Information architecture model

Recently busy graduate design, the need for document translation. I thought that before the translation of a poor, but always feel not rigorous enough. My graduation project is mainly to help the teacher's institute to be a website, already have a website, but the teacher and I think the old version of the site is very old. So volunteered.

The site needs to be refactored, and the Natural Information Architecture section is an important part of it, though it has also been seen quite a bit before, such as "thinking of the OR search in the kitchen" is a sign of thought, but systematically translating the Five Simple Steps team One of the chapters in this book, "A Practical Guide to Information Architecture," is "IA Patterns." (For this book, there are PDF documents available online, which can not be posted because of personal cleanliness, you can leave a message if you want.

--Start--

This chapter is all about patterns of information architecture - some of the common approaches that are primarily needed for the information architecture of different types of websites. Hopefully, you can start with these patterns as you build the information architecture for your site. Of course, the premise is your website to meet one of these models, or a combination of them.

What is the pattern?

A pattern is a common solution to a type of frequently encountered design problem.

The idea of ​​mode comes from architecture [1]. As you can imagine, there are many common problems waiting to be addressed in buildings - the relationship between food preparation areas and food service areas, how people enter a building and how interior interiors work in harmony with the outdoors.

Patterns in the field of software development is also very common (some problems repeatedly), to provide some software design solutions to developers. The same is true of interactive design, with some reusable solutions surfaced, and common needs such as login forms and page breaks.

The key point in the model is that all of these ideas come from real-world needs. You do not have to sit and meditate to create a pattern for a whole new world. The patterns we discuss below are all the crystallization of the wisdom of the "working people."

Information architecture model

Information architecture is related to many areas that people need to use information. Below I will explain the characteristics of each mode, the use of scenarios, and provide relevant cases.

First I'll discuss four simple patterns (Hierarchy, Database, Hypertext and Linear) and a combination of the three.

Simple mode level

In the concept of hierarchy, the relationship between categories is a parent-child relationship or generalized and narrowed - that is, extracted as a broader group or broken down into more specific groups.

Hierarchies can be described as flat and conical:

Flat hierarchy features are: there are many categories of top-level, but fewer levels; cone hierarchical structure is characterized by: the top category less, but a large number of levels;

A hierarchical structure can also be described as strict and polyhierarchy:

In a strict hierarchy, a category can only be in one place; in a multivariate hierarchy, a category can be placed in more than one place;

In the real world, a strict hierarchy is a must - after all, it is impossible for an entity to be in a different place at a time. However, in the digital world, it is easy for us to place one thing in many places, and it is a good solution to the chaos in the reality. We can put things in as many places as we want to see and allow overlapping of categories of boundaries.

Hierarchy is the easiest and most common way to organize information and is suitable for a wide range of content. Especially for small sites, just a few simple layers - the top (homepage), some second level and the bottom detail page.

Similarly, the hierarchy also applies to large sites. Especially those content type website (the content is various and varied). Even if your information complexity varies, the hierarchy also has a role. For example, you can first show review information and then allow users to break down more details as needed.

3 times the "magic spell"

For years, there has always been a "legend": every specific content should enter less than three clicks from the homepage.

For many websites, this is simply fantasy. By doing so, many categories of sites at each level will be more people no choice.

On the contrary, it is more important to allow users to easily make decisions at every level and to inform them that they are on the right path. If users feel that their actions are correct, then they will follow their own intention to browse learning.

Figure 16-3: CHISIG (Computer-Human Interaction Special Interest Group) is a small organization with a small amount of information. Simple hierarchy is very effective.

Figure 16-4. There are many elements of the White House official website, but the basic hierarchy is still adopted. (Whitehouse.gov)

database

The database discussed here is not a technical object - the digital storage of various kinds of information, but a conceptual model.

They all share the same structure or model with a specific plan, all information must be adapted to these structures. So you can not impose something that is not suitable for this model.

The database schema is aimed at content that has a consistent structure. Some content may not be directly related to others - they do not really have the necessary parent-child relationship in the hierarchy - but they all consist of the same pieces of content and have the same structure.

For example, etsy website, these two categories do not have anything in common:

Picture 16-7. Sleepy time (from esty.com)

Image 16-8. Headbands (from esty.com)

But they have the same content block to make up the list:

Title Description Tags Material Location Payment Method Creation Date Photo Category Color

That is, each product on etsy is the same structure.

Database schema application scenario is either a small set of information, or a very large data set.

The biggest advantage of a database structure is that data is stored once, and then different pieces of data and methods can be used to present the information. For example, in etsy, you can look for content by category, color, location, and even recent updates (very small levels). This gives users many ways to discover the information they are interested in.

Database structure for music, product catalogs, books, articles, blogs and so on with a consistent structure of the content. It provides users with a lot of access to the content. For example, on flickr (flickr.com) you can discover photos by photographers, album sets, groups, favorites and labels.

Metadata

Metadata is usually defined as "data about data" or "information about information." Although this definition is not very helpful, but very accurate. Metadata is all about the content of the block and describes the content block information.

There are three different types of metadata:

Intrinsic: What is the object actually Administrative: How it is used Descriptive: Description of the category

For example, UX Australia blog, metadata may include:

Type: Blog Post Article (Inherent Data) Author (Administrative) Post Date (Administrative) URL Status: Published (Administrative) Title (Descriptive) Classification (Descriptive) Label (Descriptive)

These metadata are mainly used for two things:

Generate a specific content list (for example, show everything under the 'Announcements' category) Select what to display on the page (for example, include title, author, description, category).

Of course, the most difficult part of the project is deciding what types of content to collect and what their content is (or which categories you will use).

Hyperlink

Hyperlink mode is a very interesting part of the information architecture because it seems to be an anti-structure mode. Content blocks connect based on only one another. Unlike hierarchies or databases, there is no main structure - content is simply linked by links.

The best example of a hyperlink structure is wiki. Wikis do not have a preplanned structure - the content is only linked through embedded links.

Wikipedia is the best annotation. The content of Wikipedia does not have the main structure, there is no very robust database structure. (That's right, he's stored in a database and has some basic chunks of data, such as headers, descriptions, etc. But it's not the "database structure" described above.) Each page is independent and can be accessed through related links Connect with other pages.

Hyperlink structure is particularly useful if the content has been created for some time, but you are still not sure what to build. In this case, it is not possible to predetermine the detailed architecture or even to determine the basic pattern for the website. This is how many document projects start - people first write separate document pages and then are busy creating document relationships through links.

Many websites are hyperlinked structures at the outset, and are refactored when the content is established.

The main problem with the hyperlink structure is that the success of this model depends very much on the connections that people create between the pieces of content. You can clearly see the next level for the hierarchy; for the database schema you can show everything for a particular category; however, the hyperlink structure itself does not have the ability to automatically show related content. If the author does not create a link, the user has a fundamental way to find the information.

Linear

Linear mode, as its name implies, follows one another according to a straight line rule.

Linear patterns are not common - this pattern is often used to allow users to jump in the way they understand.

If you are confronted with a situation where you have to understand one thing before moving on to another, then it is well suited to use linear patterns - usually, for example, teaching materials. If the user does not actually need to read according to certain rules, do not use the linear mode, or the user will have a sense of frustration.

[Translator's Note]

Apple's official "Start Developing iOS Apps Today" document is the use of linear mode, apparently iOS learning is a gradual linear process.

Mixed mode

Now let's see how to combine these three simple patterns to create a more complex information architecture. There will be overlap between them, so do not tangle with what kind of patterns your site belongs to.

"Simple Level + Simple Database" mode

A common pattern is the combination of a simple hierarchy and multiple database-based content.

This model is very applicable. You can create the site's hierarchy for the underlying content and then leverage the powerful database model to integrate specific information with that part. Of course, it can be any other combination.

Case Study: UX Australia

Figure 16-17. Conferencing Website - Some content pages are arranged in a small hierarchy.

Figure 16-18. The presentation part uses a database structure - this index page shows the title, speaker, brief overview, AZ order. Each link points to a detail page of the exhibit. (Www.uxaustralia.com.au/conference-2009/program/presentations)

One of the major challenges with this model is deciding which part to translate into structural content and which part to leave as a hierarchical content. May have the following consideration:

Do you want to reuse something in another part of the site? If you do not have the need for second use, then you do not have to worry about over-organization issues - completely trivial. Database structure can help you manage a large number of information sets. If you have a lot of news story material in a year, you can manage it as a hierarchical content. If there are hundreds of items a day, you may have to use the database structure for automatic presentation. table of Contents

The second commonly used than the directory mode.

This structure is in fact a database schema, but it is mentioned here specifically because it is very common, especially in the area of ​​e-commerce. The bottom is content, and thus the three top levels depend on the site size and content type.

Jared Spool has written an article that explores this pattern in depth [2] and describes three different types of home and content pages:

Gallery pages: provide direct entry of content pages; category pages: provide display page entries; Store pages: provide category page entries;

Should be changed. Needless to say, these three products may be used in large product catalogs.

Jared stressed that the display page is the most difficult part of the information architecture design, because the user will be based on the effectiveness of this page to decide whether to click Browse page.

Radiation Center (Star)

The central radiation model in the final analysis is also a kind of hierarchical model. However, I want to elaborate separately here because people are still slightly different from the hierarchical model.

In a hierarchical structure, people tend to start from the top (homepage), browse content down one level, often hovering over a branch under the hierarchy. In center-radiant structures, people go from one level to another and then go back to the starting point (center point) and then go back to the other detail pages. For example, a personal page is a focal point for LinkedIn - it's where you often hang around.

Subsites

I have participated in many large sites - government websites, university education websites and more. One of the most common patterns I use, calling myself a "sub-site" (which people used to call "portals" for a while, disappeared later).

Obviously, the entire site is composed of a series of sub-stations, and through the home page or more top-level pages connected. Sub-stations can be applied to any mode, do not have to be limited to the same kind.

In some scenarios, sub-sites are using a unified navigation and page layout, a subtle expression of a point of view: These sub-site is part of the brand website. Of course, the navigation and page layouts of some practical solutions tend to be slightly different in content and user experience, but the overall effect will still be demonstrated through some details.

This model is particularly suitable for large organizations - often with many functional departments or many sub-brands - but there is a need to present it as a whole. In addition, after the content analysis, if you can not find a single model for each part of the information architecture, then you should consider "substation mode" instead of forcing a certain pattern.

As you can imagine, a college website is a good example - the university as a whole represents an organization and brand, but its content is diverse, with each college and organization having specific communication needs (including objects, policies, etc.) . The same is true of government agencies. For example, ABC (Australian Broadcasting Corporation) owns many television stations, radio stations, news and other Internet services.

Figure 16-28. For ABC, the home page hosts many sub-site entries.

Focused entry points

Many large sites, it is impossible to organize content in a single way will be able to meet all users - the so-called outrageous.

Faced with this situation, I usually use the "centralized entry point" mode. First of all, I will be based on content and core users to determine a suitable information schema for the site (the majority of the hierarchy).

Then, assuming some users do not look for information through the main info structure, I'd be interested in providing some entry points to help them find useful information. These entry points do not have to cover all the site content - focus on the core information can be.

I have said that some taxonomies are hard to use - especially for user and task combinations. I found that the entry point solved the problem well. For example, you first use the underlying hierarchy to organize your site content and then provide reasonable access to different users or tasks.

Case - Australian water

The Australian government's Water website contains many government policies and related syllabuses. Of course, policies and programs are not always obvious - some people may know (such as government officials and media workers), but many are at a loss. Both types of people need to find the information they care about.

This site is a typical case of a "centralized entry point" model. The main information architecture is a simple hierarchy organized around policies and outlines:

Picture 16-34. Water policies and programs, grouped together as a category.

Topics such as desalination, rainwater tanks, and saving water provide topics of interest to users. Entrance:

Figure 16-35. Topic Entry Page - links throughout the site

Tagged

Tag mode in either the underlying database model or hyperlink mode has been used. Each category in the site uses keywords as labels - and they are also the entrance to content.

The label may be added by the original author or may be another authoritative reader (for example, the same team member).

This model applies to a large number of different content portfolio, especially content readers have very different ideas and ideas.

When the user feels unable to start, the label can help users explore and find relevant information. For example, Flickr lists all the tags, and you can find all the related photos based on a specific tag - both photographers and visitors have the right to tag.

Applicable Content Content Available Content Challenges and Topic Hierarchy Site with a Variety of Content Site Habits Read the Overview Information First and then Detail Balance Breadth and Depth of Content Database Content Consistency Want to Access Content Through More Means All Content Needs Adaptation to Structure and Do not collect metadata that exceeds requirements Meta content of hyperlinks Follow links to related material The author needs to know the content of the link;
When the content is complete, it may need to be refactored; the linearly ordered content users want to understand something in a specific order; only use simple hierarchies when the users have to read in sequence + the database composite content plus the content type that has a consistent structure What content needs to be structured and what does not need a directory A large number of structural content sets to find a specific category, and then follow the example of a specific product center Radiation rating content Users want to go back to the center page each time, and then check the new content sub-station large enterprises and government Sites, many separate content sections are required Consider whether the substations need uniform navigation, page layouts, and brand hubs, but usually tier users would like to browse the web with no best way to label large content sets Excavating with their own definition information;
Easily find relevant information; who has the authority to carry out labeling operations

[1] The architectural work on the most famous model is Christopher Alexander's The Timeless Way of Building (1979).

[2] Table of Contents For more information, see Jared Spool's article: The 8 Types of Navigation Pages and Galleries: The Hardest Working Page on Your Site. Alternatively, reference may be made to Web Anatomy: Interaction Design Frameworks that Work, by Jared and Robert Hoekman Jr.

Source Address: http://www.hoowolf.net/20....../ia-patterns/