Methodology
What tools did we use?
This website single-handedly allows users to access the over 826 billion web pages saved over time (1996-Present). It links to the expansive archive of the Internet Archive and is defined by its free public access.
The Wayback Machine is immensely useful in how it can display a web page as it would have looked at the time. Below, you can see the first capture ever of the Indymedia Argentina website, from April 5th 2001.
Excel was used to document and organize all of the data collected. Two methods of data collection were utilized, needing two separate Excel sheets. The first method of data collection required the researcher to manually go through each page of Indymedia Argentina archived on the Wayback Machine, searching for the gender-neutral x. For this Excel sheet, we separated the different forms of data into different pages: article posts, photos, and comments. We made separate columns for key pieces of information regarding each use of the gender-neutral x including the post’s title, author, date, and keywords used in the article. We also included the link in order to trace back to the article if needed.
The other method of data collection involved using Voyant, a text analysis tool, to find any use of the gender-neutral x. This process was a lot more organized and methodical, making the process of recording data a lot simpler. For this Excel sheet, we made separate pages for year assignments, overall findings, and each of the years analyzed. Below is an example of how we documented uses of the gender-neutral x. By organizing our data by month and year, we were able to quantify the data and visualize the usage of the gender-neutral x.
The Windows Command Terminal helps users communicate directly with their computer system. Once we had a list of the HTML links for each page of Indymedia, we instructed our system to download the HTML file for each page by referencing the list. By typing in the commands depicted below, we downloaded each HTML file to our desktops and from there were able to organize them by months and years.
Sublime Text was used to read through the HTML code on the Indymedia Argentina website. We could see how the web page was constructed, revealing details that are not apparent in the site’s public display. Additionally, after the web pages were downloaded, but before they were uploaded to Voyant for analysis, Sublime was used to shorten the file to mainly just the text, so the research and text analysis in the next step was concentrated on the articles.
Voyant is a text analysis website that allowed researchers to search for any use of the gender-neutral x. Since the gender-neutral x is usually used in words with an “*s” ending, we searched for the gender-neutral x by searching “*xs”. Since our HTML files were already compressed into zip file by month, we were able to easily transfer the findings on Voyant onto our Excel sheet.
Voyant is a text analysis website that allowed researchers to search for any use of the gender-neutral x. Since the gender-neutral x is usually used in words with an “*s” ending, we searched for the gender-neutral x by searching “*xs”. Since our HTML files were already compressed into zip file by month, we were able to easily transfer the findings on Voyant onto our Excel sheet.
Aside from its text analysis tools, Voyant also has a variety of visualization tools (graphs) that help users visualize the frequency of terms in the text.
In order to compile all of our research into a presentable format, we used WordPress to create the website you are viewing right now. We went through the university to obtain a specific URL, in the hopes that our website will be archived and preserved long into the future.
Accessing and Interpreting Data
Discuss how the books The History of Abundance and Data Feminism guided and challenged your work. What questions emerged from reading their books?
Problems Encountered and Solutions
Through our research, we encountered a variety of problems, ranging from discrepancies in data collection to lack of knowledge of how the internet worked in the early 2000s. These issues mainly stemmed from working with a site that is over twenty years old. However, we were able to work through major problems with the help of outside resources, as well as Professor Mackenzie Brooks, and ensure that all data is as accurate as possible.
One of the main issues was the lack of technological knowledge that came with managing a website in the early 2000s. Given that at the time, the internet was less than a decade old and sparsely available in Argentina, we can see that users of Indymedia were still experimenting with the process of publishing their own information online. Some of the issues that stemmed from this include lack of publication author or date, and timeliness of publications. While researching, this made pinpointing uses of the gender-neutral x with an exact date very difficult, and in some cases, impossible. However, by looking into the code, web address, and even surrounding articles, we were able to pinpoint most dates at the very least to the month.
The main problem we encountered was how we were going to search for the gender-neutral “x”, when archived versions of the website did not have a functioning search button. Our two methods of searching divided our project into two parts. Our first solution was to manually flip through every publication and use the command function to search for the “x”. Although this resulted in thorough data, this process was very tedious and was only done with the archived version of Indymedia that was available through the Wayback Machine. Our second solution involved downloading every page published in the archived version of Indymedia and processing everything through a text analysis tool to search for the gender-neutral “x”. In the text analysis tool, we used the asterisk before typing in “xs” to ensure that all words that used that conjunction would appear in our search results.
Another major issue encountered was discrepancies within the archiving of Indymedia. During our research, we read Ian Milligan’s book, History in the Age of Abundance. This book touched on the creation and rise of the internet in the 90s, as well as the process behind digital archiving. Not only did this book give us a new perspective on how the internet was used in its earlier years, but it also brought to light a problem that would have gone unnoticed otherwise. Every day, websites on the internet disappear, leaving all the information they held to disappear as well. Digital archives utilize web crawlers to capture these websites before they are taken down, ensuring that this information does not cease to exist. However, due to the imperfect process that web crawlers use, not every website or page is able to be captured, leaving lots of information to disappear from the internet. This was an issue we encountered with Indymedia, as the archived versions of the website would sometimes lead to dead ends that had not been archived. The original version of Indymedia also used some plug-ins for comment sections that no longer exist. All in all, our research was limited to the extents that web crawlers had archived the website.
Once all data was collected, we needed to find a way to organize all of our findings in detail, and so they could be retraced if needed. To do this, we separated our findings into two Excel sheets for the two methods of searching we used. From there we separated data by date and content, as well as other defining characteristics like author and category. This process is described in greater detail above. Organizing data was crucial, especially since retracing steps using the Wayback Machine proved to be a challenge.