Text Seeker (Tzeeker) Project Paper

Audience and Purpose

Text Seeker is a new text analysis environment as an expansion of TAPoR; the online Text Analysis Portal for Research. It is a “one field does all” site in which users can analyze text in a variety of different forms with a variety of analytical tools provided. Depending on the type of text entered, within seconds the site’s intelligence will select a suitable set of analytic tools, run the text through these applications, display results and suggest further tools that could also potentially be useful. The Text Seeker project will ultimately result in the construction of a simplified text analysis resource for both new and experienced users but without jeopardizing the effectiveness it possesses. The reason this project exists is because the TAPoR Portal and the TAPoRware tools can be confusing to users. Text Seeker will follow the same TA principles as TAPoR, but at the same time will provide a ready and available resource for users of any level of experience. It will minimize initial interaction and instead determine appropriate TA actions to take based on the text entered. The objective of the Text Seeker project is to provide a smart text analysis site that incorporates a variety of TA tools and eliminates initial confusion surrounding the site’s existence and capabilities.

Background Research

In order to design a site that is fit for the user and not just the site development team, it is important to come up with a set of personas and scenarios that match the profiles of users who will most likely be reaching the site. By creating personas for a developing website, a site designer is able to convey to the development team and client the type of person expected to use the site. Personas typically have a fictional name and a description of characteristics this user may possess. We have created three personas that fit the site profile. The first and most important persona we invented was a University undergraduate student in English who has never used or even heard of Text Seeker before. From this, we have been able to construct wireframes that fit this user’s character. The same has been done for a graduate student in Communication Studies, a biology professor, and a newspaper columnist.

The Text Seeker project and its own unique TA capabilities were heavily influenced by the TAPoR Portal itself. The TAPoR Portal is an online environment in which users can manage sets of texts either uploaded from a PC or on the web. Users can learn about and run many of the same tools that Text Seeker uses on these texts. One disadvantage the Portal possesses is that it has a fairly complex interface. For new users, it is difficult to grasp what the tool is capable of without exploring and researching different terms, tools, and functions. Text Seeker has taken this into consideration and learned from these flaws by designing a fully functional, easy-to-use interface. The Portal has some inventive and smart functions for text analysis, but the site is found by many to be cluttered and overwhelming with respect to the number of options available immediately on the entry screen. This is where Text Seeker will excel in conjunction with its technical capabilities. Text Seeker will provide all these functions and more but without the complications that exist in navigating through the TAPoR Portal.

The TAPoR Tools site was also a resource that was explored before the latter stages of the project had begun. TAPoR Tools is a database of all the TAPoR TA tools that exist for users to learn about and test with sample texts. It has the capability of performing functions like making text comparisons to verify that the authorship is or is not the same. By doing some background research on the various tools, I was able to alleviate some confusion I had concerning the potentials TAPoR and ultimately Text Seeker had; a fundamental understanding I needed in order to tackle the design for this project. I was able to experiment with a number of the tools which helped me construct possible personas and scenarios for Text Seeker.

Shawn Day’s TAPoR Portal Recipes also accounted for a considerable part of my research for this project. To be able to read a variety of scenarios for an already well established text analysis site proved to be very useful. These recipes explained the capabilities and true potential of the full set of TAPoR tools and how they could be used in different textual scenarios.

I have also explored other text analysis tools that can be found online including HyperPo and Textalyser. HyperPo, developed by Professor Stéfan Sinclair, is a digital text reading environment that instantly provides a statistical summary of a user’s text. Although it is a preliminary draft of a soon to be completed version, it has some problems with interface design. Hyperpo is not particularly aesthetically pleasing but does possess an amiable simplistic quality to it. We wanted to mock the lack of clutter found on the HyperPo site and apply this to Text Seeker. Textalyser is a useful quantitative TA tool, but again obviously lacks attention to interface details. This may although be affected by the site’s lack of funding. It seems that there has yet to be a TA environment hosted on the web that prioritized design in the site’s developmental stages. Typically online TA environments are also either hit or miss with respect to layout. By evaluating strengths and weaknesses of these sites both in terms of layout and design, I have been able to learn from their mistakes and increase the number of possibilities that exist with Text Seeker.

Implementation

Text Seeker will be hosted on the web when it is complete. The programming itself will not be implemented by me but by two other members of the TAPoR group. The portion that is my thesis project will be expressed in the form of the final website, my blog posts, and the wiki which will eventually incorporate an explanation of the site for users along with its capabilities.

This project is based on a set of high-profile research tools used by academics all over the world. I am responsible for doing site design for a site that could potentially become more successful than the TAPoR Portal itself. As with any client-based project, it is of course difficult to please the client 100%. Due to the fact that I have been meeting with my team members (who also happen to be my clients) every week, there is a lot of opportunity for them to express dissatisfactions with my work or suggestions to improve my contributions to the site. This has also kept me focused and on task. On numerous occasions I have presented ideas and have had to go back and rework them after receiving feedback from the TAPoR group. This has been extremely beneficial with respect to tweaking fine details which in turn make for a stronger interface. It has also encouraged me to keep the Text Seeker project a top priority this semester. The Text Seeker group is also small enough that we have become quite the tight-knit team. This has allowed me to gain a sense of comfort with the group, enough to express any thoughts or concerns I have at any point in time throughout the course of the project. The client relationship has left a positive tone on the project even proving to be beneficial to my work ethic. As a means of managing my tasks and presenting my content I have also been recording notes on progress of the project in the wiki, and constantly updating a blog that documents my personal thoughts on the project, and any problems that arise during each stage. This has been very helpful with respect to communicating ideas between the client/team members and myself.

Typically for a project of this caliber it is beneficial to conduct testing or user surveys before any significant interface decisions are made. For Text Seeker, we decided collectively as a group that any sort of user testing would have been too time consuming to conduct in the early stages of the project. As a means of replacing any extensive user analysis, we worked out potential user personas and scenarios based on the user base of other TAPoR sites which proved to be very valuable to our interface design.

Artistic and Design

In terms of interface design, the wireframes have played a pivotal role in this project. The steps taken that lead up to the current state of the wireframes were time-consuming and tedious but absolutely necessary. The three personas were created for users to ensure the site profile and design was fit for them. These were presented to the Text Seeker team in segments before their final due date. As personas were brought to the table, they were critiqued and re-worked. For each persona, it was necessary to write at least one potential scenario in which the user would enter the www.textseeker.com and navigate through it. These were constructed and presented to the Text Seeker team along with the corresponding personas. Again, they were assessed and re-worked until complete. The personas and scenarios have opened the gate to the graphic design stage of the project.

The graphic design portion of this project is underway, but still in progress. The whole philosophy of Text Seeker is to create a simple and clean text analysis resource that eliminates any initial confusion surrounding the site’s existence and capabilities. In the early stages of the project I had already begun to brainstorm design elements and had even incorporated some of them into the mock-ups of the entry screen and the result’s page. In the graphic design stage of the Text Seeker project I want the final product to reflect this philosophy of simplicity and functionality. By creating a clean design free of clutter, the site will accordingly become more user-friendly and comprehensible to users. I want to use white, black, and a soft pastel colour, still to be decided, for the graphic design. I would like to have all corners slightly rounded to create a comfortable atmosphere free of harsh angles for both amateur and experienced users. This will keep the user’s eye focused more on the content, the most important element, but while still maintaining an aesthetically attractive page.

Some of my influences for design are the Apple site and Google. Both these sites have a neat interface and use black and white heavily in their colour palette. This allows for a clean design that leaves a high level of importance to the content, but still with an aesthetic that is eye-catching and not overdone. This is a crucial design mentality with a site such as Text Seeker as the content is of principal importance. Google was an influence on not just the graphic design but also the usability of Text Seeker. As when you enter www.google.ca, we want there to be only options available that are absolutely necessary in terms of navigation. A choice between two primary buttons on the entry screen emphasizes the importance of a simplified interface. As there are really only two options to choose from in order to advance to a new page, the user will feel a sense of comfort rather than being overwhelmed with choices. A set of six buttons, three along the left-hand-side and three along the right appear to keep the page design symmetrical and tidy. These are options a more experienced user can experiment with.

Conclusion

Text Seeker is a text analysis environment that is unlike any other. It is heavily influenced by TAPoR’s Portal and TAPoRware Tools, even using certain elements from both, but is unique with respect to interface design. This site has been built with the user at absolute heart and a great deal of layout design care. The usability and simplicity of the final product will attest to this. The Text Seeker Team has full confidence that www.textseeker.com will combine the functionality of a modern interface, the charm of a simple design, and the intelligence of a well programmed data engine; a truly original and unparalleled combination.

-- BradKarelson - 04 Mar 2008