Responding to Organizing Information.

Morville and Rosenfeld reveal the types of exact and ambiguous schemes in their book Information Architecture for the Web and Beyond (4th ed.). Here I will outline the varrying types of both kinds of schemes in organizing information that make searching for certain information easier for users.

The first scheme is an exact one, and it’s Alphabetical schemes. This is how a dictionary works where the user knows that finding a word starting with “a” will be at the beginning and a word starting with “z” will be at the end. This scheme also helps us understand how some schemes can be an “umbrella” for more schemes. For example in a phonebook names are listed alphabetical but also by last name, which is a second scheme that fits beneath the alphabetical scheme of the book.

The next exact type of scheme is Chronological schemes. This is a scheme that goes off of date data and aligns information based on a document’s date. This is how the University of Washington library databases store and show newspaper articles where newest ones can be seen first, and as the user scrolls, the articles get older in aligned date degredation. This can also be done in the reverse order and still be considered chronological scheming.

The final other type of exact scheme is Geographical schemes (however these 3 listed aren’t the only types of exact schemes, but definitely some of the most prominent and common). Geographical scheming is where certain information such as weather is accessible based on what location that weather is at, so when a user accesses their mobile map they gain access to that information using the scheme that is based on their location.

The other types of schemes can be ambiguous such as Topical organization schemes. This is exactly what it sounds like where groups of information are organized based on their topic. An example of this would be a newspaper where all the info on sports is in a separate location by comparisson to the economics section, so unlike if it were alphabetical, “football” isn’t inherently behind “market-crash” simply because the “f” in football is behind the “m” in market-crash alphabetically, in reality they could be anywhere in the newspaper in relation to each other depending on where the newspaper puts the topics.

Task-oriented schemes are ambiguous in that information and content is stored in tasks that a user of the desktop or app, in which task orientation is commonly found, would need to use in succession or alongside using the app. An example of this would be the tool bar in Photoshop or Word.

Audience-specific schemes is where a site or app’s creators know they have a varied user-base and so they create a page that allows the differing users to choose which themes of content they want to see. This can be seen on Pinterest where some users might not want to see knitting pictures so they don’t choose that theme to be shown, but a pro-knitter will choose to see that content in excess.

Metaphor-driven schemes is where common knowledge user’s have from interacting in their daily lives is used to convey the usage or idea of something in a scheme. A key example of this is the filing system on desktops where the reason for us knowing the image of a file means I can drop more stuff into that icon and it will be stored exactly like the real world usage of what that icon displays.

Hybrid schemes usually are shown as a confusing mix of differing schemes which tend to create no mental model for users. However clear communication can be used to designers to get the ideas across and use differing schemes to the best of a site’s abilities. This is seen on drop down menus that let you choose categories or alphabetical filtering.

Changing the focus from Morville and Rosenfeld’s ideas of schemes we turn to look at the article “How Netflix Reverse-Engineered Hollywood” who talks about the microgenres they used and not use user tags to youtube or flickr.

The reason fo creating a database of reference keywords for movies that goes in the 70,000+ range is because it accomplishes the main goal Netflix has as a super company now; retaining memberships. Studies showed that “Members connect with these [genre] rows so well that we measure an increase in member retention by placing the most tailored rows higher on the page instead of lower,”. In simple it’s worth paying viewers to meticulously watch shows and movies and rate them with common words to help users later identify with the platform and it’s content when searching for content.

They don’t let users tag the movies themselves because they don’t need to now that they has testers do that already which hides the wall of work behind the massive genre list. They also don’t do it because it would allow a disruption in what users want, without user input in the Netflix app the general expectations of what people want to watch can be inferred by the data collected from testers which tells Netflix Originals what kinds of shows they should make based of searched for genres. This could be disrupted by constant massive open user input vs controlled testing input. Alongside this their AI is able to use basic tags testers put in for movies and then calculate more tags out of those which heightens and calculates the tagging process rather then keeping the data constantly base level.

Related posts:

You fight with yourself — you want what’s best for you, or you want to try something new, but your mind does not; it wants comfort and familiarity instead. You think you could’ve done more, but…

Did you know every one of us has an entire biological system based around interacting with cannabis? Well, we do, welcome to the Endocannabinoid system. The enzymes help break down endocannabinoids…