2/7/2019 Project Update

[brenda275]

During our first meeting we finally began formulating a strategy to prepare our corpus. We decided we would only want to focus on memes relevant today. That being said, we mostly likely will not include memes from the Baby Boomer group. We're still not quite certain on what features we will focus on, but at least we now know that we will be transcribing texts from images, gifs, and tweets. Since we are still in the process of trying to limit our corpus we would like to be opened minded before defining what a meme is or isn't. Unfortunately our first meeting was before we were assigned a TA, so while we started brainstorming about our corpus we were still unsure on how we wanted to upload our files to GitHub. Our goal for the next meeting is to organize our collected data into one space so we can begin narrowing down our focus and point out any common linguistic features that are left out in memes.

[engeljonathan98]

Can I ask how you determine "what is relevant today", considering how rapidly memes change? Maybe one approach would be to determine what is enduringly popular and still popular now. What I mean by that is: what memes are popular now and have been for a long enough time that we can say they have permeated the public consciousness in a deep sense? That way, you can narrow your scope and eliminate memes that are only brief trends, focusing instead on memes that have lasting impact on people and the way they express themselves.

[emmamamula]

Although you have not clearly defined what a meme is, there is almost an infinite amount of them related to various topics and subjects. Since your corpus is still being prepared, it might be helpful to decide the category of memes that will be analyzed. For example, will you be looking mainly at memes related to pop culture, politics, etc. It could help to choose a category, which will then help with constraining your corpus and making the project more manageable as a whole. When you mention transcribing tweets, do you mean pulling the text from images shared on twitter, or transcribing the actual words in tweets that include popular jargon and linguistic trends?

[spkellyo]

To help you narrow it down, as mentioned above you need to be specific about what type of memes you are looking at. Are you looking at memes associated with a social issue, for example? The NFL protests last year produced a lot of memes that were for/against the players, for/against the NFL, and for/against specific teams. So, you could narrow down your question and then decide which type you want to look at and then determine how specific/broad within that type you want to be in order to decide on your corpus. Do not transcribe tweets. It is a long an arduous process. I would, personally, stick to just the memes. Staying with the example above, if you looked at the memes posted onto the twitter page of The Steelers during a specific time period, like the first week of the 2017 season, then you would have a manageable number of Tweets to look at that contained memes (they all won't have memes). Then you could classify them by in support of the players/ against the players/ against the NFL/ against The Steelers which would give you a nice grouping of different aspects to look at. I don't know if you have thought about this already, but some images will be copy righted so beware of putting them on your page without attributing them to the creator/owner. I do a lot of work with social media and rhetoric, but I have to do it all by hand. There are ways to mine the web for what you are looking for, but that is a whole other beast. Good luck.

[pickettj]

WRT the problem of selection, what about letting another repository site do the selecting for you? E.g. scrape material from https://knowyourmeme.com/ or something like that. You could choose a category on that site, and your object would become that particular website as an artifact. Just a thought.

[ddaud02]

Do you plan to analyze just what it is written on your memes? or the semantics?