.11.2010
59
Tracking customers
15.11.2010
60
Content Management Systems for the Web
Good news!
In many cases you don’t
don t really have to worry for all those lower level
content management system repositories, keyword indexing etc.
There are many existing systems that take care of those issues, and you
can simply get one of them, install it and configure it for your website.
15.11.2010
61
Example: Plone
•
•
•
•
•
A content
t t managementt system
t
for
f the
th web
b
Open Source
Supported by a large number of community members
Easy to use
Customizable and extendable
15.11.2010
62
Plone
15.11.2010
63
Key features
• No need for knowing HTML. You can create and edit pages based on a
WYSIWYG editor
• Collaboration of multiple people on the site
• Enables interaction between the users and the site
15.11.2010
64
A lot of add-ons
15.11.2010
65
A Plone site
15.11.2010
66
Content editing options (once you login)
15.11.2010
67
Simple content editing (using Kupu editor)
• Style based formatting
• Add media (photos, etc.)
• Copy-paste text
• Automatic
u o a c ccreation
ea o o
of
table of contents
15.11.2010
68
Working copy
Case: you have published a page, and you want to update it, but keeping
the old version on the web site until you have finalized and published the
new one.
You also want the new p
page
g to replace
p
the current one,, but still yyou would
like to keep the history of the old one (just in case).
It is possible with “Working copy”
15.11.2010
69
Working copy
15.11.2010
70
Working copy
If someone else tries to edit the same document
15.11.2010
71
Working copy
When you are done, simply check-in
15.11.2010
72
Version control
• Track changes
• Revert to older version
15.11.2010
73
Multi-language sites (with LinguaPlone add on)
• Authors can easily add the
translated content
• Transparent for any layout
used
15.11.2010
74
Side-by-side translating (with LinguaPlone add on)
15.11.2010
75
Dynamic forms
• Easily add forms to a
web site
• Export data to
• e-mail
• Spreadsheet
15.11.2010
76
Dynamic forms
• Easy design
• Re-arrange fields by
dragging and dropping
15.11.2010
77
Events
Add calendar events
No need to program/code
15.11.2010
78
Events
Add calendar events
No need to program/code
15.11.2010
79
Dynamic content with ATSuccessStory add-on
• Easily create a space where
random stories/links to other
documents appear
• Feature photos, text of some
success stories
15.11.2010
80
Live Search
• All documents
d
t are indexed
i d
d and
d searchable,
h bl in
i a live
li manner (AJAX)
• Search results can appear in a drop-down list, as you type
• PDF
PDF, Word
Word, Excel etc
etc. files
can also be indexed
15.11.2010
81
Search results as a web feed
• Every search result set can be represented by a web feed (eg. RSS)
• Clients can subscribe to it to follow any changes…
15.11.2010
82
Search results as a web feed
• With any standard feed reader, the user will be notified if new results
appear in the future…
15.11.2010
83
Administrating a Plone site
15.11.2010
84
Administrating a Plone site
15.11.2010
85
Administrating a Plone site
15.11.2010
86
Administrating a Plone site
15.11.2010
87
More Plone reading
Plone quick startup guide video
• http://ia331414.us.archive.org/3/items/SeanKellyIntroducingPlone/ploneintro.mov
Guides, tutorials, etc:
• http://plone.org/documentation
15.11.2010
88
Other content management systems
15.11.2010
89
Blogs
A blog is a type of website or part of a website. Blogs are usually
maintained by an individual with regular entries of commentary,
descriptions
desc
pt o s o
of e
events,
e ts, o
or ot
other
e material
ate a suc
such as g
graphics
ap cs o
or video.
deo
Entries are commonly displayed in reverse-chronological order. "Blog"
can also be used as a verb, meaning to maintain or add content to a blog.
(wikipedia)
Blogs are typically created with blogging platforms (ie. specialized content
management systems)
15.11.2010
90
Example: Blogger
Blogger by Google
•
•
•
•
•
Free
Comments & access control
Templates, custom colors
Photo uploading support
Posting from mobile phone (via email)
15.11.2010
91
Example: Blogger Tutorial
A step-by-step guide:
http://danielcraig.wikispaces.com/file/view/Blogger+Tutorial.pdf
15.11.2010
92
References
• JoAnn T. Hackos, "Content Management for Dynamic Web Delivery", Wiley; 1st
edition (February 14, 2002).
http://www.amazon.com/Content-Management-Dynamic-WebDelivery/dp/0471085863
• http://plone.org/documentation/manual/plone-3-user-manual/managingcontent/working-copy-3-0
• Christopher Johnson, "if PEOPLE: Plone CMS Demo",
http://www.slideshare.net/ifPeople/plone-content-management-system-demopresentation
• Christian Vinten-Johansen, "Content Management Systems", Penn State Web
Conference 2003 http://its.psu.edu/webconference/Web2003/Web2003Materials/CmsJohansen.ppt
15.11.2010